{' '} {' '}
Limited time offer
SAVE % on your upgrade.

Page 1



Copyright © Field Notes: Journal of the McGill Undergraduate Geography Society, McGill University, Montréal, Canada, 2012. Editorial selection, compilation, and material © by the 2012 Editorial Board of Field Notes and its contributors. Field Notes is an academic journal of McGill University with submissions by undergraduate students. Printed and bound in Canada by Solutions Rubiks Inc. All rights reserved. Except for brief passages quoted and cited from external authors, no part of this book may be reprinted or reproduced or utilized in any way or form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or any information storage or retrieval system, without permission in writing from the publisher. Special thanks to the Arts Undergraduate Society of McGill University, the Science Undergraduate Society of McGill University, and the McGill Geography Undergraduate Society for enabling us to publish this journal. Cover photo by Brendan Buchanan Dee.

FIELD NOTES J o u r n a l o f t h e M c G i l l U n d e rg r a d u a t e G e o g r a p h y S o c i e t y

EDITORS-IN-CHIEF Miriam Dreiblatt & Jeremy Keyzer JOURNAL ADMINISTRATOR Joshua Fagen UNDERGRADUATE EDITORS Courtney Claessens Katie Cundale Corey Lesk Matthew Shields GRADUATE EDITORS Drew Bush Blanaid Donnelly Sarah Wilson DESIGN EDITORS Joseph Ariwi & Brendan Buchanan Dee


FROM THE EDITORS-IN-CHIEF We are thrilled to introduce the first issue of Field Notes, the journal of McGill’s Undergraduate Geography Society. What you will see in the following pages began as an impromptu conversation in the Geographic Information Centre on how best to promote undergraduate students’ research and highlight the diversity within McGill’s Geography department. The development of this first volume has been a learning process for both the authors and the editorial board on peer review and publication. We have thoroughly enjoyed producing this volume and hope that it will act as a precedent for future issues, engaging the Geography department’s students and faculty in the upcoming years. This issue contains six papers and one freestanding map produced by undergraduate students. The journal touches on a range of contemporary issues, from the effect of climate change on the lemur community in Madagascar to food security in the Sahel. There are many who have contributed towards the launch of Field Notes, without whom this journal would not have been possible. First and foremost, we would like to thank the undergraduate and graduate editors. Each editor came to the journal with a distinct geographic background and project vision, but collaborated to make this publication a reality. We are also grateful for the design team’s hard work, which synthesized abstract ideas and images in the assembly of this beautiful book. Lastly, we appreciate the support of our journal administrator as well as that of the McGill Undergraduate Geography Society, who together ensured the journal’s funding and created an environment conducive to completing this project. We hope you enjoy what follows. Miriam Dreiblatt & Jeremy Keyzer Editors-in-Chief


FIELD NOTES, VOLUME I Planting the Seeds for an Agricultural Revolution: Evaluating Barbados’ Potential for Increased Food Security through Small-Scale Agriculture CALHOON, DESMOND, LEBUTKIN & MARCHAND......................................... 6

The Role of El Niño: Southern Oscillation in Shaping Cholera Disease Dynamics in Bangladesh ALLNUTT.......................................................................................................... 20

Review of Temporal GIS WONG............................................................................................................. 30

The Impact of French Colonial Rule on Contemporary Food Security in the Central Sahel LESK................................................................................................................ 40

Vulnerability Assessment of Lemurs on the Island of Madagascar SARRAZIN....................................................................................................... 50

The Sea-Ice Carbon Pump: The Carbon Biogeochemistry of Sea Ice and its Role in the Global Carbon Cycle WARD.............................................................................................................. 62

Total Watt Hours per Rooftop Building in Ville-Marie ARAOS-EGAN & HABERMAN......................................................................... 74 OUR CONTRIBUTORS...................................................................................... 76

Planting the Seeds for an Agricultural Revolution: Evaluating Barbados’ Potential for Increased Food Security through Small-Scale Agriculture

DEREK LUBETKIN, ERICA CALHOON, JUSTINE DESMOND, AND KIRYA MARCHAND The small island nation of Barbados is currently facing a crisis of food security. The transition from an agricultural to a predominantly service economy has led to an exaggerated dependence on imports. This has in turn forced high commodity prices onto consumers. The object of this report is to discern to what extent a small-scale agricultural revolution could empower Barbadians and render the island more food secure. This information was collected from the combination of academic literature and a series of personal interviews, the results of which are organized and explored thematically. The final product of our research was an identification of the major barriers and potentials to small-scale agriculture that stem from the geographical, historical, and socio-cultural systems that qualify life on Barbados. The final section of this report will attempt to look forward to a potential future in which small-scale agriculture will help Barbados to navigate uncertainties such as population growth, economic hardship, and climate change.




country’s primary responsibility is to feed its people, which can prove daunting for small island nations such as Barbados. With a population of just under 300,000 people, the country has, in the past several decades, been shifting to a service-oriented economy centered on the tourism industry (Brathwaite, 2012). The island experienced rapid economic growth following this transition. This may be good news for the country’s current economy, but it comes at a price. According to the FAO (2006), the nation’s heavy dependence on imported food and beverages represents a violation of the four standards of food security: availability, access, utilization, and stability. Consider, for example, the fact that high mark-up prices on imports render some products unavailable to Barbadians of lower incomes, or the risk that exogenous shocks such as hurricanes might cut the island off from its primary food sources (Brathwaite, 2012). These factors suggest that Barbados is currently in a state of food insecurity, and thus the welfare of its population may be in jeopardy. Yet the situation is far from hopeless. Personal and community gardens represent a compromise between the extremes of absolute devotion to either agriculture or services and enable community development through collective action (Armstrong, 2000). The goal of our research was to discover to what extent a small-scale agricultural revolution could empower Barbadians, rendering the island more food secure. This paper weighs the predicted benefits of personal and community gardening against the geographic and societal limitations of the island and examines the present potential of gardening within the greater narrative of the island’s history. How is the proposed agricultural revolution informed by the events in Barbados’ past; and how, in turn, will it

influence its future? This paper uses the terms “small-scale agriculture,” “home gardening,” and “horticulture” interchangeably. These terms refer to backyard and community food production, modest in harvest and informal in nature. Environmental overview of Barbados

The majority of Barbados has a Pleistocene coral-limestone base created by ancient continental drift and uplifting (Randall, 1970). Water percolates through this rock to create complex subterranean channels, deep reservoirs, and freshwater springs, limiting easily available water abundance (Schomburgk, 1848; Randall, 1970). The United Nations has classified Barbados as a water-scarce country (Government of Barbados, 2001). Indeed, the population currently uses 98% of renewable freshwater resources, and increasing urbanization and population growth will create even greater demand for water resources (Cashman, 2011). The island’s relatively low annual precipitation and high rates of evapotranspiration also limit plant growth in Barbados (Watts, 1970). However, the rainy season (from June to November) boasts an average rainfall of about 168.4 mm. During this period, flash floods are common over low-lying areas, leading to soil erosion on agricultural land (ibid.). Barbados’ original dense tropical rainforest was cleared for plantation agriculture (Watts, 1970). Forests were future altered by the establishment of several problematic exotic species, as Barbados’ isolation and geological affects made it extremely vulnerable (ibid.). Today, the island’s unaltered wild vegetation is limited to small fragments found in narrow gullies, steep slopes and the coastal mangrove forest (ibid.). The main vegetation types include seasonal forest, grassland FIELD NOTES | VOL I | 7

and thicket, dune woodlands, strand vegetation and secondary seasonal woodlands (ibid.). History: Intersections of agriculture and social institutions

Barbados’ climatic conditions and soil were superbly suited to the growth of sugar cane, sparking the island’s economic and population boom during the 17th and 18th centuries (Beckles & Downe, 1987). The rise of the Barbados sugarcane industry was accompanied by the institutionalization of large-scale slavery operations on the island. Due to the massive influx of slave labour, the population of African-born slaves in Barbados ballooned from 50 (0.03% of total population) in 1629 to 62,115 (79% of total population) in 1786 (Momsen, 2005). This growth was driven almost exclusively by the sugar industry, and it is estimated that from the time of colonization in 1807 more than 380,000 native Africans were brought from the Gold Coast, or present day Ghana (Watson, 1979). This system of slave agriculture remained in effect until the mid-19th century, when the profitability of sugarcane began its slow decline (Drummond & Marsden, 1995). A major shift in Barbados’ agricultural practices occurred following the abolition of slavery in 1834. At this time, plantations accounted for nearly all of the arable land (Barrow, 1995). As a result, recently freed slaves could only acquire land after large estates were broken up and sold (Greenfield, 1960; Momsen, 2005). This process of land acquisition caused small “villages” to appear as the land of a single estate was divided to create many smaller plots (ibid.). As the profitability of the sugar industry decreased, plantation owners began to sell their divided land to ex-slaves, primarily women, for subsistence agriculture (Momsen, 2005).


Consequently, the Barbados tradition of small-scale agriculture and backyard gardens began to take root. Most of the land sold to freed slaves, however, was marginally productive at best and often plagued by low fertility, steep slopes, and inaccessibility (Momsen, 2005). New forms of small-scale agricultural land tenure systems in Barbados accompanied this redistribution of land, including rented and non-rented property. Non-rented land was further divided into two categories: freehold and family land. Freehold land refers to the ownership of a land plot by a single individual, while an entire family collectively owns family land (Momsen, 2005). Although single ownership still represents about 70% of agricultural land in Barbados, the quantity of family-owned land doubled twice between 1963 and 2003 (ibid.). The increase in family-owned land occurred due to the desire to teach younger family members the importance of cultivating the land, while also avoiding further subdivisions of already small land parcels (ibid.). Small-scale farming expanded throughout the 19th and early 20th centuries as land was further distributed among the Barbadian population (Momsen, 2005). However, recent decades have seen an increased demand for non-farming education as well as a steady climb in real estate prices, both results of the growing tourism sector. The combination of these phenomena led to a transition away from agricultural activities towards service economies (Momsen, 2005). Simultaneously, the sugarcane industry’s profitability continued to decrease, causing large plots of land previously utilized for agriculture to be sold for development projects (Barrow, 1995; Momsen, 2005). The shift away from agriculture is evident in the amount of remaining farmland in Barbados, which has decreased by half be-

tween 1950 and 1990 (Momsen, 2005). Evidence of the decreasing Barbadian interest in farming is similarly exemplified by the large amounts of agricultural land that have been sold for non-farming purposes: between 1970 to 1985 alone, Barbados lost 1.4 hectares of arable land per day, increasing to 5.3 hectares per day between 1987 and 1989 (Barrow, 1995). The agricultural sector in today’s national economy

Today, the agricultural sector ranks third among Barbados’ top grossing economies, but contributes a mere 3.4% to the nation’s GDP (CIA, 2011). This relatively low contribution can be partly attributed to the size of the small island, of which only 37.2% is arable land and 34.4% of that land is devoted solely to the harvesting of sugar cane (ibid.). Although raw sugar continues to front the list of the island’s top grossing commercial exports, the industry is suffering a steady decline. In 2010, the total production of sugar cane fell by 23.5%, while the price paid per bulk unit plummeted by over 32% (Government of Barbados, 2010). This market failure has effectively prompted a re-evaluation of Barbados’ land and agricultural legislation, and the government is now encouraging plantation owners to convert plots of land towards the production of non-sugar cane fruits and vegetables (ibid.). The failure of the sugar industry is thus being re-marketed as an opportunity for Barbados to revolutionize its agricultural sector, and to lessen its costly dependence on the import of foreign foods and beverages. According to Brathwaite (2012), Barbados currently imports an estimated 70% of all the food consumed on the island, which annually costs an estimated $313 million USD or 16.5% of the total annual value of imports (FAOSTAT,

2011). The most ubiquitous foreign foodstuffs are brute foods such as maize, soybean, and wheat, but the most costly are prepared foods belonging to multinational brands that predominantly arrive from the United States (FAOSTAT, 2011; Chandler, 2011). This high price of food commodities has been identified as the leading factor driving high inflation rates in the Caribbean (Brathwaite, 2012). In 2011, Barbados underwent 7.9% inflation on commodity prices, compared to a global average of only 5.4% (CIA, 2011). This can be attributed to the high cost of living associated with higher food prices, the effects of which are more heavily borne by the average consumer (Scott-Joseph, 2009). Although Barbados boasts one of the most economically prosperous populations in the Caribbean, total public debt represents 103.9% of the nation’s GDP (CIA, 2011), and the island’s high cost of living was named the top concern by voters in the 2008 federal election (BBC News, 2012). When an exaggerated dependence on foreign food stresses on Barbados’ national economy and household budgets, it becomes necessary to re-evaluate the state of the country’s agricultural sector. A resurgence of small-scale home gardening would provide economic relief to Barbadians, the most disadvantaged of whom would be the first to profit from the revolution.

METHODS Participants

In order to assess prevailing attitudes toward food production, thirty interviews were conducted over the course of three days in the field. Respondents from the first day of data collection were from the Orange Hill neighbourhood of SaintJames. During the next two days vendors FIELD NOTES | VOL I | 9

and other participants at Agrofest—a two-day festival in Bridgetown that showcases Barbados’ agricultural sector—were interviewed. The festival features local food growers and vendors as well as livestock farmers, persons involved in horticulture, representatives from the Barbados Ministry of Agriculture and other agricultural organizations such as the Food and Agriculture Organization (FAO) of the United Nations and the Caribbean Agriculture Research and Development Institute (CARDI). On the first day at Agrofest, the site was open exclusively to schoolchildren and their instructors and thus offered a unique opportunity to learn about agriculture-related programs in the Barbados education system. The second day at Agrofest was open to the general public and allowed us to gain insight into the local population’s interest in agriculture and related activities. All of the respondents were adults over the age of majority who gave their informed consent prior to beginning the interview. Interview techniques

The interviews were semi-structured and featured a list of prepared, open-ended questions evoking the general themes we wished to explore. This informal format enabled a natural conversational flow and for new questions to arise.

Sources of error

The Orange Hill neighbourhood was sought out precisely because it contained a high concentration of households with gardens, which indicates that residents may have biased views in favour of home gardening. Any of the respondents interviewed at Agrofest were predisposed to attending a festival about agriculture, and so may have held similar biases. We conducted interviews with a non-random sample of thirty Barbadians from amongst the is-


land population of 288,000 (CIA, 2011) and thus our results may not be generalizable. Further, we are culturally and ethnically isolated from the local population so may have received different answers than might have been given to a Barbadian researcher. Despite these potential sources of error, the research presents a starting point for further inquiries into the potential of increasing Barbados’ food security through the expansion of small-scale farming.

RESULTS The generation gap and the role of education

The results show distinct and noticeably different attitudes towards farming and gardening amongst disparate Barbadian age groups. The morning spent with students at a primary school and the information provided by their teachers demonstrated a deep interest in gardening among school-aged Barbadians. However, our research suggests a considerable decline in this interest and an attitudinal disregard for agricultural activities among secondary school students, which continued through young adulthood. Respondents from the young adult generation looked down upon agricultural work as “uncool” and menial. Some respondents explained they did not need to farm because they could simply purchase the food they required from the market, exemplifying the typical attitude of the developed, post-subsistence state. Thus the overarching picture depicted by young Barbadian respondents echoed a distinct lack of desire to engage in agricultural activities. This was coupled with a general sentiment amongst young adults that the island’s food security is not a major issue of concern. Most of the respondents explained it was more important to learn other skills in order to succeed in the modern economy and questioned the

potential to earn a living through agriculture. Contrary to the young adults interviewed, many of the middle-aged and older Barbadians cited the importance of gardening as a means to supplement personal income, especially during times of economic hardship. Many of the older respondents were taught gardening techniques by their parents, and still remember the island’s agricultural heritage. However, many of these respondents noted the difficulty of maintaining a garden while working full-time, a viewpoint shared by the young adult population. The older respondents also consistently cited the lack of interest among the younger generation as a major problem for the future of Barbados’ agricultural sector and referred to many younger Barbadians as lazy and lacking work ethic. While this notion certainly carries a bias (and is itself characteristic of generational gaps around the world), it is emblematic of the growing divide between younger and older Barbadians. This divide represents one of the most significant problems facing a proposed revolution in Barbadian agriculture. Barbados must first and foremost overcome this lack of knowledge and interest among young islanders, without whom the future of home gardening stands barely a chance. Perceptions of agriculture and the environment

Our results show that Barbadian perceptions of agriculture are influenced by historical effects, have changed throughout time, and vary depending on age and income. Among older Barbadians our research team noted a variety of responses regarding environmental issues. Some respondents demonstrated acute concern for the island’s environmental future, and grieved a general lack of environmental risk perception amongst their fellow

Barbadians. Similarly, this group demonstrated a keen interest in issues of island sustainability. Conversely, others claimed that Barbados had no environmental problems, with one respondent proposing that technological advances could solve all environmental problems. One respondent cited an issue of complacency, claiming that “Bajans are very good at suffering in silence,” and that intervention will not happen until issues with food security become more critical. These responses reflect the diversity of our respondents, but also show that there is not a shared or generalizable perception of the island’s environmental threats. In addition, many non-gardening respondents were under the impression that Barbados did not have good soil for growing; a misconception, as evidenced by the island’s rich agricultural history and current population of successful gardeners. It also became evident that Barbadian perceptions of gardening differed greatly from those of North Americans. In an interview on Orange Hill, one woman refused to identify herself as a gardener despite her abundant and diverse garden that contained a banana orchard, mango trees, and four large containers of herbs. Evidently, this cultural difference is intrinsically linked to Barbados’ history of sugar plantations and slavery. In several interviews respondents expressed that due to the history of slavery on the island, there is a cultural stigma surrounding agricultural practices. This stigma has resulted in limited gardening, as many Bajans “want a better life for their children” than can be provided by working the soil. As described by one respondent, there is also an important link between perceptions of income and agriculture. As household wealth increases, Barbadians practice floriculture as a display of status, whereas agriculture is apparently practiced for subsistence. Thus FIELD NOTES | VOL I | 11

the negative stigma associated with fruit and vegetable gardening is deeply rooted in the cultural history of the island, and represents one of several barriers to the expansion of novel agriculture in Barbados. Respondents also shared the opinion that growing food at home was too time consuming and required too much hard work to be compatible with other means of employment. David Bynoe of the Ministry of Agriculture mentioned that the media consistently documents instances of financial losses from farming and never includes stories of economic success or of money saved. This leads people to perceive agriculture as inherently costly, or as presenting a financial pitfall to those who undertake it. Contrary to this shared belief, however, several gardeners mentioned that agriculture has been making a comeback following the economic downturn of 2008. Not only have people been growing their own food as a way to make ends meet and avoid expensive imported produce, but one respondent explained the decrease in economic activity and opportunity for employment has created more time for being at home and tending to his garden. Bynoe recommended that Barbados start a campaign to feature successful farmers in the media, creating “role-model” status for young audiences and altering attitudes towards home gardening. Despite these barriers and beyond even the economic benefits, many of the garden-owning respondents mentioned that growing food was above all an inspirational and fun activity. They believe that gardening could be used as a tool for community building, especially through school and church groups that have the appropriate space and soil. Furthermore, Barbados’ dependency on foreign imports, rising food prices, and the link between agriculture and health were all cited as key reasons why small-scale agriculture 12 | LUBETKIN ET AL.

has the potential to grow. While there was variety in our respondents’ answers about their perceptions of agriculture and environment, the general sentiments showed potential for the growth of small-scale agriculture even in the face of some limiting cultural barriers. Buying and selling locally

Our interviews indicated that Barbadians have a strong interest in buying local food. Of ten shoppers interviewed in the produce section of a chain grocer, nine admitted preferring local over imported goods. The consensus was that local food is of a superior quality and freshness compared to imported food. This was the general opinion for fruits and vegetables grown locally, but not necessarily the case for goods manufactured on the island. These findings were confirmed in an interview with the manager of Jordan’s Supermarket in Speightstown, who stated that customers are more likely to buy Barbadian produce when both local and imported varieties are in stock. For this reason, an estimated 95% of in-season vegetables on the shelves are locally produced. By comparison, however, only a third of the total stock of the store is produced or manufactured in Barbados. Shoppers at Jordan’s Supermarket were also of the shared opinion that local produce not only tasted better, but was healthier for consumers. “You know what you’re buying,” said one respondent, who went on to explain that locally grown fruits and vegetables often had less chemicals and were GMO-free. One respondent identified this association between homegrown produce and healthy eating as the product of a recent nationwide campaign that was advertising proper nutrition through local foods. Despite this fervent interest, our interviews also identified perceived cost as a

potential barrier to local food markets in Barbados. Three interviewees named the perceived higher cost of local foods as a possible deterrent, and one more claimed to buy local “despite the fact that it’s more expensive.” Although this shared sentiment implies that the price of local food often exceeds that of imported goods, vendors reported commodity prices on local fruits and vegetables that are competitive with—or even marginally below—the prices of the their imported counterparts. The manager at Jordan’s indicated that inseason fruits, vegetables, and herbs from Barbados retail at the same price as products imported from abroad. A representative of the “Buy Bajan” movement blamed part of this public misconception on the average mark-up of food processed versus locally grown food. Because Barbados has transitioned from an economy based on large-scale agriculture to one based on tourism and services, the island lacks the infrastructure to cost-effectively process foodstuffs. When shoppers see that Barbadian-made food is more expensive, they logically but erroneously apply the same principle to food that is Barbadian-grown. This misconception presents a barrier to the development of the island’s small-scale agriculture movement, and a possible deterrent to growers looking to enter the industry. Our team also identified poor labeling standards as a second barrier to local food markets in Barbados. Our respondents indicated that locals typically know which foods can be grown on the island, but not which products on the shelves are authentic local produce. Following a brief survey of the stock, chives, chayotes, and bananas were discovered as the only products clearly advertised as domestic produce. Alongside them found Mexican limes, Canadian carrots, Costa Rican pineapples, and other imported versions of fruits

and vegetables commonly grown by local farmers. While discussing the ability of Barbados to grow so much good food, one local even motioned what he believed to be local peppers in his shopping cart, which were actually Californian imports. Much of this confusion can be credited to the lack of a clear, standardized system of food labeling. Although the manager of the store claimed that 95% of the produce was local, few items had country stickers, few were labeled as “local” or “imported,” and most gave no indication of their origin. Many respondents felt as though a clear label—such as the proposed “Buy Bajan” sticker or the “100% Bajan” tag of years gone-by—would help shoppers distinguish between local and imported products. All ten of the interviewees in the aisles of Jordan’s said that they would like to see such labels. If shoppers are willing to buy into local small-scale agriculture, the nation does itself a disservice by obscuring local produce and confusing it with imported goods.


We identified several barriers to the spread of small-scale agriculture through our interviews and background research. These barriers are categorized into three groups: environmental, economic, and cultural constraints. Of these, we suggest that transforming Barbadians’ unfavourable cultural perception of gardening and agriculture should become a national priority. The environmental issues limiting the spread of small-scale agriculture are diverse. First, Barbados’ currently has limited arable land and freshwater resources due to it’s topography combined with a history of intense natural resource exploitation. Furthermore, human activities have reduced previous levels of biodiverFIELD NOTES | VOL I | 13

sity in forests. Exotic species such as the Green Monkey and Giant African Snail have destroyed large areas of crops (see Caldwell-Rafferty, Roan, Sharp & Templeton, 2012). Furthermore, due to the agricultural sector’s need for climatic and environmental stability, uncontrollable events such as tropical storms act as a deterrent to farmers wary of environmental risk. Economic constraints also act as barriers to small-scale agriculture. Since land in Barbados typically finds its highest capital returns under tourism and related services, much land has been converted from agricultural uses. The research demonstrated that many respondents perceived home gardening as requiring a time investment too significant to be balanced with other sources of employment. Officials at the Ministry of Agriculture acknowledge that the media frequently documents instances of financial losses from farming, leading people to perceive agriculture as inherently risky. Meanwhile, the producers of locally grown food are losing profits due to the widespread perception of local foods as costlier than imported alternatives, though this claim was proved to be inaccurate. The Barbadian cultural perceptions of agriculture are the most significant barrier to the adoption of small-scale agriculture. The generation gap in Barbadian society is inherently cultural and it identifies both a lack of knowledge about gardening techniques and a general lack of interest among young adult Barbadians. Barbadian attitudes are further characterized by divergent visions of gardening as a leisure activity rather than a productive one, an association of agriculture with the sugar plantations and slavery of the past, and the view that agriculture is a lowly and disrespectable activity not to be practiced by ‘progressive’ Barbadians interested in 14 | LUBETKIN ET AL.

development. Climate change: Uncertainties in the future of Barbados’ agricultural sector According to the Government of Barbados’ (2001) the following impacts of climate change are expected to pose significant problems for Barbados: increases in temperature; sea level rise and its attendant impacts of erosion; inundation and saline intrusion; and changes in weather patterns including changes in the amount and timing of rainfall along with changes in storm intensity. What is most significant for this analysis are the ways in which climate change will directly or indirectly affect Barbados’ current economy, the consequences of which will inform the future of its agricultural sector. The island’s resort hotels sit almost exclusively within inundation zones, placing them at high risk of structural damage from sea-level rise (Government of Barbados, 2001). Reduced and irregular rainfall will exacerbate water scarcity, impacting sanitation and the ability to provide amenities (ibid.). Increased storm activity will increase insurance premiums, lessening the cost competitiveness of Barbados as a tourism destination (ibid.). Warmer sea temperatures will kill coral reefs and wildlife, an important tourist attraction, and coral reef death will mean a decrease in sand for beaches (ibid.). The decline of the tourism sector, which now accounts for 14.5% of GDP (ECLAC, 2010), will inevitably lead to general economic decline, similar to that caused by the 2008 global recession. A 6.6% decline in tourism earnings from 2008 to 2009 was partially responsible for a 3.6% decline in GDP over the same period (ibid.). Thus the industry of sugar cane production retains importance as a foreign exchange earner (Government of Barbados, 2001). Sugar yields are expected to fall as a result of increased wa-

ter scarcity, soil salinity, and atmospheric CO2 concentrations (ibid.), exacerbating any economic decline. A contracting economy will cause individual households to experience a decreased ability to purchase imported foodstuffs. In fact, economic decline was cited by many of our respondents as a reason for a comeback in domestic agriculture. To the extent that climate change will compound economic woes, there is an increased need for home gardening to compensate for losses in ability to purchase food. The question of the impact of climate change on global food security is also important for Barbados, a net food-importing country. Urban populations of island countries are very much dependent on cheap foreign imports for their daily sustenance (FAO, 2008b). Given that 44% of Barbados’ population is urban (CIA, 2011), the effects of climate change, which impacts international food prices, will create a critical situation in food security. Globally, food security has already been adversely affected by high oil prices and growing demand for agro-fuels, as well as droughts and floods linked to climate change (FAO 2008a). Access to food will be worsened by climate change events that lead to damages in infrastructure and loss of income and employment opportunities (ibid.), both of which are likely to occur given the vulnerability of the tourism industry. According to the FAO’s State of Food Security in the World report (2011), high and volatile food prices are likely to continue due to supply side challenges stemming from scarce natural resources and declining rates of yield growth for some commodities. Volatility may even increase due to the strong linkages between agricultural and energy markets and the increased frequency of weather shocks (ibid.). Improvements in local food production may strengthen market

resilience, especially in under changing climate conditions (FAO, 2008b). Looking forward

Although we found several barriers to the expansion of small-scale agriculture, there are also many factors promoting the proliferation of gardening in Barbados. Programs ranging from governmentbacked initiatives to grassroots movements are taking hold, and consumer awareness regarding local food is growing. Therefore, the future of Barbados’ food production is bright. In particular, our team identified government funded projects, producerdistributor co-operatives, and the shifting views of the general population as being the strongest drivers for change. These factors must be reviewed in order to better understand the current trend of agriculture in Barbados. One of the brightest stars of Barbados’ agricultural scene is the annual Agrofest, organized by the Barbados Agricultural Society. This multi-day festival has grown exponentially during the past few years and is now one of the island’s largest annual public gatherings. Many Barbadian farmers and agricultural organizations attend the festival, together representing the island’s multi-faceted agricultural sector. This event has helped raise awareness concerning the state of food security in Barbados, and provides a stage for sharing agricultural knowledge and novel ideas. Agrofest is visited by every school child on the island, and thus represents one vehicle for overcoming the generational gap previously identified as a barrier to smallscale agriculture. Agrofest also provides a direct link between the general public and the initiatives currently being put forward by the Ministry of Agriculture. One such initiative is the “Land for the Landless” program. This program provides Barbadian FIELD NOTES | VOL I | 15

farmers with access to public and private arable land for cultivation. The program began in 2001 and was designed to empower disadvantaged Barbadians through agrarian reform. The initiative lists three objectives: to increase the level of production in agriculture, to provide increased employment opportunities in agriculture, and to increase the contribution of agriculture to the GDP. Similar government programs emphasize recent initiatives in greenhouse planning, workshops in gardening and horticulture, and better business practices. Government involvement may foster the growth of Barbados’ agriculture output while encouraging greater social participation in food production. Reducing the generation gap with regards to agricultural interest is one of the most important issues facing Barbados’ small-scale agriculture. This problem is beginning to be addressed through school run gardening clubs and the Barbados

Figure 1. Garden at St. Silas Primary School 16 | LUBETKIN ET AL.

4-H organization (see Figure 1). An interview with a primary school teacher who supervised her school’s gardening program revealed that children are increasingly enabled to become active gardeners at their schools, while simultaneously learning the knowledge necessary to become successful gardeners at home. At this point in time, the key to maintaining this interest is to implement more exhaustive programs at the secondary level, where only two of 23 public schools currently offer gardening programs. However, more programs are being developed to address issues of space and urban agriculture. Both the Bellairs Research Institute and the 4-H program address the concerns many Barbadians have regarding their lack of land to cultivate with multi-tiered container gardening systems. The popularization of container gardening eliminates the need for large backyards and proper natural soil

conditions, thus allowing a much greater number of Barbadians to participate in small-scale food production. Several successful container gardening projects were on display at Agrofest demonstrating the potential for this aspect of agriculture to expand. Our team was also impressed by the potential of container gardening as performed by one gardener in Orange Hill, who increased the productivity of her backyard space by lining discarded bathtubs with soil for growing herbs and companion flowers (see Figure 2). In addition to the numerous programs and initiatives focused on expanding Barbadian agriculture, one of the most influential forces for change is the current global economic downturn. Several respondents commented on the importance of personal food production as a method to offset rising food prices and declining income. These respondents explained that if food imports continue to rise in

cost relative to household incomes, many Barbadians will seek to save on household expenses. Therefore, future economic crises may lead Barbados towards strong, continued investments in small-scale agriculture, and non-farmers may yet be convinced by the success of their neighbours. Although not as popular today, Barbados nevertheless retains a rich history of gardening and food production. This history has yet to be forgotten, and its model serves as yet another asset to the proliferation of gardening on the island. While there are concerns over the historical significance of slavery to agriculture, Barbadians have also historically utilized small-scale agricultural practices when the circumstances deemed necessary. One respondent described that this tradition of gardening was crucial to the island’s survival following the Second World War, a time when food imports were in short supply and the island was threatened with

Figure 2. Container Gardening in Orange hill FIELD NOTES | VOL I | 17

famine. Although the situation is not yet as dire as during WWII, the current state of Barbados’ food insecurity calls for a partial return to traditional means of feeding the island. The seeds of this agrorevolution are already sown, and Barbadians need only decide to broaden their support.


The aim of this research was to explore the potential for small-scale agriculture on the island of Barbados. Given Barbados’ current situation and anticipated changes, a return to traditional methods and the development of small-scale agricultural initiatives are now becoming crucial. Although there currently exist significant cultural, economic and environmental barriers to the wide adoption of smallscale agriculture, they must be overcome for the good of the island’s people. The existing environmental barriers can be overcome through the promotion of gardening techniques that are flexible and made to mitigate the anticipated impacts of climate change. Cultural and economic barriers can be overcome through the empowerment of Barbadian youth and by increased media portrayal of agriculture as a source of national pride, instead of a burden to be borne by only the country’s lowest constituents. Technological changes such as container gardening can complement the necessary social changes. The research concludes that a revolution in household and community gardening has the power to render the nation more food secure, and thereby act as an impetus to national development.


REFERENCES Armstrong, D. (2000). A Survey of community gardens in upstate New York: Implications for health promotion and community development. Health & Place, 6(4), 319327. Barrow, C. (1995). The Plantation Heritage in Barbados: Implications for Food Security Nutrition, and Employment. ISER Working Paper, 41. Institute of Social and Economic Studies, University of the West Indies. Beckles, H.D., and A. Downes. (1987). The Economics of Transition to the Black Labor System in Barbados, 1630-1680. Journal of Interdisciplinary History, 18(2), 225-247. Brathwaite, C. (2012, February 25). Speech presented at Agrofest, Bridgetown, Barbados. British Broadcasting Corporation (BBC) News. (2012). Barbados country profile. BBC News. Retrieved from http://news. bbc.co.uk/2/hi/americas/country_profiles/1154116.stm Caldwell-Rafferty, N., S. Roan, K. Sharp & M. Templeton. (2012). “An Everyday Affair”: Managing Human-Wildlife Conflict in Barbados. Montreal: McGill University Press Cashmann, A. (2012, February). Water Management. Speech presented at the University of the West Indies, Cave Hill, Saint Michael, Barbados. CIA. (2011). World Factbook: Barbados. CIA. Retrieved from https://www.cia.gov/library/ publications/the-world-factbook/geos/ bb.html Chandler, Dr. Frances. (2011). How Can We Reduce Our Food Import Bill? Business Barbados. Retreived from http://businessbarbados.com/industry-guide/agricultureand-fisheries/how-can-we-reduce-our-foodimport-bill/ Drummond, I. & T. Marsden. (1995). A Case Study of Unsustainability: The Barbados Sugar Cane Industry. Geography, 80(4), 342-354.

Economic Commission for Latin America and the Caribbean (ECLAC). (2010). Barbados. Economic Survey of Latin America and the Caribbean 2009-2010 (pp. 199-202). Santiago, Chile: United Nations Publication. Food and Agriculture Organization of the United Nations (FAO). (2006). Food Security. Retrieved March 27, 2012 from ftp:// ftp.fao.org/es/esa/policybriefs/pb_02.pdf Food and Agriculture Organization of the United Nations (FAO). (2008a).Climate Change Adaptation and Mitigation in the Food and Agriculture Sector. Retrieved March 14, 2012 from ftp://ftp.fao.org/docrep/fao/meeting/013/ai782e.pdf Food and Agriculture Organization of the United Nations (FAO). (2008b).Climate Change and Food Security in Pacific Island Countries. Retrieved March 14, 2012 from http://www.fao.org/fileadmin/user_upload/ foodclimate/f... Food and Agriculture Organization of the United Nations (FAO). (2009). Imports: commodities by country. Retrieved March 12, 2012 from http://faostat.fao.org/desktopdefault.aspx?pag eid=342&lang=en&country=14 Food and Agriculture Organization of the United Nations (FAO). (2011). The State of Food Insecurity in the World 2011. Rome, Italy: United Nations Publication. Government of Barbados. Ministry of Physical Development Environment. (2001, October). Barbados’ First National Communications. St. Michael, Barbados: Ministry of Physical Development Environment. Retrieved March 14, 2012 from http://unfccc. int/resource/docs/natc/barnc1.pdf. Government of Barbados. Ministry of Finance & Economic Affair. (2010). Economic and Social Report 2010. Bridgetown, Barbados: Ministry of Finance and Economic Affairs. Retrieved March 5, 2012 from http:// www.economicaffairs.gov.bb/archive-detail. php?id=243

Greenfield, S.M. (1960). Land Tenure and Transmission in Barbados. Anthropological Quarterly, 33(1), 165-176. Momsen, J. H. (2005). Caribbean Peasantry Revisited: Barbadian farmers over four decades. Southeastern Geographer, 45(2), 206221. Randall, R.E. (1970). Vegetation and Environment on the Barbados Coast. The Journal of Ecology, 58(1), 155-172. Schomburgk, R.H. (1848). The History of Barbados. New York: Cambridge UP. Scott-Joseph, A. (2009). The Nature of Rising Food Prices in the Eastern Caribbean: An Analysis of Food Inflation During the Period 2005 – 2008 in a Context of Household Poverty. UNICEF Office for Barbados and the Eastern Caribbean. Retrieved on March 20, 2012 from http://www.unicef.org/barbados/The_Nature_of_Rising_Food_Prices_in_the_Eastern_Caribbean.pdf Watson, K. (1979). The Civilised Island, Barbados: A Social History, 1750-1816. Caribbean Graphic Production. Watts, D. (1970). Persistence and Change in the Vegetation of Oceanic Islands: A Example From Barbados, West Indies. Canadian Geographer/Le Géographe Canadien, 14(2), 91-109.


The Role of El Niño: Southern Oscillation in Shaping Cholera Disease Dynamics in Bangladesh


Environmental determinants of cholera are important component of its occurrence because the vibrio cholera bacterium, the cholera pathogen, is known to inhabit aquatic ecosystems. A growing body of research suggests a correlation between cholera epidemics and climate variability associated with El Niño – Southern Oscillation (ENSO). The aim of this paper is twofold: (1) Describe the relationship between cholera and ENSO by drawing on a case study from Bangladesh where the disease is endemic, and (2) Determine if the correlation between the two can play a role in forecasting future cholera outbreaks. The results suggest that there is plausible evidence for ENSO influencing cholera outbreaks in Bangladesh; however, there are other factors that need to be considered in terms of cholera prevention. These include access to clean drinking water, sanitation, hygiene education, and proper water infrastructure. In addition, the internal disease dynamics must be considered in determining variability in disease outbreak. Overall, it is argued that El Niño is a factor in the ecological basis of cholera in Bangladesh, and can be effective in disease prediction. It is suggested that the presented link should be considered by policy makers and health officials to help target timing and location of health interventions.




holera remains a major public health problem in many developing nations. For example, areas with large cholera epidemics include Bangladesh, India, and countries in Africa and South America (Kovats et al., 2003). Figure 1 shows areas that reported cholera outbreaks in 2009-2010. Prior to the 1970s cholera has been viewed as a fecal-oral infection and person-to-person contact was perceived to be the method of cholera transmission, where epidemics began with contaminated water and food (Constantin de Magny et al., 2008). More recently, the environmental determinates of the disease have been recognized, since Vibrio cholera—the bacteria causing cholera—may inhabit coastal waters, estuaries and rivers (Pascual et al. 2000). Outbreaks of cholera emerge when V. cholerae is present in food and drinking water, and thus more likely to occur in regions with limited or damaged infrastructure (Cash et al., 2005). A growing body of research indicates

that the correlation between cholera epidemics and climate variability associated with El Niño Southern Oscillation (ENSO) (WHO, 1999). Increased attention has been paid to ENSO and cholera since the re-emergence of cholera in 19911992 in Peru coincided with an El Nino event, which motivated the hypothesis that ENSO events may influence cholera dynamics (Pascual, 2008). The aim of this paper is to critically analyze and characterize the role of El Niño Southern Oscillation in cholera outbreaks in Bangladesh where the disease is endemic. The objectives include: (1) To describe the associations between cholera outbreaks and ENSO events using a case study from Bangladesh, and (2) Determine if the relationship between cholera and ENSO can be used to forecast outbreaks.


Cholera is an acute diarrheal illness caused by an infection of V. cholerae bac-

Figure 1. Reported Cholera Outbreaks in 2009-2010; Source: World Health Organization, 2011 FIELD NOTES | VOL I | 21

Figure 2. Cholera transmission in nature; Source Schoolnik, 2008 teria in the small intestine (CDC, 2010). Symptoms of the disease include profuse diarrhea, vomiting, leg cramps, circulatory collapse, and shock (CDC, 2010). Untreated cases result in mortality of up to 50 percent. Mortality is reduced to less than one percent with appropriate medical intervention (Cash et al., 2010). Transmission occurs in two stages: primary and secondary. Primary transmission is a result of eating contaminated shellfish (copepods) or aquatic plants, or drinking contaminated water. Secondary transmission occurs from person-to-person contact (Ruiz-Mureno et al., 2010). Primary transmission shapes the seasonal patterns in endemic areas and secondary transmission determines the level of infection (RuizMureno et al., 2010). Figure 2 depicts how 22 | ALLNUTT

cholera is transmitted in nature. Environmental signatures have been associated with cholera epidemics, as V. cholerae is autochthonous to riverine, estuarine, and coastal ecosystems consisting of phytoplankton, aquatic plants and with its host, the copepod (Constantin de Magny et al., 2008). Temperature, salinity, and the presence of plankton are suggested to be important factors to the ecology of V. cholerae (Patz et al., 2005). Bacteria growth increases with warm temperatures, mainly in combination with blooms of phytoplankton, aquatic plants, and/or algae (Colwell, 1996). These blooms increase the amount of food available for copepods which the V. cholerae attach to, protecting the bacteria from exterior environmental factors and allowing it to multiply (Lipp et al., 2002).


ENSO is the strongest naturally occurring source of climate variability around the globe, following typical seasonal variability (Kovats et al., 2003). El Ni単o describes the warming of the waters in the equatorial Pacific and the southern oscillation is defined as the oscillation in air pressure mirroring the warming and cooling associated with El Ni単o (Kump

et al., 2009). ENSO is often referred to interchangeably with El Ni単o. The ocean warming associated with El Ni単o has a significant impact on climate around the globe, as described in Figure 3. With the exception of North America and Australia, the countries that experience the greatest environmental impacts from ENSO are in developing nations where the events are atypically hot or wet. In Bangladesh, the

Figure 3. ENSO Impact on regional climate; Source: Kump et al. 2009 FIELD NOTES | VOL I | 23

effects of ENSO are typically warmer than average temperatures from June to August and dry from December to February (Patz et al., 2005). ENSO occurs on time scales of 2-7 years and is associated with extreme weather conditions such as floods and drought (Kovats et al., 2003). Given that ENSO is a large source of climate variability and V. cholera is greatly influenced by environmental conditions, an association between the two is hypothesized (Cash et al., 2010). According to the World Health Organization (WHO) (2011), cholera incidence is underestimated in the Indian subcontinent and Southeast Asia, as countries within these regions typically do not report their cholera cases. Although there are no reported cases from Bangladesh, experts estimate that there may be one million cases per year in the country (WHO, 2011). Therefore, the WHO (2011) estimates the Burden of disease in Bangladesh is much higher than is reflected among reported cases. Further, cholera estimates can be made as the International Centre for Diarrheal Disease Research; Bangladesh (ICDDR, B) operates hospitals in Matlab and Dhaka, Bangladesh and keeps records of cholera patients (Emch, 2010). The WHO (2011) suggests that the number of patients arriving to hospitals in Bangladesh in shock due to cholera is increasing. Moreover, a large body of literature discusses the role of ENSO events in shaping cholera disease dynamics (Patz et al., 2002, Pascaul et al., 2000, Koelle et al., 2005, Cash et al., 2010). Given the quantity and quality of work on the subject in Bangladesh and the fact that the disease is becoming more virulent, it provides an informative case study for how ENSO affects cholera outbreaks over time.



There exists a clear climatic sensitivity to cholera epidemics in Bangladesh. Many researchers suggest a bimodal seasonal pattern to the outbreaks (Cash et al., 2010, Kovats, 2000, Kovats et al., 2003). That is, cholera incidence is at its maximum during fall following the abatement of summer monsoons and reaches another maximum in the spring before summer monsoons begin (Cash et al., 2010). ENSO is emerging as a significant predictor of cholera risk because it is observed that ENSO events alter the traditional bimodal distribution (Patz et al., 2005). Cholera in Bangladesh has varied with climatic fluctuations and sea surface temperatures (SSTs) influenced by ENSO phenomena over multi-decadal time periods (Patz et al., 2005). An association between fall cholera incidence and Winter ENSO events has been observed since cholera incidence has been shown to increase following December, January, and February El Ni単o months (Cash et al., 2010). The physical basis for this is that Bangladesh summer rainfall increases as a result of warm winter SST anomalies in the tropical pacific initiated by winter El Ni単o events. Cholera incidence, consequently, rises through increased flooding and malfunctions in sanitation systems (Cash et. al., 2010). In addition, outbreaks also increase because rainfall increases insoluble iron levels, which advances the survival of V. cholerae in marine ecosystems (Robo et al., 2005) and suggests that increased rainfall associated with ENSO events may improve conditions for the bacteria and increase food and water contamination.

Historical Data

The correlation between cholera and ENSO in Bangladesh has become more pronounced in recent years. An analysis of historical data in Bengal from 1890-1940

indicates that the effects of El Ni単o on cholera was historically confined to coastal regions and may have caused Spring epidemics (Kovats et al., 2003). Trends have become more pronounced since the 1980s in Bangladesh as the ENSO events have become stronger (Rodo et al, 2002). Pascual et al. (2000) examined the association between cholera and ENSO using 18-year records from Bangladesh. Monthly time series data for cholera incidence between January 1980 and March 1998 and SST anomaly data in the equatorial Pacific was used to create an ENSO index during the same period. A significant correlation was observed between cholera and ENSO, with strong associations occurring during more pronounced ENSO fluctuations (Pascual et al., 2000). This is largely due to a relationship between cholera cases in Bangladesh and sea surface temperature in the Bay of Bengal. Copepods bloom in response to warming sea surface temperatures that are associated with ENSO (Patz et al., 2005, Kovats, 2003). However, the possible mechanism by which increased SST increases disease transmission is not greatly understood.

Forecasting the risk of a cholera outbreak

ENSO events are becoming easier to predict with dynamic and statistical climate models. Models can predict the likelihood of typical ENSO temperature and precipitation anomalies. The magnitude of an event, however, has high levels of uncertainty and the severity of impacts is hard to predict (McPhaden et al., 2006). Pascual et al. (2008) attempted to predict cholera outbreaks with ENSO data retrospectively by using a semi-mechanistic time series model that incorporates ENSO and cholera disease dynamics. The results illustrate that ENSO is a key covariate and confirms its interplay with immunity levels. It is suggested that when V. cholera is considered in predictive models a robust early warning system for cholera in endemic regions can be developed (Constantin de Magny et al., 2008). The feasibility of using a model as a forecasting tool is established and it is argued that regardless of model limitations valuable information for the development of risk management strategies for climate sensitive diseases can be provided (McPhaden et al., 2006). Therefore, the environmen-

Figure 4. Time-scale model for El Ni単o early warning Systems; Source: Kovats, 2000. FIELD NOTES | VOL I | 25

tal determinants of cholera epidemics can help provide a foundation to build predictive mechanisms for cholera outbreaks and reduce regional vulnerability. If ENSO is considered in predictive models a robust early warning system for cholera in endemic regions can be developed (Constantin de Magny et al., 2008). The feasibility of using a model as a forecasting tool is established and it is argued that regardless of model limitations, valuable information for the development of risk management strategies for climate sensitive diseases can be provided (McPhaden et al., 2006). Therefore, the environmental determinants of cholera epidemics can help provide a foundation to build predictive mechanisms for cholera outbreaks and reduce regional vulnerability. Kovats et al. (2000) propose early warning systems for enhancing public health measures in Bangladesh and other at risk areas. Seasonal forecasts indicate areas where the probability of some deviation from the climate mean is increased, such as wet/dry or warmer/colder conditions (Kovats et al., 2003). For operational use, however, seasonal climate trends are not sufficient due to the chaotic nature of the climate system. Local weather conditions, therefore, are needed to provide early warning of epidemic risk (Patz et al., 2005). In addition, it is argued that seasonal forecasts are more accurate during ENSO events in comparison to other times (Kovats, 2000, Kovats et al., 2003) This is likely because it is known that ENSO SSTs in the second half of the year persist for two seasons and the evolution can be monitored quite accurately (McPhaden et al., 2006). Early warning systems can help limit health impacts because they can influence response times. Figure 4 depicts the time-scale of El Ni単o early warning systems with both weather and climate. 26 | ALLNUTT

In addition, given that cholera occurs from ingesting contaminated food or water, it is not always possible to provide adequate treatment in a timely manner once the epidemic has begun (Cash et al., 2010). Consequently, it is suggested that better methods of forecasting cholera risk would be beneficial to societies that face cholera outbreaks, as medical supplies could be prepared in advance. For example, in 1997 the regional WHO Cholera Surveillance Team was aware of an El Ni単o related drought forecasted for South-East Africa and responded by instituting measures to decrease the severity of the outbreak by preparing health care institutions and increasing supplies (WHO, 1999). Thus, disease surveillance and forecasting systems can be effective in reducing vulnerability to cholera associated with climate extremes as the ability to predict high or low transmission seasons can help target the timing and place of public health interventions. Although research suggests the association between cholera and ENSO can act as a predictive mechanism for cholera outbreaks, there are other key issues in the prevention of the disease that should not go unmentioned. Cholera prevention is a multifactorial disease and prevention can be targeted to a variety of factors. For example, access to clean water, proper sanitation methods, health education and good food hygiene are important prevention measures (WHO, 2011). Proper water infrastructure is important as cholera outbreaks generally emerge in areas that lack the necessary infrastructure, such as proper plumbing systems (CDC, 2011). The flooding associated with El Ni単o can cause damage to local infrastructure such as sewage systems and water supply systems, which can exacerbate the epidemic (WHO, 1999). Moreover Emch et al. (2010) posit

that local socio-economic status (SES) modifies the effect of regional environmental forces in Matlab, Bangladesh. SES has a significant impact on cholera incidence, with lower incidence rates in higher SES households (Emch, 2010). Thus, there are important social factors such as infrastructure quality and economic status that contribute to cholera risk. These social determinants of the disease are important to consider because they can have a large impact on disease prevention. According to infectious disease epidemiology, limitations to the analysis described here emerge because it is rare for a disease to be consistently attributed to one factor. In addition, there are multiple pathways in which endemic cholera can emerge. In order to portray a more complete picture, the internal dynamics of the cholera disease must be more thoroughly researched in relation to ENSO events. Although there is a plausible link between ENSO and cholera, it is suggested that “non-linear disease hosts dynamics can generate cycles of variability independently of climate” (Hashizume et al., 2011). Koelle et al. (2005) mention these dynamics were accounted for and a correlation was observed that suggested an ENSO link. Another body of research, however, that suggests cholera incidence is characterized by internal disease dynamics. For example, Hashizume et al. (2011) fit a semi-parametric times series SIRS model to the observed number of cases and seasonal long-term transmission, and decay of immunity. The model results suggest 37 percent of observed variance is unexplained; that is, there is a “significant fraction of observed variability of cholera incidence that can be attributed to internal disease dynamics” (239). Thus, further research needs to reflect a more complete picture.

How ENSO dynamics will change due to climate change is still speculative (Ohtomo et al., 2010). Collins (2005) posits that there may be an increase in the probability of ENSO-like events due to anthropogenic climate change and Rodo et al. (2002) claim that most observational and modeling evidence suggest an increase in both variability and amplitude of ENSO under global warming model scenarios. Furthermore, it is suggested that areas that are currently strongly affected by ENSO could experiences greater risks if ENSO variability or the strength of ENSO intensifies (Patz et al., 2005). Ohtomo et al (2010) argue that clarifying the correlation between prevalence of infectious disease and climate change is important given our current warming trends. Enhanced warming—including merely the intensification of ENSO—has the ability to change disease transmission by changing human behavior, especially with increased contact with contaminated water before and during the spring where cholera outbreaks normally start in Dhaka, Bangladesh (Rodo et al, 2002). Therefore, risk associated with climate change and ENSO should be investigated further because it can have a significant impact on the emergence of cholera epidemics.


The associations between cholera outbreaks and ENSO climate patterns can have a profound impact on public health. El Niño is associated with increased risk of cholera in Bangladesh because climate anomalies are linked with El Niño. The relationship with El Niño illustrates the ecological basis of cholera in Bangladesh and represents one pathway of cholera emergence. However, the social determinants of the disease need to be examined in conjunction with the environmental determinants, since they are other factors FIELD NOTES | VOL I | 27

that play a big role in the disease transmission, something that is often overlooked in the ENSO literature and future research should consider. Moreover, since cholera incidences appear to increase following winter El Niño event there is a window of opportunity to improve cholera risk through forecasts. In areas where El Niño can be reliably associated with regional climate variations, such as Bangladesh, forecasting can provide decision makers with early warning of increased risk, as seen in the aforementioned WHO Cholera Surveillance Team example. That is, the effect ENSO has on cholera can be a useful predictor of disease outbreaks and can help preparedness for disease outbreaks and reduce population vulnerability to climate extremes. Thus, the relationship between the two needs to be considered by policy makers and healthcare professionals because of its use in predicting epidemic risk and can determine location and timing of public health interventions. Although the notions presented here focus on Bangladesh the concepts can be transferred to other locations that are vulnerable to the effects of El Niño climate variability, such as Africa and South America. El Niño affects sea surface temperatures around the world, which may increase risk for a disease epidemic in non-endemic regions.

REFERENCES Cash, B.A., Rodo, X., Kinter, J.L (2008). Links between Tropical Pacific SST and cholera incidence in Bangladesh: Role of the Eastern and Central Tropical Pacific. Journal of Climate, 21, 4647-4663. Cash, B.A., Rodo, X., Kinter, J.L.,& Yunus, M. (2010). Disentangling the impact of ENSO and Indian Ocean variability on the regional climate of Bangladesh: Implications for cholera arisk. Journal of Climate, 23, 2817-2831.


Collins, M. (2005). El-Nino or La Nina like climate change? Climate Dynamics, 24, 89104. Colwell, R.R., (1996) Global climate and infectious disease: the cholera paradigm. Science, 274, 2025–2031. CDC, (2010). Centres for Disease Control and Preventions disease index: Cholera. Retrieved online at http://www.cdc.gov/ cholera/index.html October 19, 2011. Constantin de Magny, G., Murtugudde, R.,Sapiano, M.,& Colwell, R. (2008). Environmental signatures associated with cholera epidemics. PNAS, 105(46), 1767617681. Emch, M., Yunus, M., Escamilla, V., Feldacker, C.,& Ali, M. (2010). Local population and regional environmental drivers of cholera in Bangladesh. Environmental Health, 9(2) Hashizume, M., Faruque, A.S.G., Terao, T. (2011). The Indian Ocean Dipole and Cholera Incidence in Bangladesh: A TimeSeries Analysis. Environmental Health Perspectives, a119(2), 239–244. Koelle, K., Rodo, X., Pascual, M., Yunus, M.,& Mostafa, G. (2005). Refractory periods and climate forcing in cholera dynamics. Nature, 436(4), 696-700. Kovats, R.S., Bouma, M.J., Worrall, E., Haines, A. (2003). El Niño and health. The Lancet. aRetrived online at http://image. thelancet.com/extreas/02art5336.web.pdf Kovats, R.S. (2000). El Niño and human health. Bulletin of the World Health Organization, 78(9), 1127-1136. Kump, L.R., Kasting, J.F., & Crane, R.G. (2009). The Earth System, 3nd ed. PrenticeHall. Lipp, E.K, Huq, A., Colwell, R.R. (2002). Effects of global climate on infectious disease: The cholera model. Clinical Microbiology Review, 15, 757–770 McPhaden, M. J., Zebiak, S. I., & 2 Glantz M.I. (2006). ENSO as an integrating concept in earth science. Science, 314(1740),

1740-1745. Pascual, M., Rodo, X., Ellner, P., Colwell, R., & Bouma, M.J. (2000). Cholera dynamics and El Nino Southern Oscillation. Science, 289, 1766-1769. Pascual, M., Chaves, L.F. Chaves, Cash, B., Rodo, X., Yunus, M. (2008). Predicting endemic cholera: The role of climate variability and disease dynamics. Climate Research, 36, 131-a140. Patz, J.A, Gibbs, H.K., Foley, J., Rogers, J.V., & Smith, K.R. (2007). Climate change and global health: Quantifying a growing ethical crisis. Eco-Health, 4, 397-405. Patz, J.A., Cambell-Lendrum, D., Holloway, T., & Foley, J.A. (2005). Impact of regional climate change on human health. Nature, 438(17) 310-317. Ohtomo, K., Kobayashi, N., Sumi, A., Ohtomo, N. (2010). Relationship of cholera incidence to El Ni単o and solar activity elucidated by time-series analysis. Epidemiol. Infect., 138, 99-107. Rodo, X., Pascual, M., Fuchs, G., Faruque, A.S.G. (2002). ENSO and cholera: A nonstationary link related to climate change? PNAS, 99(20), 12901-12906. Ruiz-Moreno, D., Pascual, M., Emch, M., Yunus, M. (2010). Spatial clustering in spatiotemporal dynamics of endemic cholera. Infectious Diseases, 10(15), 1417-1487. Schoolnik, G. (2008). Environment degradation begets epidemics: Cholera in Bangladesh. Retrieved online December 2, 2011 at http://med.stanford.edu/medcast/2008/ cholera-bangladesh.html WHO (1999) Task force on climate and health: El Ni単o and Health. Geneva 1999. Retrieved online http://www.who.int/ globalchange/publications/en/elnino.pdf October 19, 2011. WHO (2011) Health topics: cholera. Retrieved online at http://www.who.int/topics/cholera/en/ October 19, 2011.


Review of Temporal GIS


Temporal Geographic Information Science (GIS) deals with the modeling of spatial phenomena over time and the indexing of this temporal information into databases.Various conceptualizations of time have been resolved into a linear path structure due to the limitations of GI systems. Fundamental differences between time and space necessitate separate ontologies and topological structures. Data models that have been adapted to accommodate temporal information and initial applications are discussed in terms of their ability to manage uncertainty, to store data efficiently, and their temporal continuity.


30 | WONG


ractitioners of Geographic Information Science (GIS) have grappled with how to model and index changes over time into databases. The issue of temporality intersects with many other aspects of GIS, such as scale, uncertainty, ontology, data models, and database design. The results of these debates, in turn, affect many practical domains, such as environmental modeling, transportation, and location-based services (Dragicevic and Marceau, 2000; Shaw, 2010; Tattoni et al., 2010). The inclusion of time in analyses allows for powerful and informative explorations of causality (Peuquet, 1994). In discussing temporal GIS, important terms include entity, state, event, and granularity. Entities are the abstraction or computer representations of real-world things (Wang et al., 2004). State or version is used to refer to the specific condition of an entity at a given time (Frihida et al., 2002; Wang et al., 2004). Events or mutations are the occurrences that change the state of an entity (Wang et al., 2004). Granularity refers to the resolution at which something is being observed, where a fine granularity indicates a lower scale and smaller temporal or spatial units (Xu et al., 2004). These definitions beg the question: what is temporal GIS? In this context, temporal GIS is defined as an analytic approach that incorporates temporal topology into traditional spatial databases and allows queries along the temporal axis. It explicitly recognizes different versions of entities, maintains the history of an entity’s changes, and strives to identify when state-changing events occur. The development of tools that answer the above criteria is important in establishing patterns and improving our understanding of complex processes, such as those found in landscape ecology. For example, the abil-

ity to model state-changing events can allow researchers to predict with greater precision when a forest fire is likely to occur and what its impacts will be. This paper will review the significance of temporality in geographic phenomena and provide a summary of the development of temporal GIS applications. Further, major strains of thought on how to represent time in GI systems will be discussed, bearing in mind the formative force of technical limitations. Finally, the implications of these representations will be explored in terms of ontologies and uncertainty. In particular, object and raster data models will be examined in terms of their ability to address fundamental differences between time and space. Throughout the paper, the three characteristics of (1) Temporal continuity: the ability to trace an object through various states in time, (2) Hierarchical organization: the organization of events and states as logical results of prior events, and (3) Incompleteness: the uncertainty of the position of objects through time as identified by Peuquet (1994) will tie salient points together.


Temporal GIS finds its roots in philosophy, where thinkers postulated various conceptions of time (Langran and Chrisman, 1988). Notable visualizations of time include multiple parallel tracks, tree structures, circularity, and discreteness (ibid.). Moreover, planners concerned with migration, optimal routing, and efficient planning of services assessed the intersection of time and space (Hagerstrand, 1970). A classic application of GIS involves the “travelling salesman� problem, where a number of stops must be made along a network (i.e. roads; Jung, Lee, and Chun, 2006). The network is analyzed to FIELD NOTES | VOL I | 31

find the shortest path that takes the user to each stop (ibid.). Notably, Hagerstrand (1970) suggests a tree-like structure of time, where the number of branches of possible outcomes or paths is limited by constraints. These constraints are based on the geographic distance an individual can reach in an allotted amount of time and on the intersections of paths of different entities (ibid.). Given these constraints, one must choose a single path out of all possible branches (ibid.). This path-centric conceptualization envisages GIS as a way to track the history of a path and to identify the previous states and decisions (events) that led to its current status. The linking of temporal processes with geographic analyses (i.e. maps) was further solidified by diffusion theory (Peuquet, 1994). This sociology-oriented definition of diffusion sought to map the spreading pattern of phenomena through time to establish causal links (ibid.). Again, attention is directed to the spatiotemporal paths of individual entities and enabled researchers to map, for instance, the spread of AIDS or how farmers have been impacted by the Green Revolution (ibid.). Researchers in these fields, however, soon discovered several technological problems of modeling complex interactions between multiple entities at various scales. For example, temporal patterns that were cyclical or random were difficult to model due to a paucity of knowledge of processes at work (Peuquet, 1994). These limitations were temporarily amended with qualitative descriptions (ibid.).


Initial forays of incorporating time into GIS analyses typically took a “snapshot” approach, which compiles a chronologically ordered series of images of the same spatial extent at different times

32 | WONG

(Goodchild et al., 2007). By comparing consecutive images, researchers are able to identify the state of a pixel on any of the snapshots (Peuquet and Duan, 1995). The snapshot method, however, is limited to only presenting various states and there is no recognition or tracking of an entity’s history (ibid.). Furthermore, the events that cause state changes are not explicitly identified, which prohibits the user from querying by event (Peuquet and Duan, 1995). In general, the snapshot method can answer the question: “Is a section of the Arctic covered by ice at Time 1 and Time 2?” The snapshot method, however, is not able to answer: “how has the section of the Arctic’s ice cover changed from Time 1 to Time 2?” (Langran and Chrisman, 1988; Peuquet and Duan, 1995). This lack of continuity is troubling for users of remote sensing, which has, for example, led to the development of change matrix tools in ENVI software that track state changes of pixels between two snapshots (Lung and Schaab, 2006). These manipulations typically generate a one-time report of statistics (reflecting a snapshot approach) instead of a functional database that tracks entities through time. Deriving significance from several change matrix reports is also a visualization issue, as user abilities and preferences must be considered. Lastly, since the snapshot method records every entity for each time interval regardless of whether or not the entity has changed, redundancy in data storage is a major concern (Peuquet, 2001). As a result of the above limitations, a temporal topology was recognized as necessary to move beyond limited exercises in temporal GIS (Langran and Chrisman, 1988). Recall that topology identifies the spatial relationships between features (such as whether two roads intersect) and allows for error checking (ibid.). A temporal topology would thus link an entity

with its past versions/states and maintain logical consistency rules (e.g. a forest object cannot become a desert object without spending time as a transitional grassland/shrubbery object; ibid.). The ability to track an entity’s temporal neighbours (its previous states) is key to satisfying the condition of temporal continuity where cause-effect events link an entity from one state to the next (Hagerstrand, 1970; Langran and Chrisman, 1988)


The increasing proliferation of data produced with temporal information has prompted renewed focus on temporal GIS. Advances in remote sensing, more affordable sensors, and real-time streaming services have resulted in a rich mine of unorganized spatio-temporal data. For example, Twitter yields data in the form of ‘tweets’ to users in chronological order and along a social network composed of accounts the user has subscribed to. This data is useful in that an observer can analyze where and when reports have been generated. For example, an individual living on top of a hill may report and tweet the event of a fire in a nearby neighbourhood faster than others. The location of the second wave of “re-tweets” is also compelling in identifying whether or not there is a shared geographic commonality that motivated a user to re-tweet an event. For the average user, the streaming of tweets means that much of the data are no longer visible after a relatively short period of time. For researchers that deliberately harvest tweets, a problem of translating coordinates or an IP address to a street address (geo-coding) exists: does one geo-code by the IP address within each tweet or by the geographic content of the message? Challenges such

as this have renewed focus on database design that integrates time and space in meaningful ways (Goldberg, 2011). The problem of unifying temporal data with spatial data lies in the fundamental differences between the two. Whereas objects can move in multiple directions in space (left, right, forward, and back), the temporal axis is unidirectional (Xu et al., 2004). Time is also measured in different units than space, with the former tending to have a larger granularity (discrete seconds, minutes, hours, etc.) than continuous space (ibid.). As a result, resolving issues of temporality is not accomplished by simply extending a relational table model to include a column for time (Shaw, 2010). Treating temporality and an attribute (e.g. elevation) equally ignores the dimensionality and dependency between entities and time. For example, a given forest’s foliage, biodiversity, and extent changes over time and are entirely dependent on their past instances.


The use of raster and vector data models in temporal GIS has documented advantages and disadvantages. The raster model grids the entire extent into cells where each cell maintains a record of its state changes and the time at which these changes occurred (Peuquet and Duan, 1995). Since a cell or an aggregation of cells represents every spot in the extent, the raster data model is suitable for location-based queries of change over time (ibid.). In addition, advanced mathematical functions such as Bayesian Maximum Entropy allow for probability-based interpolation between scales as well (Christakos et al., 2001, pg. 4). The raster data model is easily integrated with remote sensing applications, where most images consist of pixels. On FIELD NOTES | VOL I | 33

the other hand, the vector model represents entities of interest as discrete objects (Renolen, 2002; Goodchild et al., 2007). The focus of Hagerstrand’s model on time pathways and directionality lends itself well to the vector data model (Egenhofer and Golledge, 1994). An object-based model is arguably intuitive to humans in the way that entities are perceived and can contain an entire state history in one contiguous unit (Renolen, 2002). As mentioned, however, redundancies in storage can occur if each instance of an object for every time interval, regardless of whether a change has occurred or not, is recorded. Furthermore, an element of uncertainty exists in the visualization of an object’s boundaries (Goodchild et al., 2007). Over time, for example, the area of households that are serviced by a bus route can shift, expand, and split into discontinuous shapes in response to user preferences, road conditions, and cost of transporta-

tion. How a user defines what constitutes a single entity and what event precipitates a new, separate entity has profound impacts on the temporal database in terms of continuity. Importantly, questions arise as to how best to communicate the uncertainty present in geographic data. In a bid to solve this issue, Goodchild et al. (2007) have discussed how phenomena can be interchangeably perceived as fields and objects in an attempt to unify the two data models into a more robust composite for tracking change over time. The object-oriented model has been popularized and extended for applications such as land cover change. This popularity is due in part to convenient software such as ECognition that automate the segmentation of images into objects (Trimble, 2010). To track changes explicitly, Langran and Chrisman (1988) have proposed alternative vector models that only show new state changes. The first method involves a blank base map of the original data set. New maps (either past or future) are overlaid onto the base map where the only features are new changes of state as shown in Figure 1 (Langran and Chrisman, 1988). Such a method stores layers only when a new object version is created (a temporally significant event) and keeps track of when the change occurred (ibid.).

Figure 1. Langran’s overlay model (Langran and Chrisman, 1988)

Figure 2. Langran’s space-time composite model (Langran and Chrisman, 1988)

34 | WONG

The second model shown in Figure 2 begins with an original dataset of objects. A change in state of any part of these objects results in the creation of a new object, thus fragmenting the original dataset over time (Langran and Chrisman, 1988). The geometry of each state is tracked throughout time as shown in Figure 2. The principal point of these two models is that they store a continuous temporal topology of their objects in their rural or urban states (Peuquet, 1994). It is possible, therefore, to ask not only when an object changed from an urban to rural state but also what changed in a period of time. In other words, this model incorporates temporal topology into spatial databases and allows queries along the temporal axis.


The interdisciplinary nature of temporal GIS has led to collaboration on many aspects of critical GIS, which examines the impacts of implicit assumptions made within geo-spatial principles. Given the growing sources of heterogeneous temporal data, discussion on how to conceptualize, categorize, and relate data have emerged (Xu et al., 2004). These ontological debates have emphasized the need to make flexible, interoperable databases (ibid.). The possibility of objects changing from one state to another, new objects emerging from old objects, or changing size and position must be accommodated in temporal GIS (Renolen, 2002). To address these challenges, a hierarchical structure has been developed to facilitate a logically consistent transition flow of entities from one state to another (Goodchild et al., 2007). A general taxonomy has also been presented to allow subclasses and inheritance, with categories such as Uniform Stationary Rigid, Evolving Moving Elastic, and

Uniform Moving Rigid to describe heterogeneity, mobility, and dynamism (ibid.). At a database design level, it is recommended that objects maintain records of their temporal position relative to other objects (i.e. past and future neighbours) to ensure temporal continuity (Frihidi et al., 2002). Improvements in this regard include intelligent self-learning by the software to enable prediction of future states (ibid.). Wang et al. (2004) suggest a modular approach in which temporal, spatial, and thematic data are inter-related but stored separately. A modular system avoids storage redundancies such as identical thematic states for a given time range, while enabling powerful queries on the basis of Object ID, Version ID, or Event IDs (Peuquet, 2001; Wang et al., 2004) At a higher ontological level of domain classification, experimentation has been carried out to specify separate ontologies for objects and events (Xu et al., 2004). Xu et al. (2004) suggest that specialized ontologies that take advantage of semantic meanings are needed to combine data from different sources. For example, linking phrases such as “affects whom,” “when,” and “how” can elucidate the relationship between an event and the entities it affects. Xu et al. (2004) also suggest more detailed semantic phrases that adjust accordingly to the scale of observation, so that a smaller spatial scale is supported by more ontological relationships or a finer “semantic granularity.” The lack of functional temporal geodatabases is a reflection of how GIS has developed. Traditionally used by city planners and engineers, ArcGIS excels at solving routing problems and this bias is evidenced in its available tools (Shaw, 2010). As of the 10.0 version of ArcMap, the Network Analyst extension features three subcategories and numerous tools to manipulate data. Conversely, the TrackFIELD NOTES | VOL I | 35

ing Analyst extension has only two tools, which are primarily oriented towards appending a timestamp column to an existing dataset. While these temporal GIS manipulations may suffice for expressing a bus route or tracing the paths of migratory birds, complex analyses such as those in landscape ecology struggle with these simple tools (Lung and Schaab, 2006). A unilinear view of time, in other words, makes it difficult to assess seasonality and to identify certain entity states as part of a larger pattern over time (Hagerstrand, 1970). The lack of weight given to temporality is again evidenced by the timestamps for various extents of the globe in Google Earth. As of October 2012, there is no option to view the extent in years other than the most recent capture, and the image date is relegated to the corner of the interface in small text. The paucity of information and poor under-representation of patterns as dynamic and changing engenders a public perception of maps as static, immutable, and complete. It is in such a lacklustre socio-historic context that the practical development of temporal geo-databases has stalled.


The adoption of the object data model has necessitated an acceptance of spatial and temporal uncertainty in object boundaries. Spatial uncertainty arises from instrument accuracy and representation issues (e.g. representing a building as a point or a shape), while temporal uncertainty stems from the chronic incompleteness of a dataset (Peuquet, 1994). Aside from technical limitations, decisions on how to spatially represent an entity may reflect local politics such as disputed borders or a conscious decision to be appropriate/intuitive for the user’s purposes.

36 | WONG

Spatial uncertainty and public demand has led to biased development of real-time location-based services that use points and line routes (Frihida et al., 2002). These features are simpler to model than polygon features whose area may change over time (ibid.). To adjust for uncertainty, fuzzy sets have been utilized to represent transitional phases through time that may later be aggregated into trends (Dragicevic and Marceau, 2000; Wang et al., 2004). Fuzzy sets embrace uncertainty by assigning probabilities that an entity is one thing or another (ibid.). A lack of temporal resolution due to a coarse granularity has also prompted the use of interpolative models. For instance, healthcare practitioners who might only collect data on a yearly basis may need to know which month heralds the beginning of flu season. In such situations, predictive and stochastic models are used to provide some answers (Urban and Wallin, 2002). The Markov method is a common instance of interpolative fuzzy logic, where objects at T2 are assigned a probability of being a certain state depending only on the distribution of states at T1 (Urban and Wallin, 2002). This simple model may be expanded and executed in piecemeal fashion to simulate different rates of state change at various time intervals (ibid.). Through a fuzzy logic approach, Dragicevic and Marceau (2000) interpolate the intermediate states of urban and rural objects between two given image snapshots to reduce temporal uncertainty and incompleteness. It is important to note that the chosen temporal resolution or granularity has implications on interpretations; an overly large decadal time interval may generalize over seasonal patterns, while a very small time interval may present too much variability and obscure general trends (Peuquet 2001). These principles of temporal GIS were

applied in a personal effort to model land cover change in Ecuador. Satellite imagery for a north-western region named Intag was obtained for three different year intervals and classified into different types of land use using remote sensing techniques. A Markov model designed to predict land cover was calibrated using the first two time periods. The model was then run to predict the landscape for the last time interval, and the results were compared to the actual image to assess the model’s accuracy. This project allowed for iterative interpolations of states in-between the given time intervals, thus increasing the granularity of the dataset. The project’s accuracy rate of 41 percent, however, demonstrates the difficulties of temporal GIS. Stochastic interpolation requires strong knowledge of how the subject matter behaves and which events are possible. In modeling deforestation of an Ecuadorean cloudforest, it is arguable that a first order Markov method is more representative of a forest undergoing the regrowth of succession (Urban and Wallin, 2002; Tucker and Anand, 2004). Similarly, first-hand knowledge is often required to learn of unexpected anthropogenic events that can cause state changes (Wilson, Sarah, pers. comm. 2011). Finally, differences in data quality of the original datasets (newer sensors tend to be of higher quality than older ones) and the difference in image capture times complicates the compilation of information into one database. In short, incorporating a temporal dimension requires careful consideration of the rules that govern an extent to ensure a consistent, hierarchical organization as expressed earlier.


In the paradigm-shifting environment of cloud-based computing, interactive applications, interoperability, and growing

sources of data, the average user is likely becoming more aware of issues pertaining to temporality. Development is mainly directed to service an instantaneous, ondemand attitude of ‘when the next bus arrives’ (Shaw, 2010). GI scientists face the challenge of tracking changes over time, grounding them in space, and socially engineering average users to understand the geography of long-term patterns to inform more sustainable practices. The inclusion of time in GIS represents an additional dimension that can provide a significant amount of useful information in establishing causality of phenomena. To take advantage of new sources of temporal data, databases must be designed to protect temporal topology and allow queries of temporally significant events (Langran and Chrisman, 1988). Improvements in algorithms and integration of agent-based modeling promise less uncertain forecasts of spatio-temporal processes such as landscape ecology (Dragicevic and Marceau, 2000; Christakos et al., 2001). As has been demonstrated throughout this paper, future development of temporal GIS should be directed towards hierarchical, interoperable ontologies and effective representations that emphasize temporal continuity in order for this to occur.

REFERENCES Christakos, G., Bogaert, P., & Serre, M. (2001). Temporal GIS: Advanced Functions for Field-Based Applications. Berlin: Springer-Verlag. Dragicevic, S. & Marceau, D. (2000). An application of fuzzy logic reasoning for GIS temporal modeling of dynamic processes. Fuzzy Sets and Systems, 113(2), 69-80. Egenhofer, M., & Golledge, R. (1994). Time in Geographic Space: Report on the Specialist Meeting of Research Initiative 10. National Centre for Geographic Information


and Analysis. Report 94-9. Web. Accessed from http://www.ncgia.ucsb.edu/Publications/Tech_Reports/94/94-9.PDF Frihida, A., Theriault, M., & Marceau, D. (2002). Spatio-temporal object oriented data model for disaggregate travel behaviour. Transactions in GIS, 6(3), 277-294. Goldberg, D. (2011). “Advances in Geocoding Research and Practice”. Transactions in GIS, 15(6), 727-733. Goodchild, M. F., May, Y., & Cova, T. (2007). Towards a general theory of geographic representation in GIS. International Journal of Geographic Information Science, 21(3), 239-260. Hagerstrand, T. (1970). What about people in regional science? Papers of the Regional Science Association, 24(1), 6-21. Jung, H., Lee, K., & Chun, W. (2006). Integration of GIS, GPS, and optimization technologies for the effective control of a parcel delivery service. Computers & Industrial Engineering, 51(1), 154-162. Langran, G., & Chrisman, N. (1988). A framework for temporal geographic information. Cartographica, 25(3), 1-14. Lung, T., & Schaab, G. (2006). Assessing fragmentation and disturbance of west Kenyan rainforests by means of remotely sensed time series data and landscape metrics. African Journal of Ecology, 44(1), 491-506. McGuirk, P., & O’Neill, P. (2012). Critical Geographies with the State: The Problem of Social Vulnerability and the Politics of Engaged Research. Antipode, 44(4), 13741394. Peuquet, D. (1994). It’s about time: A Conceptual Framework for the Representation of Temporal Dynamics in Geographic Information Systems. Annals of the Association of American Geographers, 84(3), 441-461. Peuquet, D. J., & Duan, N. (1995). An eventbased spatiotemporal data model (ESTDM) for temporal analysis of geographical data. International Journal of Geographic Information Systems, 9(1), 7-24.

38 | WONG

Peuquet, D. (2001). Making space for time: Issues in space-time data representation. Geoinformatica, 5(1), 11-32. Renolen, A. (2002). Modelling the Real World: Conceptual Modelling in Spatiotemporal Information System Design. Transactions in GIS, 4(1), 23-42. Shaw, S. L. (2010). Geographic information systems for transporation: from a static past to a dynamic future. Annals of GIS, 16(3), 129-140. Tattoni, C., Ferretti, F., & Cantiani, M. G. (2010). Monitoring spatial and temporal patterns of Paneveggio forest (northern Italy) from 1859 to 2006. Biosciences and Forestry, 3(1), 72-80. Trimble (2010). eCognition Developer 8.64.0 Reference Book. Munich: Trimble. Tucker, B. C., & Anand, M. (2004). On the use of stationary versus hidden Markov models to detect simple versus complex ecological dynamics. Ecological Modelling, 185(1), 177-193. Urban, D. L., & Wallin, D. O. (2002). Chapter 4: Introduction to Markov Models. Learning Landscape Ecology. New York: Springer-Verlag. Wang, S., Nakayama, K., Kobayashi, Y., & Maewkawa, M. (2004). Considering events and processes within GIS: An event-based spatio-temporal data model. IEEE International Symposium on Communications and Information Technologies, 1(2), 770-773. Xu, W., Qin., Y., & Huang, H. K. (2004). Spatio-temporal ontology oriented to geographic information system. Proceedings of the 2004 international conference on machine learning and cybernetics, 1(7), 1555-1560.


The Impact of French Colonial Rule on Contemporary Food Security in the Central Sahel

COREY LESK ABSTRACT. Since the 1960s, the central Sahel region of West Africa encompassing

Niger, Mali and Burkina Faso is experiencing recurring widespread famine due to food and fodder shortages resulting from inadequate or inconsistent rainfall. There are two possible explanations for this precarious state of the central Sahelian population. The first is that the present dry period is a particularly severe one which has drastically altered the environment to which Sahelian peoples were once adapted. The second is that prolonged droughts are a common occurrence in the Sahel and that the adaptive capacities of its peoples have somehow been reduced. This study attempts to explain contemporary hunger in the central Sahel by arguing in favour of the latter view. Extreme rainfall variability is intrinsic to the central Sahelian climate and the current dry period falls well within ordinary variation. Economic, ecological and sociocultural adaptations have for millennia provided the people of the central Sahelian with the capacity to survive frequent and sustained droughts, and the disruption of these systems by invasions and conquests has historically resulted in their failure. French colonial rule (approximately 1875-1960) profoundly disturbed central Sahelian drought resilience by ending the lucrative transSaharan trade, co-opting institutionalized and informal food redistribution, and replacing sound agronomic systems with a destructive and unreliable cash crop economy. Though not the sole instance in history of such a destabilization, it was a particularly devastating one whose effects are still being felt today.

40 | LESK


s Niger, Chad, Burkina Faso, Mali, and northern Nigeria enter another dry season with food shortages (All Africa, 2011; UN News Center, 2011), past vulnerabilities of the central Sahel linger in recent memory. Chad and Niger suffered severe food and pasture shortages in 2010 following a growing season crippled by a cruel juxtaposition of torrential downpours and drought (Rice, 2010; Foy, 2010). Failed rains in 2004 coupled with unfavourable terms of trade pushed parts of Mali, Burkina Faso, and Niger into famine during the 2005 lean season, the time of year before harvest when food supplies are low (Doyle, 2005). This recent fragility in the ability of the central Sahelian population to feed itself, which may be referred to as food insecurity, extends back in time to the late 1960s when precipitation patterns began to trend markedly into deficit (Dai et al., 2004). The much-publicized famine of 1968-1973 is attributed to this desiccation (Williams, 1974; Ottaway, 1973). The association between climate and hunger in the central Sahel invites the question of how a population could come to exist in such a precarious state. Either of two dichotomous scenarios may offer an explanation. The first possibility is that the present dry period is an anomaly which has drastically altered the environment to which Sahelian peoples were once adapted. The second is that prolonged droughts are a common occurrence in the Sahel and that the adaptive mechanisms of its peoples have somehow been altered. This study will argue in favour of the latter scenario by examining a sampling of the climatic, archaeological, historical and anthropological literature on the central Sahelian economy and society in order to present a bifurcated thesis. To begin, it will be argued that the current dry period is not abnormal for the Sahelian cli-

mate and that the Sahelian peoples of the pre-colonial period managed to survive droughts of equal magnitude. Then, the food insecurity experienced by contemporary central Sahelian countries will be attributed to the profound disruptions of French colonial rule, first in terms of traditional modes of subsistence and trade and then with regard to the sociocultural system of protection against famine.


The Sahel is a semi-arid ecological and climatic zone located between the southern margin of the Sahara and the northern edge of the Sudanese savannah, extending from the Atlantic Ocean to the Red Sea. This study will focus on the central Sahel within the borders of contemporary Mali, Niger, and Burkina Faso. The term western Africa will refer to the ensemble of the Sudanese savannah, the Sahel and the Sahara desert. Rainfall decreases nearly continuously from south to north across the Sahel producing a variety of geographies and populations (Raynaut, 1997). For simplicity, this study will consider the Sahel as the pairing of a drier northern steppe region, characterised by relatively low population density and a nomadic pastoralist population, and a southern savannah region, characterised by higher population density and a mainly sedentary agriculturalist population. The first part of the study examining historical droughts and adaptations will be concerned with the time period ranging from the Neolithic era to the invasion of the Songhai Empire in the 16th century. The investigation into colonial disruptions of Sahelian society will refer to the period beginning with the French expansion into the central Sahel in the late 19th century through to the present day. FIELD NOTES | VOL I | 41

In order to consider the etiology of contemporary food insecurity in the Sahel, the current drought must be situated in the greater history of western African climate and human settlement. Studies into the paleoclimate of the southwestern Sahara and the Sahel estimate that about 10,000 before present (bp), after roughly eight millennia of arid conditions (Rain, 1999), the Sahara began to experience a humid period which resulted in the northward spread of the Sudanese savannah (Vernet, 2002). A desiccating trend began in about 7000 bp and by 4000 bp, most of the savannah of the green Sahara had returned to desert or semi-arid Sahelian ecosystems. Thus, the climate of western Africa is seen to be dynamic over long time periods. Another feature of the western African paleoclimate was a tendency towards deep and sustained droughts on the order of centuries. Nick Brooks (2006) reports that the green Sahara period was interrupted by “a period of aridity in the Sahara . . . that lasted for several centuries” (p. 2). The end of the green Sahara period was marked by “a severe drought, lasting several centuries” (Vernet, 2002, p. 47), while “severe dry spells occurred between 3800 to 3600 bp” (Breuning & Neumann, 2002, p. 125). A reconstruction of precipitation data from lake sediments in Ghana suggests that “the severe drought of recent decades is not anomalous in the context of the past three millennia” (Shanahan et al., 2009, p. 1). Karl Butzer’s (1984) review corroborates this finding, adding that “the Sahel drought of 1968 . . . falls well within the range of short- and medium-term variability” (p. 6). The archaeological and historical records of human inhabitation of the savannah-Sahel-Sahara ensemble demonstrate a history of adaptation to climate fluctuations and resistance to drought. A promi42 | LESK

nent theory on the origin of pastoralism in the Sahel considers cattle herding an adaptation to the desiccation of the green Sahara beginning about 7000 bp (Brooks, 2006). With the onset of semi-arid conditions, the adoption of cattle herding allowed hunter-gatherers to relocate when the environment could no longer support them: “mobile animals were less vulnerable than plants to localized, short-term droughts” (Marshall & Hildebrand, 2002, p. 131). A similar etiology is proposed for the domestication of plants in the southwestern Sahara, which Munson (1976) claims took place as a result of dry conditions of about 3000 bp. Innovation as a means of adaptation to rainfall deterioration thus has ancient origins in semi-arid western Africa. Institutionalized famine prevention measures emerged during in the Sahel during the second millennium A.D. The empires of Ghana (c. 500 A.D. – c. 1200 A.D.) and Songhai (c. 1473 A.D. – c. 1600 A.D.) were adept at organizing their agriculture and economy in favour of drought resistance (Franke & Chasin, 1980; McCann, 1999). Two examples of the devastation that resulted from the disruption of these mechanisms by invading forces illustrate the delicate ecological and economic balance struck by the Ghana and Songhai Empires. During the relatively wet period of approximately 800 A.D. to 1300 A.D. (Webb, 1995), the Ghana Empire thrived and expanded on a base of trade and agriculture (Franke & Chasin, 1980; McCann, 1999). In the eleventh century, the Almoravids invaded Ghana and captured the Sahelian city of Kumbi-Saleh causing its inhabitants, who had kept carefully irrigated gardens, to flee. The invading nomads were oblivious to the methods necessary for sustainable cultivation of the Sahel and ecological destruction en-

sued. As Levtzion (1980) writes, “with the retreat of the sedentary population, the wells were neglected and became filled with sand, the nomads’ herds destroyed the vegetation, and erosion impoverished the barren land . . . [the] balance, so well maintained by the Sudanese, was violated by the nomads” (p. 8). The Songhai Empire fell prey to a similar effect of invasion. From the fourteenth century through the mid-fifteenth century, “efficient organization of agriculture, of communications, and of transSaharan trade brought [the Songhai Empire] peace and prosperity . . . [with no record of ] famine” (Swift, 1977, p. 470). After a Moroccan army invaded and took control of the empire in 1591, “the least . . . drought . . . was liable to turn into a famine. Great famines were recorded every 7 to 10 years during the seventeenth century, every 5 years during the eighteenth century. In 1738 nearly half of the population of [the capital] Timbuktu, and probably of the whole region, died” (p. 471). This period of food insecurity coincided with general political and commercial disarray in the Middle Niger region (Webb, 1995). By the 1620s, the Moroccan invaders abandoned their posts, having failed to make any use of the kingdom (Franke & Chasin, 1980). Adaptation to a difficult environment did not always take place in top-down orientation. Particular skills, practices and mutualistic agreements among agriculturalists and pastoralists in stateless societies of the central Sahel enabled them to sustain themselves through droughts and preserve the fragile Sahelian ecology (Scott, 1984; Raynaut, 1997). Sahelian societies, either stateless or under the Songhai and Ghana Empires, exhibited noteworthy tolerance to the droughts of the pre-colonial period. What changes have made the unremarkable

droughts of the post-colonial period so problematic? The remainder of this study will attempt to explain the contemporary food insecurity of Sahelian societies in terms of French colonial interference. It is important to appreciate that French colonialism was not the first event in the history of the Central Sahel to disrupt a delicate system of drought resistance, but rather one which took place in a context of periodic endogenous (i.e., African) disturbances. What is peculiar about the colonial disruption is that it introduced deep and lasting reorganizations of the economic and social realities of the central Sahel. It will be demonstrated that the current period of sensitivity to drought is largely a result of those changes.


The French colonial endeavour began in the early 17th century on the Atlantic coast of present-day Senegal, where the trade of French cloth, arms, and paper for West African slaves, ivory, animal skins, and Arabic gum grew and thrived (Franke & Chasin, 1980). In the last quarter of the nineteenth century, France began to play its part in the “scramble for Africa,” conquering West African territory by force (Swift, 1977). By 1900, colonial rule had mostly been established from Senegal to Chad (Franke & Chasin, 1980), and French presence began to deeply impact the population of the central Sahel. The Tuareg pastoralists of the northern Sahel and the Sahara had thorough knowledge of the trans-Saharan routes, kept herds of camels and had control over certain tradable dryland resources such as salt and dates. Hence, they “were very much involved with the caravan trade that passed through the Sahara” (p. 44), hiring their herds as pack animals, serving as guides for the trans-Saharan caravans, and FIELD NOTES | VOL I | 43

trading for grain in the southern portions of the Sahel (Lovejoy & Baier, 1976). The Tuareg received “well over half ” of their grain requirements through the trade (p. 155), and invested part of their earnings in slaves, gold, and jewels as buffers against times of scarcity (Caldwel, 1975). The arrival and expansion of the French in the Sahel spelt the demise of the trans-Saharan trade, as the modality of transport was shifted from camel to ship and eventually to road and rail, and a new foreign supply of salt was introduced (Baier, 1976; Franke & Chasin, 1980). By the end of World War I, “the salt trade had become insignificant” (p. 67), and by the 1960s the trans-Saharan trade had all but ceased to exist, a condition that persisted into the post-colonial period. At the dawn of the 20th century, one camel load of salt traded for fifteen to twenty loads of millet, but by the 1970s the trade yielded at best two loads of millet. This collapse in terms of trade coupled with the general dislocation of the Tuareg from the regional economy caused a decline in Tuareg purchasing power, forcing them to become more reliant on livestock as a good to trade for grain at market (Franke & Chasin, 1980). The result was environmental degradation and economic hardship. Enlarged herds came to replace gold, jewels and slaves as stores of wealth which taxed the available pasture, increasing the environment’s susceptibility to scant rainfall – one estimate suggests that by 1970 the Sahelian livestock population was twice the land’s carrying capacity (Swift, 1977). During droughts, the value of livestock tends to plummet due to the deterioration of animal health and the flooding of the livestock market by herders seeking grain. Meanwhile, drought-induced scarcity drives up the price of millet further diminishing pastoralist purchasing power (Swift, 1977). Moreover, after the col44 | LESK

lapse of the salt trade, the Tuareg tended to reorient their herds from camels towards more marketable but less droughtresistant species such as sheep and cattle, whose health declined rapidly at the onset of drought. Thus, colonial interference in regional trade in the central Sahel caused the food security of the Tuareg to become extremely sensitive to livestock price fluctuations, such as those that typically accompany droughts. French colonial policy also transformed the livelihoods of sedentary agriculturalists, to the detriment of their drought resistance capabilities. Beginning with the expansion of the French Empire from the coast into the hinterland in the late nineteenth century, the cultivation of cash crops such as peanuts and cotton for export was encouraged and often forced upon the agrarian population of the central Sahel (Caldwell, 1975). The French colonial authority installed largescale cotton plantations in Upper Volta (now Burkina Faso) and Mali in 1937 and 1945 respectively (Comité d’Information Sahel, 1975; Filipovich, 2001). Explosive growth in cotton output began towards the late 1950s – cotton production in Burkina Faso swelled from 1,650 tons in 1955 to over 37,000 tons in 1973 (Comité d’Information Sahel, 1975), and from 1960 to 1972, the area dedicated to cotton production in Mali nearly quadrupled from 26,000 tons to 90,000 tons (Somerville, 1986). Peanut cultivation was established after World War I in Southern Niger and began to burgeon in the late 1930s (Raynaut, 1997) – in the Maradi region, production expanded from 18,000 tons in 1929 to 182,000 tons in 1968 (Somerville, 1986). Cash cropping came to take precedence in the central Sahelian economy. The proliferation of export-oriented cash cropping was devastating to the en-

vironment. Peanut production in the central Sahel was widespread prior to the colonial era but it was traditionally conducted in alternation with millet and with a minimum fallow period of six years in order to preserve soil nutrients and structure. Whereas the millet harvesting practice left the stalks and root bulbs intact as buffers against aeolian erosion, the peanut harvest loosened the soil and stripped the surface of vegetation, leaving the soil vulnerable (Franke & Chasin, 1980). Under pressure from French colonial authorities to produce more, peasants reduced the fallow period to as little as a single year (Somerville, 1986), crippling the soil’s capacity to hold moisture and withstand erosion. Abbreviated fallow periods also compromised the mutualistic agreement between farmers and pastoralists through which herds were allowed to graze on fallow fields in return for fertilizing the soil with their manure (Caldwell, 1975). In the central Sahelian ecosystem, tree roots play an important role in preserving soil humidity and foliage provides a crucial input of organic matter to the soil. As old cash crop plots became infertile and new lands were deforested in order to cultivate, “a spreading wave of environmental degradation” propagated across the central Sahel (Franke & Chasin, 1980). A similar effect has been documented with regard to cotton production in Burkina Faso (Gray, 2005). The desertion of “traditional agricultural methods for guaranteeing the ecological balance” thus increased the Sahelian peasantry’s food vulnerability (Somerville, 1986, p. 21). Cash crop production flourished not only at the expense of the Sahelian ecology but also at the expense of food production. Agricultural research both under French rule and after independence favoured cash crops while technologies for food production stagnated (Psychas

& Malaska, 1989). Somerville (1986) reports that in the Sahel, “the most arable land [was appropriated] for cash crops” forcing food production onto marginal land (p. 21). Beginning in the 1940s, France encouraged production of peanuts and cotton by offering preferential prices and market protections to West African producers and supplying peanut farmers with tools, fertilizer, and credit. However these privileges did not extend to subsistence crops (Franke & Chasin, 1980). As a result, “the area devoted to millet and other subsistence crops declined” (p. 95), an effect that alarmed colonial authorities concerned with food insecurity. By 1961, the average farmer in central Niger was dependent on peanuts for over half of his income, and throughout the 1980s and 1990s, the area under peanut cultivation in the Maradi department continued to grow at twice the rate of millet, beans, and corn (Rain, 1999). This reality was paralleled in Mali where from 1968 to 1972, the cotton and peanut industries grew by 400 and 77 percent, respectively, corn yields fell by over a third and millet harvests stagnated (Lofchie, 1975). By 1970, Mali, Niger, and Upper Volta were each large part dependent on cash crops, receiving over half of their export revenue from them (Lateef, 1980). Apart from the environmental destruction, cash cropping presented itself as a sound means of maintaining food security in the first decade after independence, as growing income from exports of a few cash crops enabled households to purchase enough imported food in the market. However as terms of trade began to plummet along with cotton and peanut prices beginning in the 1960s, “the Sahelian economy collapsed” and farmers could no longer afford to feed themselves (Psychas & Malaska, 1989, p. 110). Whereas prior to the imposition of a cash crop economy FIELD NOTES | VOL I | 45

Sahelian peasants would store grain from bumper harvests for drought periods (Franke & Chasin, 1980), the onset of a desiccating trend in the late 1960s caught farmers with languishing cash crops, decimated purchasing power and no stores of grain. Such were the conditions that contributed to the infamous famine of 1968 to 1973 and their persistence explains much of the contemporary food insecurity of the central Sahel. The precariousness of dependence on cash crops elucidates why the peanut-exporting departments of Maradi and Zinder, the most prosperous rural departments of Niger, suffered the most severe malnutrition in 2005 (Mousseau et al., 2006). In addition to drastically altering the regional economy, colonial policy disrupted the sociocultural systems that ensured the survival of populations through unfavourable climates. During the precolonial period, central Sahelian societies exhibited a tendency to prevent destitution through grain redistribution and the gift economy, to use slavery as a means of extracting surplus for storage, and among the Tuareg pastoralists, to raid sedentary villages for grain and extortion. These adaptations were interrupted either immediately or gradually by the imposition of French colonial rule. Institutional redistribution, assisted by slave production, trade, and pillage, was a major focus of the Songhai Empire. Portions of the vast wealth amassed through trans-Saharan trade and a thriving system of slave estates were invested in grain reserves (Franke & Chasin, 1980). Widespread slavery was the lifeblood of the estates, enabling them to produce enough grain to contribute substantially to the drought buffer. As Lansiné Kaba (1984) writes, “without this extensive slavery, the labor costs of large estates . . . would be prohibitive” (p. 41). Royal granaries were 46 | LESK

occasionally filled with grain pillaged during annual campaigns of pacification (Franke & Chasin, 1980). The kings of Songhai took redistribution seriously: the agricultural minister, who was usually son of the king, was tasked with “the redistribution of grains to the poor” (p. 57), and royal gardens worked by slaves were established to feed villages with food deficits. “Local famines occurred, but they . . . were alleviated with stores from the royal granaries” (p. 58). The Hausa societies of central Niger during the 19th century exhibited a complex hierarchy of redistributive mechanisms on the levels of lineage, village, and state. After the harvest, peasants would present seed for the next season’s crop and a portion of grain for the lean season to the village chief, who would oversee their storage until after the first rains of the next planting season (Watts, 1984). In the event of a poor harvest for an individual household, interest-free loans of grain were expected from households with surplus, particular from ones within the extended family. In case of village-wide shortages, regional high-yielding farmers were obliged to offer similar loans at harvest festivals, a practice which was regarded as a personal honour under the Islamic principles of charity. Women and youth commonly exchanged comparable redistributive gifts and loans between households and villages. When severe drought struck and regional grain stocks were depleted, the population turned to the ruling classes who contributed to stocking the central granaries through the taxes they paid on their slave estates and transSaharan commerce, and the levy they extracted from peasant’s harvests. Thus, the population in Hausa society was triangularly protected against variability in food supply (Watts, 1984). Tuareg pastoralists employed gift giv-

ing, military might and slavery to protect themselves against famine. Loans and outright gifts of fertile female livestock were given to groups who had lost their herds in order to help them re-establish themselves, and “there were mechanisms for redistributing the livestock so that more people had access to the animal’s milk” (Franke & Chasin, 1980, p. 43). The Tuareg raided farming villages of the southern Sahel, particularly when their herds had been diminished by drought, either to pillage grain directly or to claim ownership over them and extract an annual tribute of grain (Baier, 1976). During raids, slaves were also taken to serve as labour in captured southern villages or to garden and gather wild edibles in Tuareg camps (Franke & Chasi, 1980). Tuareg aggression frequently led to war, the casualties of which may have played a role in population control (Baier, 1976). The imposition of French colonial rule deeply disturbed the social institutions that were central to pre-colonial central Sahelian food security. After the conquest of the central Sahel towards the end of the 19th century, France discouraged slavery in its colonies and by 1947 slaves were no longer used as estate labour in Niger and Mali – the traditional mode of production that had enabled societies to produce and store vast quantities of grain had become politically unfeasible (Raynaut, 1997). Though the Songhai Empire collapsed long before the arrival of the French, the tradition of slavery-based production that its history of drought resistance exemplified had become impracticable. The abolition of slavery was particularly severe among the Tuareg, who were heavily dependent on their southern Sahelian slaves to survive droughts (Franke & Chasin, 1980). Furthermore, the French colonial military’s exertion of order and control prevented raiding as a means of acquir-

ing food (Franke & Chasin, 1980), and may have eliminated warfare as a means of keeping the population on par with the land’s carrying capacity (Baier, 1976). This reality invites the question of whether the relative freedom, equality, peace and tranquility brought about by French conquest merit the loss of these traditional sociocultural adaptations to climatic variability. Under French rule, the Songhai and Hausa models of taxation for redistribution during drought were abandoned and replaced with taxation for the purpose of filling colonial coffers (Watts, 1984). Furthermore, whereas the timing and size of Hausa taxes were subject to adjustment based on the timing and success of the harvest, colonial taxes were inflexible and led to peasants into debt, hunger and generalized vulnerability. The institutionalized Islamic ethics obligating the rich to ensure the wellbeing of the poor, which characterised the Hausa and Songhai states, ceased to exist in the modern secular governments of the colonial and postcolonial periods, and the centralized storage of grain fell from its former efficacy, leaving the populations of Mali, Burkina Faso, and Niger vulnerable to the vicissitudes of Sahelian rainfall (Somerville, 1976, Ch. 3; Lateef, 1980, Ch. 2). Social changes brought about by colonial rule have compromised the gift economy in the central Sahel. Raynaut (1997) identifies the rise of individualistic economic values and the decline of lineage-based and collective economies as consequences of the growth of the market economy in the Sahel. Watts, who attributes these changes to the rise of cash cropping, articulates the implications of these social changes: “the distribution . . . of foodstuffs was no longer based upon the old, socially established norms of giftgiving” (p. 136). He goes on to ascribe the recent correspondence between drought FIELD NOTES | VOL I | 47

and famine to the colonial disruption of the Hausa sociocultural system.


The complex and delicate economic and sociocultural systems that have emerged and flourished in the central Sahel have historically provided its people with the capacity to survive the frequent droughts intrinsic to its climate. The disruption of these systems by invasions and conquests has historically resulted in their failure. French colonial rule in the central Sahel resulted in the demise of the lucrative trans-Saharan trade and the replacement of sound agronomic systems with destructive and unreliable cash crop economy. The colonial disturbance further altered the sociocultural institutions of redistribution, gift-giving, slavery, and raiding which played central roles in the food security of the central Sahelian peoples. The French colonial episode is not the sole instance of a disruptive shock to the sensitive food security systems of the central Sahel, but it is a particularly devastating one whose effects are still being felt today.

REFERENCES All Africa. (2011). Sahel the Danger Zone for Food Insecurity. Baier, S. (1976). Economic History and Development: Drought and the Sahelian Economies of Niger. African Economic History, 1, 1-16. Breunig, P., & Neumann, K. (2002). From Hunters and Gatherers to Food Producers: New Archaeological and Archaebotanical Evidence from the West African Sahel. In Hassan, F. A. (Ed.), Droughts, Food and Culture (123-156). New York, NY: Plenum. Brooks, N. (2006). Climate change, drought, and pastoralism in the Sahel. Discussion note for the World Initiative on Sustainable Pastoralism.

48 | LESK

Butzer, K. (1984). Late Quaternary Environmental Change in the Sahel. Resource Management for Arid and Semiarid Regions. Caldwell, J. C. (1975). The Sahelian Drought and its Demographic Implications. Washington, D.C.: Overseas Liaison Committee. ComitÊ d’Information Sahel. Qui se nourrit de la famine en Afrique? Paris: Maspero, 1975. Dai, A., et al. (2004). The Recent Sahel Drought is Real. International Journal of Climatology, 24, 1323-1331. Doyle, M. (2005). Niger livestock face food crisis. BBC News. Filipovich, J. (2001). Destined to Fail: Forced Settlement at the Office du Niger, 1926-45. The Journal of African History, 42, 239260. Foy, H. (2010). Millions Face Starvation in West Africa. The Guardian. Franke, R. W., & Chasin, B. H. (1980). Seeds of Famine. New York, NY: Universe. Gray, L. C. (2005). What Kind of Intensification? Agricultural Practice, Soil Fertility and Socioeconomic Differentiation in Rural Burkina Faso. The Geographical Journal, 171, 70-82. Hassan, F. A. (2002). Droughts, Food and Culture. New York, NY: Plenum. Kaba, L. Power, prosperity, and social inequality in Songhay. In Scott, E. (Ed.), Life before the Drought (29-46). Boston, MA: Allen & Unwin. Lateef, N. V. (1980). Crisis in the Sahel: A case study in development cooperation. Boulder, CO: Westview. Levtzion, N. (1980). Ancient Ghana and Mali. New York, NY: Africana. Lofchie, M. F. (1975). Political and Economic Origins of African Hunger. The Journal of Modern African Studies, 13, 551-567. Lovejoy, P. E., & Baier, S. (1976). The DesertSide Economy of the Central Sudan. In Michael H. Glantz (Ed.), The Politics of Natural Disaster, 145-168. New York, NY: Praeger.

Marshall, F., & Hildebrand, E. (2002). Cattle Before Crops: The Beginnings of Food Production in Africa. Journal of World Prehistory, 16, 100-143. McCann, J. C. (1999). Climate and Causation in African History. International Journal of African Historical Studies, 32, 261-79. Mousseau, F., et al. (2006). Sahel: Prisoner of Starvation? The Oakland Institute. Munson, P. J. (1976). Archaeological data on the origins of cultivation in the Southwestern Sahara and their implications for West Africa. In Harlan, J. R., de Wet, J. M., & Stemler, A. B. L. (Eds.), Origins of African Plant Domestication (187-210). The Hague: Mouton. Ottaway, D. (1973). Long drought threatens existence of West African nations. The Boston Globe. Psychas, P., and Malaska, P. (1989). An Analysis of the Problematique of African Famine. In Lemma, A., & Malaska, P. (Eds.), Africa Beyond Famine (81-131). London: Tycoon. Rain, D. (1999). Eaters of the Dry Season. Boulder, CO: Westview. Raynaut, C. (1997). Societies and Nature in the Sahel. New York, NY: Routledge. Rice, X. (2010). Severe Drought Causes Hunger for 10 Million in West Africa. The Guardian. Scott, E. (1984). Life before the Drought. Boston, MA: Allen & Unwin. Shanahan, T. M., et al. (2009). Atlantic Forcing of Persistent Drought in West Africa. Science, 324, 377-380. Somerville, C. M. (1986). Drought and Aid in the Sahel. Boulder, CO: Westview. Swift, J. (1977). Sahelian Pastoralists: Underdevelopment, Desertification, and Famine. The Annual Review of Anthropology, 6, 457-478. UN News Center. (2011). Niger’s Food Crisis Affecting More Than Half of Country’s Villages. Vernet, R. (2002). Climate change during the Late Holocene in the Sahara and the

Sahel: Evolution and Consequences on Human Settlement. In Hassan, F. A. (Ed.), Droughts, Food and Culture (47-64). New York, NY: Plenum. Watts, M. J. (1984). The demise of the moral economy: food and famine in a SudanoSahelian region in historical perspective. In Scott, E. (Ed.), Life before the Drought (124-146). Boston, MA: Allen & Unwin. Webb, J. L. A. (1995). Desert Frontier. Madison, WI: University of Wisconsin Press. Williams, M. J. (1974). On U.S. assistance to parched West Africa. The New York Times.


Vulnerability Assessment of Lemurs on the Island of Madagascar


The objective of this study is to examine the projected impacts of climate changes on the island of Madagascar and the expected responses of the lemur community. Madagascar is likely to experience a warming above the global mean, to be affected by more frequent droughts, to be characterized by a shorter but stronger rainy season and to be hit by more intense tropical cyclones. Rising temperatures will cause habitat loss and a shift in the range of species that will require migration. Extreme events such as drought, heavy rainfall and cyclones will have an impact on food availability and on the reproductive capacity of lemurs. This vulnerability assessment demonstrates that, despite the strong adaptive capacity of lemurs to aridity and extreme events, uncertainty remains about their ability to cope with future climate change due to habitat fragmentation and to the increased frequency of destabilizing events. ABSTRACT.



limate change has been the subject of numerable studies in the last decades. Recently, work on the causes of recent climatic variations and possible mitigation options have been complemented by examinations of the necessary adaptation to climate change. Vulnerability assessments are important tools to evaluate the adaptation needs of subjects to future climate change. According to the Intergovernmental Panel on Climate Change (IPCC), vulnerability is: The degree to which a system is susceptible to, and unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate of climate change and variation to which a system is exposed, its sensitivity, and its adaptive capacity” (IPCC, 2007a, p. 883).

for primate conservation since all 36 primate species are endemic to the island— lemurs are not found in the wild outside Madagascar, except for a few introduced species on the Comoro Islands—and in 2000, eleven lemur species were critically endangered (CEPF, 2000). In addition, in the year 2000 less than 10% of the island’s primary vegetation remained (Myers et al., 2000), and clearing has continued since then. Deforestation for agricultural and grazing purpose has led to a fragmentation of the original forest ecosystems on the island and, coupled with strong rainfall and frequent tropical cyclones, has engendered strong erosion rates on the island (Randrianarijaona, 1983). The exposure unit examined in this study is the entire Lemuriformes infraorder through case studies of particular species and populations. The study area is the entire island of Madagascar. This assessment begins with a brief description of

As such, vulnerability is a combination of the strength of the changes experienced by the subject, its propensity to be affected by those changes and its ability to function despite those changes. The objective of this study is to assess the vulnerability of lemurs to future climate change in Madagascar. The lemur clade includes a great variety of species, ranging from tiny mouse lemurs to large Indri lemurs (figure 1). Lemurs are primate species endemic to the island; their safeguard is thus important for preserving future global biodiversity (Myers, Mittermeier, Mittermeier, da Fonseca & Kent, 2000). Madagascar is classified as a biodiversity ‘hotspot’ due to its high level of endemism and species’ endangerment (Myers et al. 2000, p. 854); indeed, Madagascar is among the ‘hottest’ of these ‘hotspots’ (Myers et al., 2000, p. 856). Madagascar is the highest priority

Figure 1. The ring-tailed lemur (Lemur catta) is certainly the most internationally known lemur species FIELD NOTES | VOL I | 51

the methods used and the environmental conditions of the study area. Then, current climate conditions and projected climate changes are examined for Madagascar. The core of this study follows with an impact and vulnerability assessment regarding the different consequences of projected climate changes on lemurs. The study ends with an assessment of the autonomous adjustment of lemurs to a changing climate with emphasis on the uncertainties surrounding their future adaptive capacity.


The divisions of this assessment are adapted from the seven-steps framework for assessing climate change impacts and adaptations developed by the IPCC (1994). Climate change impacts for Madagascar are based on the global emissions and impact scenario produced for the Fourth Assessment of the IPCC (2007b and 2007c). Global and local impact scenarios have been combined. Specific impact projections on lemurs are based on a review of empirical studies and expert judgement from the literature on the impact of current climate variations on lemurs in Madagascar.


Madagascar is an island located in the Indian Ocean, on the east of the African continent. The island is commonly divided into five major ecoregions, roughly corresponding to biomes (see figure 2). The Eastern ecoregion is characterized by lowland rainforest patches and contains the highest species diversity of the island (Critical Ecosystem Partnership Fund [CEPF], 2000). The Northern Mountains ecoregion encompasses two high mountain massifs, and a small transition zone of mixed forest (CEPF, 2000). The Western ecoregion is the largest one, with its dry


deciduous forests ranging from the coastal plain to the central plateau (CEPF, 2000). The Central ecoregion contains the central plateau and some remaining patches of moist and dry forest (CEPF, 2000). Finally, the Southern ecoregion, referred as the spiny desert, is a dry forest composes of deciduous shrubs (CEPF, 2000).


Madagascar contains multiple climatic zones as defined by the Köppen climate classification (figure 3). Those different climates are determined both by local geographical factors and global patterns; thus, local climate patterns to depend on a multiplicity of factors, such as topography, the dynamics of the trade winds, the monsoon flow and the subtropical anticyclones (Nassor & Jury, 1998). Temperature and precipitation vary both seasonally and by location on the island. The mean temperature in the highlands is of about 18°C compared to approximately 26°C on the West-coast lowlands (National Aeronautics and Space Administration [NASA], n.d.). With the exception of the dry South, every location receives most of their precipitation during the austral summer or wet season, occurring from December to March (Nassor & Jury, 1998). The amount of precipitation varies notably because of the monsoon wind influence. The Northwest receives the most precipitation, with an average of 508 mm per month during the wet season, and the Southeast receives the least (Dunham, Erhart & Wright, 2011). The Southwest rarely receives more than 400 mm of precipitation per year, which has recently shifted toward March and April (Bohr, Giertz, Ratovanamona & Ganzhorn, 2011). Madagascar’s climate is also characterised by tropical cyclones during the austral

Figure 2. Map of Madagascar’s biomes FIELD NOTES | VOL I | 53

summer. The passage of those cyclones tends to be irregular: between 1961 and 1990, five tropical cyclones hit Madagascar during the wet summers and zero or few during the dryer summers (Nassor & Jury, 1998). Tropical cyclones are notably associated with sea-surface temperatures (SSTs) of more than 280C, weak easterly wind shear and monsoon surges (Nassor & Jury, 1997). More cyclones are thus generated near Madagascar during El Nino years of the ENSO (Kuleshov et al., 2009).

Figure 3. The climatic zones of Madagascar, based on the Köppen climate classification map. Tropical rainforest climate (Af ), tropical monsoon climate (Am) and humid subtropical climate (Cfa) on the West coast; temperate climate with dry winters (Cwc) and humid subtropical climate (Cwa) on and around the central highlands; tropical wet and dry climate (Aw) on the Northeast coast and in the South; and steppe climate (BSh) and small desert climate (BWh) in the Southeast



According to the Fourth Assessment of the IPCC, all emissions scenarios expect an increase in global mean temperature between 1.10C and 6.40C (IPCC, 2007b). Africa’s median temperature increase is projected to be about 1.5 times higher than the global mean, varying between 3 and 40C for all seasons (IPCC, 2007c). Madagascar lies in the South African sub-region, where the increase will be more intense in September, October and November (IPCC, 2007c), the transition period between the dry and the wet seasons. Climate change models also predict that the Southern African sub-region will experience decrease in precipitation in austral winter and spring, causing the dry season to be dryer and the rainy season to be delayed (IPCC, 2007c). This delay will be coupled with an increase in precipitation in the Southern part of the West African monsoon (IPCC, 2007c). The lower frequency of rainy days will be compensated by an increase in the intensity of rainy events (IPCC, 2007c). As such, “in southern Africa, the frequency of extremely dry austral winters and springs [will] increas[e] to roughly 20%, while the frequency of extremely wet austral summers [will] doubl[e]” (IPCC, 2007c, p. 871). Episodes of drought and of heavy precipitation could thus be expected to increase (IPCC, 2007b). However, the east and the west portion of Madagascar are likely to be affected differently due to their already different conditions (IPCC, 2007c). Increase in tropical sea-surface temperatures (SSTs) will also influence rainfall patterns during the warm season (IPCC, 2007c). Increasing SSTs will be linked with a likely increase in tropical cyclone intensity, in which wind speeds and precipitation will become more extreme

(IPCC, 2007b). Changes in the frequency and location of future tropical cyclones are still unknown, due to the uncertainty in modelling of future El Niño-Southern Oscillation (ENSO) (IPCC, 2007c). Tropical cyclones will hit Madagascar more violently, and their already variable frequency will fluctuate depending on changes in ENSO (IPCC, 2007c).


Due to species’ habitat requirements, temperature increase leads to the displacement of species’ range in the pole direction and in higher altitude (Groupe d’Experts Intergouvernemental sur l’Évolution du Climat [GIEC], 2002). Malcolm, Liu, Neilson, Hansen and Hannah (2006) have estimated that climate change in Madagascar will cause a habitat loss between 11% and 27% if total range migration is possible, and between 17% to 50% if no migration is possible at all. The species living in the already arid south will be more exposed to change, requiring them to migrate (Hannah et al, 2008). This exposure to temperature change is coupled with an obstacle to adaptive capacity: habitat fragmentation in the region makes migration nearly impossible (Bohr et al., 2011). Bohr et al. (2011) studied two populations of gray-brown mouse lemurs (Microcebus griseorufus), one inhabiting the dry spiny brush and one living in the mesic dry forest. The gray-brown mouse lemurs in the spiny brush have lower population densities due to comparably lower fruit production in the area; the mouse lemurs in the arid niche have to include more gum into their diet, leading to a smaller fat accumulation (Bohr et al., 2011). With increased aridity, food availability will further shift toward gum at the

expense of the more caloric fruits (Bohr et al., 2011). To compensate for this reduction in available calories, the lemurs can increase their home range; however, over a certain threshold, the increase in range fails to profitably compensate for the decrease in food availability (Bohr et al., 2011). For that reason, the gray-brown mouse lemur seems “[un]able to extend its home range much beyond its present size” (Bohr et al., 2011, p. 910). The connectivity between the spiny brush and the dry forest has been dramatically reduced, inhibiting migration (Bohr et al., 2011). In sum, habitat fragmentation increases species’ vulnerability by impeding their migration, thus lessening their capacity to adapt to increasing aridity. In fragmented habitats, even adaptive species struggle to compensate for environmental changes. Mountainous areas are particularly sensitive to temperature change (IPCC, 2007b). Elevation creates multiple micro-climates, which means that the climatic component plays a great role in restricting the range of species (Hannah, Midgley and Millar, 2002). Raxworthy et al. (2008) examined climate and species distribution on the Tsaratunana Massif of northern Madagascar in 1993 and 2003. They recorded a warming higher than the global average and observed an upslope displacement of amphibians and reptiles that ranged between 19 and 51 meters, showing a variation in sensitivity between species (Roxworthy et al., 2008). The most vulnerable species in the northern massifs are those that cannot expand their range upslope because they are restricted to 600 meters of the summit. A warming of 1.70C will lead to the extinction of three high-montane endemic species within the next 50 to 100 years (Roxworthy et al., 2008). Montane species more vulnerable to rising temperature are thus the ones living near the summit FIELD NOTES | VOL I | 55

because their adaptive capacity is limited by geomorphological features. Since lemur communities tend to decrease with elevation, they are not highly vulnerable to an upslope shift (Goodman & Rasolonandrasana, 2001). Exposed and thus potentially vulnerable populations however do exist, such as the ring-tailed lemur (Lemur catta) group living near the summit of the Andringitra Massif (Goodman, Rakotoarisoa & Wilmé, 2006). More frequent drought

With rising temperature during the dry season and a delaying of the rainy season, the frequency and the extent of droughts is likely to increase in Madagascar (IPCC, 2007b). Droughts in turn increase the risk of wildfires (IPCC, 2007b). Wildfires frequency is strongly linked to the dryness of the austral winter; in 1987, a particularly dry El Nino year, Madagascar had the greatest number of recorded wildfires in the world (Ingram & Dawson, 2005). Wildfires are part of the island’s ecology, yet an increase in the area affected or in wildfires frequency will impact forested areas, representing a risk to lemurs. An increase in dryer years and in droughts frequency will also affect lemurs’ population size. Dunham, Erhart, Overdorff and Wright (2008) found that fecundity rate for the Milne-Edward’s Sifaka (Propithecus edwardsi) was 65.6% lower during dry ENSO years from 1986 to 2003, compared to other years. This reduced fecundity, attributed to a lower fertility and to an increase in the malnourishment of youth, led to a decrease in the population size (Dunham et al., 2008). An increase in the frequency of ENSO also increases the risk of decline in the Milne-Edward’s Sifaka population (Dunham et al., 2008, p. 293). Gould, Sussman and Sauther (1999) also observed a decline in the population of ring-tailed


lemurs after a drought episode in 1991 and 1992 in the dry forest of Southwestern Madagascar. Two years after the prolonged drought, the lemur population decreased by 31%, caused by an increase in mortality: the infant mortality rate was as high as 80% in 1992 (Gould et al., 1999). Drought reduces food availability, which prevents mothers from producing enough quality milk (Gould et al., 1999). Drought thus has a higher impact on the fecundity of lemurs when it occurs during the lactation period, and is augmented when occurring in dry forest compared to rain forest (Dunham, Erhart & Wright, 2011). Mothers and infants of populations inhabiting the dry forest ecosystem are thus particularly vulnerable to drought (Gould et al., 1999). Heavier rainfall

Species living in the tropical forest of the east coast are more affected by the changing patterns of rainfall. Lemur reproduction is seasonal, so weaning corresponds to the peak production in fruits and young leaves (Dunham et al., 2011). Contrarily to the south of the island, the eastern region receives more rainfall and in a more intense manner during the warm phase of the ENSO (Dunham et al., 2011). During those warm phases, the fecundity of the Milne-Edward’s Sifaka substantially decreases (Dunham et al., 2011). When the heavy rainfall episode occurs before conception, it affects the ovulation of females; when it occurs during the weaning period, it leads to a decrease in infant survival (Dunham et al., 2011). A significant increase in precipitation adds a stress on trees and contributes to tree mortality (Dunham et al., 2011). More intense rainfalls thus directly impact the reproduction of the Milne-Edward’s Sifaka by affecting ovulation, but also indirectly by influencing food availability

for recently weaned lemurs. With climate change, heavy rainfall is predicted to become the norm; the abundance of lemurs may thus be expected to decrease in the absence of adaptation in their reproductive behavior (Dunham et al., 2008). This case study of the Milne-Edward’s Sifaka thus supports the claim that Malagasy primates are affected by change in precipitation and may be vulnerable to long-term changes in the precipitation cycle brought by climate change. More intense tropical cyclones

The short-term impact of cyclones on mammal species is considerable (Dunham et al., 2011). Among the population of Milne-Edward’s Sifaka studied by Dunham et al. (2011), cyclones occurrence was the most important factor determining the reproductive capacity of lemurs. A strong negative relationship exists between the passage of tropical cyclones and the birth rate of female Milne-Edward’s Sifaka (Dunham et al., 2011). This association is mainly attributable to the timing of the cyclone season, which happens from December to April and corresponds to the gestation period of the lemurs (Dunham et al., 2011). Since cyclones engender strong precipitation events, they increase the nutritional stress on the wildlife by having an impact on tree productivity and mortality, which then increases mortality rates among lemurs (Dunham et al., 2011). Nevertheless, the lemur population has been able to cope with this disturbance, proven by the continued presence of lemur species in cycloneprone environments. As previously stated, uncertainty still surrounds the frequency of tropical cyclone in a context of climate change, but cyclone intensity is likely to increase (IPCC, 2007b). This increased intensity is likely to cause more uprooted trees, further reducing food availabil-

ity and putting more pressure on lemurs (IPCC, 2007b). According to Johnson et al. (2011), the resilience and resistance of lemurs to tropical cyclones strongly depend on their feeding requirement. The authors studied a population of gray-headed lemurs (Eulemur cinereiceps) at Manombo, in the Southwest of Madagascar, before and ten years after the Cyclone Gretelle of 1997. Cyclone Gretelle was detrimental to the lemurs’ habitat, uprooting or severely damaging 62.3% of the trees at Manombo (Johnson et al., 2011). Ten years later, the forest structure was still significantly different from the pre-cyclone period; nonetheless, the population density of the gray-headed lemurs ten years after the cyclone was similar to the pre-cyclone period (Johnson et al., 2011). This finding was surprising, since another species of lemur in Manombo suffered a significant decline in its population due to Cyclone Gretelle and was still rare ten years after the event (Johnson et al., 2011). Johnson et al. (2011) attributed the resilience of the gray-headed lemur to its dietary flexibility. Being frugivores, the gray-headed lemurs should have faced a high risk of starvation in the aftermath of the cyclone; however, the inclusion of endemic flowers and exotic plants in their diet has permitted them to maintain a stable population (Johnson et al., 2011). Gray-headed lemurs were thus able to adapt to the changes in habitat induced by the landfall of Cyclone Gretelle. As such, variations exist in the extent of the short- and long-term vulnerability to cyclones within the lemur community. Since cyclones already have a substantial impact on lemurs, particularly frugivore species, uncertainty remains as to the resilience of some lemur species to more intense cyclones.



Lemurs are highly adaptive species, categorized as “edge” or “weed” species (Goodman et al., 2006, p. 12). Historically, they have been able to cope with climatic and environmental variations (Lehman, Rajaonson & Day, 2006). Lemurs are tolerant to edge effect; they are not restricted to the core of the forests and could be relatively unaffected by diminution in habitat size due to climate change (Lehman et al., 2006). However, Lehman et al. (2006) suggest that there may be a “threshold of habitat disturbance” due to the edge-intolerance of the food trees on which lemurs depend, a threshold that should be further investigated in order to avoid lemurs’ extinction (p. 239). In other words, lemurs’ adaptive capacity may be compromised by the lower adaptive capacity of the trees on which they depend. Another adaptive strategy to changing climatic condition is migration. Mobile, lemurs are able to use transformed landscapes such as Eucalyptus plantations to find food, to travel and to rest (Ganzhorn, 1987). However, lemurs can only survive in old plantations and their density is lower than in rain forests (Ganzhorn, 1987). In addition, the current habitat fragmentation on the island limits migration opportunity for some lemur species (Bohr et al. 2011). As for the rare high-montane lemur population, migration is not an option, but adaptation is likely; the high adaptive capacity of the ring-tailed lemur may balance this physical restriction on its range (Goodman et al., 2006). Lemurs are able to cope with extreme events by changing their reproductive and dietary behaviors. For instance, the ring-tailed lemur population affected by a two-year drought had started to recover as soon as four years after the event (Gould


et al., 1999). This was made possible by an increase in birth rate, early sexual maturity and dietary adaptation (Gould et al., 1999). The gray-headed lemur community of Manombo was able to survive to a strong tropical cyclone by including flowers into its diet (Johnson et al., 2011). However, adaptability is not evenly distributed among lemurs, with infants and females of frugivore species being the most vulnerable to disturbance (Gould et al., 1999; Godfrey & Irwin, 2007). In addition, extreme events caused by climate change may increase lemurs’ vulnerability to other external factors: If a natural disaster such as a drought or cyclone markedly affects availability of food resources, malnourished animals may not be able to withstand disease and injuries, or to protect themselves from predators, as well as animals living in a less stressful environment (Gould et al., 1999, p. 77). As such, even if many lemur species have shown their ability to cope with the effects of climatic disturbances, repeated exposure to extreme events may increase their sensitivity to other elements of their environment.


Madagascar is likely to be greatly affected by climate changes, which will impact the lemur community. Lemurs will thus be exposed to climate variations, and their sensitivity has been demonstrated by their responses to current variations. Lemurs are known for their great adaptive capacity; however, their ability to cope with increasing aridity is likely to be compromised by the current habitat fragmentation of Madagascar. In addition, their physiological and behavioral capacity to adapt to drought, heavy rainfall episodes and cyclones could be compromised by the increased frequency and intensity of

such extreme events. Lemurs should be carefully monitored in order to assess their future ability to adapt in order to avoid the loss of such unique species. Strategies aimed at increasing the adaptive capacity of lemurs should thus try to re-establish or protect existing connections between habitats and to limit other stresses affecting lemurs during extreme events.

REFERENCES Bohr, Y.E.B., Giertz, P., Ratovonamana, Y.R., & Ganzhorn, J.U. (2011). Gray-brown Mouse Lemurs (Microcebus griseorufus) as an Example of Distributional Constraints through Increasing Desertification. Int J Primatol, 32, 901–913. Critical Ecosystem Partnership Fund (2000). Madagascar ecosystem of the Madagascar & Indian ocean islands biodiversity hotpsot. 22 pp. Retrieved from http://www.cepf.net Dunham, A.E., Erhart E.M., Overdorff, D.J., & Wright, P.C. (2008). Evaluating effects of deforestation, hunting, and El Nino events on a threatened lemur. Biological conservation, 141, 287-297. Dunham, A.E., Erhart, E.M., & Wright, P.C. (2011). Global climate cycles and cyclones: consequences for rainfall patterns and lemur reproduction in southeastern Madagascar. Global Change Biology, 17, 219-227. ESRI (2011). World Countries [Data file]. Retrieved from ArcGIS Online Ganzhorn, J.U. (1987). A Possible Role of Plantations for Primate Conservation in Madagascar. American Journal of Primatology, 12205-12215. Groupe d’Experts Intergouvernemental sur l’Évolution du Climat (2002). Les changements climatiques et la biodiversité. Document technique V du GIEC. Montreal : CBD/UNEP/WMO. Godfrey, L.R. & Irwin M.T. (2007). The Evolution of Extinction Risk: Past and Present Anthropogenic Impacts on the Primate Communities of Madagascar. Folia Prima-

tol, 78, 405-419. Goodman, S.M. & Rasolonandrasana, B. P. N. (2001). Elevational zonation of birds, insectivores, rodents and primates on the slopes of the Andringitra Massif, Madagascar. Journal of Natural History, 35, 285-305. Goodman, S.M., Rakotoarisoa, S.V., & Wilmé, L. (2006). The Distribution and Biogeography of the Ringtailed Lemur (Lemur catta) in Madagascar. Developments in Primatology: Progress and Prospects, Part I, 3-15. Gould, L., Sussman, R.W., & Sauther, M.L. (1999). Natural Disasters and Primate Populations: The Effects of a 2-Year Drought on a Naturally Occurring Population of RingTailed Lemurs (Lemur catta) in Southwestern Madagascar. International Journal of Primatology, 20(1), 69-84. Hannah, L., Midgley, G.F., & Millar, D. (2002). Climate change-integrated conservation strategies. Global Ecology & Biogeography, 11, 485–495 Hannah, L., Dave, R., Lowry II, P.P, Andelman, S., Andrianarisata, M., Andriamaro, L., Cameron, A. Hijmans, R., Kremen, C., MacKinnon, J. Randrianasolo, H.H., Andriambololonera, S., Razafimpahanana, A., Randriamahazo, H., Randrianarisoa, J., Razafinjatovo, P., Raxworthy, C., Schatz, G.E., Tadross, M., & Wilmé, L. (2008). Climate change adaptation for conservation in Madagascar. Biological Letters, 2, 590594. Ingram, J.C., & Dawson, T.P. (2005). Climate Change Impacts and Vegetation Response on the Island of Madagascar. Philosophical Transactions: Mathematical, Physical and Engineering Sciences, 363(1826), 55-59. Intergovernmental Panel on Climate Change. (1994). IPCC technical guidelines for assessing climate change impacts and adaptations. Part of the IPCC special report to the first session of the conference of the parties to the UN framework convention on climate change. London: WMO/UNEP.


Intergovernmental Panel on Climate Change. (2007a). Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth assessment of the IPCC. Cambridge: Cambridge University Press. Intergovernmental Panel on Climate Change. (2007b). Climate Change 2007: Synthesis Report. Contribution of Working Group I, II and III to the Fourth assessment of the IPCC Cambridge: Cambridge University Press. Intergovernmental Panel on Climate Change. (2007c). Climate Change 2007: the scientific basis. Contribution of Working Group I to the Fourth assessment of the IPCC. Cambridge: Cambridge University Press. Johnson, S.E. Ingraldi, C., Ralainasolo, F.B., Andriamaharoa, H.E., Ludovic, R., Birkinshaw, C.R., Wright, P.C., & Ratsimbazafy, J.H. (2011). Gray-headed Lemur (Eulemur cinereiceps) Abundance and Forest Structure Dynamics at Manombo, Madagascar. Biotropica, 43(3), 371–379. Kuleshov, Y., Chan Ming, F. Qi, L., Chouaibou, I., Hoareau, C., & Roux, F. (2009). Tropical cyclone genesis in the Southern Hemisphere and its relationship with the ENSO. Annales Geophysicae, 27, 2523– 2538. Lehman, S.M., Rajaonson, A. & Day, S. (2006). Edge Effects and Their Influence on Lemur Density and Distribution in Southeast Madagascar. American journal of physical anthropology, 129, 232-241. Malcolm, J.R., Liu, C., Neilson, R.P. Hansen, L., & Hannah, L. (2006). Global Warming and Extinctions of Endemic Species from Biodiversity Hotspots. Conservation Biology, 20(2), 538–548. Mittermeier, R.A, Ganzhorn J.U., Konstant W.R, Glander, K., Tattersall, I, Groves, C.P., Rylands A.B., Hapke, A., Ratsimbazafy, J., Mayor, M.I., Louis Jr. E.E., Rumpler, Y., Schwitzer, C. & R. M. Rasoloarison, R.M. (2008). Lemur Diversity in Madagascar. Int


J Primatol, 29, 1607–1656. Myers, M., Mittermeier, R.A., Mittermeier, C.G., da Fonseca G.A.B, & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403, 853-858. National Aeronautics and Space Administration. (n.d.). GISS Surface Temperature Analysis. Retrieved from http://data.giss. nasa.gov/gistemp/ Nassor, A., & Jury, M.R. (1997). Intra-Seasonal Climate Variability of Madagascar. Part 2: Evolution of Flood Events. Meteorologic and Atmospheric Physics, 64, 243-254. Nassor, A., & Jury, M.R. (1998). Intra-Seasonal Climate Variability of Madagascar. Part 1: Mean Summer Conditions. Meteorologic and Atmospheric Physics, 65, 31-41. Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., Powell, G. V. N., Underwood, E. C., D’Amico, J. A., Itoua, I., Strand, H. E., Morrison, J. C., Loucks, C. J., Allnutt, T. F., Ricketts, T. H., Kura, Y., Lamoreux, J. F., Wettengel, W. W., Hedao, P., & Kassem, K. R. (2001). Terrestrial ecoregions of the world: a new map of life on Earth. Bioscience, 51(11), 933-938. Randrianarijaona, P. (1983). The Erosion of Madagascar. Ambio, 12,(6), 308-311. Raxworthy, C.J., Pearson, R.G, Rabibisoa, N., Rakotondrazafy, A.M. Ramanamanjato, J., Raselimanana, A.P., Wu, S., Nussbaum, R.A., & Stone, D.A. (2008). Extinction vulnerability of tropical montane endemism from warming and upslope displacement: a preliminary appraisal for the highest massif in Madagascar. Global Change Biology, 14, 1703–1720.

The Sea-Ice Carbon Pump: The Carbon Biogeochemistry of Sea Ice and its Role in the Global Carbon Cycle


Ice covered oceans have previously been ignored in global carbon models because it was believed sea ice impeded gas exchange between the ocean and atmosphere. Research in the last decade is showing this is not the case. This paper will review the processes of the sea-ice carbon pump and provide recent estimates of annual carbon fluxes. When ice forms, impurities (including dissolved carbon dioxide) are rejected from the ice matrix and become concentrated in a hypersaline solution within pore spaces of the ice column. The solution drains into the ocean and sinks into intermediate and deep waters. The precipitation of a calcium carbonate mineral and biological activity within the sea ice further increases the amount of carbon that enters the ocean. During the summer, the melting sea ice and photosynthetic activity of organisms further enhances air-sea carbon dioxide exchange. A recent model has estimated the sea-ice carbon pump makes up 17-42% of the total net carbon uptake in high latitude oceans.


62 | WARD


ceans worldwide are currently a sink for atmospheric carbon dioxide (CO2) and annually absorb 30% of anthropogenic emissions (Else et al., 2010). The process and quantification of CO2 uptake rates have been well studied globally, with the exception of remote areas, such as ice covered high latitude oceans. Carbon fluxes in these regions have been ignored because it was thought that sea ice cover impeded gas exchange between the atmosphere and the ocean (Miller et al., 2011). Recent studies within the last decade, however, are indicating that this is not the case; sea ice plays an active role in the uptake of CO2 and has been termed the sea-ice carbon pump. This realization of the role of this carbon pump is important to accurately build climate models to predict future climate change. As temperatures are expected to increase in high latitude regions, sea ice extent is expected to decrease. Knowing the processes and rates of the sea-ice carbon pump will enable scientists to monitor these carbon flux changes and apply it to a global scale. The following paper is a synthesis review of the processes and mechanisms of what is currently known regarding the seaice carbon pump. First abiotic processes, particularly the formation and decay of sea ice, will be covered. Next, the impacts of the precipitation and dissolution of ikaite, a calcium carbonate polymorph mineral will be reviewed; followed by the role of biological communities within sea ice. The final section will deal with the most recent estimates on the annual uptake of CO2 via the sea-ice carbon pump. This paper will conclude with a summary of areas of future research.


The processes controlling the air-sea gas exchange are largely dependent on physical controls of sea ice, most predominantly porosity (a function of temperature and salinity; Rysgaard et al., 2011). In autumn, as air temperatures decrease, sea ice begins to form as a porous matrix. Impurities, made up of dissolved salts, dissolved gases (including CO2) and solutes are unable to be incorporated into the crystalline structure of ice and are therefore rejected. Approximately 10-40% of impurities remain within liquid intrusions and gas bubbles within the solid ice ma-

Figure 1. Brine pockets resembling elongated tubes within sea ice. These ‘brine tubes’ are highlighted by sunlight. The image width is 5mm (Maksym, 2012). FIELD NOTES | VOL I | 63

trix (Petrich and Eicken, 2010). This results in a CO2-supersaturated brine (salinity > 35ppm) that accumulates within the pores of the ice matrix and is then drained into the ocean via gravity (Rysgaard et al., 2011). The higher density brine containing dissolved CO2 sinks down the water column and is incorporated into the in-

termediate and deep waters of the ocean. Some CO2 fluxes into the atmosphere from a small fraction of the CO2-supersaturated brine that is transported upwards as a consequence of the contraction of the ice and brine volume on a rapid cooling surface. As temperatures continue to decrease

Figure 2. Summary of the seasonal carbon fluxes within the sea-ice carbon pump. The figure consists of the upper 100m of the water column. In autumn, as ice forms, salt and impurities are rejected from the ice forming front and accumulate within pore spaces of the ice. The brine, containing dissolved carbon, drains from the ice (by gravity drainage) into the water column and sinks to the intermediate and deep waters. Ikaite precipitates within the ice column increasing the carbon uptake in the winter; it is then released into the ocean via gravity drainage of the brine or when the ice melts in spring. When the ice melts, less dense and less saline water is released causing the water column to become stratified. The less saline water at the surface increases the solubility of CO2 and the uptake of CO2 into the ocean is enhanced. Sea ice algae uptakes CO2 via photosynthesis during the spring and summer and by producing exopolymeric substances during the winter that enters the water during the spring melt, (Rysgaard et al., 2011). 64 | WARD

during autumn and winter, the porosity and connectivity of pores spaces decrease as well, further increasing the concentrations of solutes within the brine solution (Rysgaard et al., 2011). The brine drainage during ice formation causes the waters underneath the ice to be supersaturated with CO2; however the ineffective porosity of the ice column prevents the vast majority of CO2 to efflux into the atmosphere. In spring, increasing temperatures cause the sea ice to melt. The melt water—low in salinity and therefore less dense then the seawater—causes a stratification of the water column (Rysgaard et al., 2011). This lesser saline water further increases the solubility of CO2 and enhances the air-sea gas exchange. This process of ice formation and decay that runs the sea-ice carbon pump is largely a function of first year sea ice. The Arctic Ocean has a maximum sea ice extent of 15 x 106 km2 in the winter and a

minimum of 7 x 106 km2 in the summer (NSIDC, 2012). The Antarctic has a winter maximum of 18 x 106 km2 and a summer minimum of 3 x 106 km2. Based on the winter maximums measured, together both oceans contribute 23 x 106 km2 of sea ice actively engaging in the sea-ice carbon pump process.

Figure 3. Maximum and minimum sea ice extent for the Arctic and Antarctic Oceans (NSIDC, 2012)

Figure 4. Ikaite crystals (Meiners et al., 2012).


It had been widely theorized that calcium carbonate (CaCO3) could precipitate within sea ice, further enhancing the uptake of CO2 into the ocean. The mineral had been successfully precipitated in laboratory sea ice but had never been observed in nature. This changed in 2008 in Antarctica and in 2010 in the Arctic when ikaite (CaCO3 *6H2O), a mineral polymorph of calcium carbonate was observed for the first time in sea ice (Dieckmann et al., 2008; 2010). The exact mechanisms that control the precipitation of ikaite are currently poorly understood, and warrant more research. What is currently known regarding the precipitation of ikaite is that sufficient quantities of calcium and bicarbo-


nate ions are concentrated in the brine solution within sea ice for precipitation to occur (Geilfus et al., 2012). This occurs during the winter when porosity is at its lowest. It also requires ice to precipitate since it is incorporated into its crystalline structure. Ikaite appears to be insensitive to dissolved species such as ions, and dissolved inorganic phosphorus that are known to inhibit calcium carbonate polymorph like calcite in ionic media (Thomas et al., 2010). The following is the chemical reaction for ikaite, and the equations describing dissolved inorganic carbon (DIC) and total alkalinity (TA): Ca2+(aq) + 2HCO3-(aq) + 5H2O(l) -> CaCO3*6H2O(s) + CO2(aq) DIC= [HCO3-] + [CO32-] + [CO2] TA= [HCO3-] + 2[CO32-] + [B(OH)-] + [OH-] – [H-] Every mole of ikaite that precipitates within the ice matrix decreases DIC by one mole, TA by two moles, and further increases the concentration of dissolved CO2 within solution (Geilfus et al., 2012). The precipitation of ikaite varies spatially within the ice column; if it precipitates near the bottom (and most porous) segment of the ice column the resulting mineral form of CO2 may enter the underlying water through gravity drainage (Rysgaard et al., 2011). Dissolution does the opposite and is thought to be the source of excess TA observed in the surface waters during the spring ice melt.


Despite the extreme environment of

66 | WARD

sea ice (made up of low temperatures and high salinity) a range of organisms have exploited this niche. This biological community, made up of ice algae, bacteria, viruses and protists, are responsible for cycling carbon and nitrogen between its organic and inorganic reservoirs (Stedmon et al., 2011). This organic carbon is the main source of carbon for pelagic consumers during the winter ice coverage and spring ice melt (Riedel et al., 2008). Thus sea ice communities play an important role in the food web of these ecosystems. During the summer, algae are released from the sea ice and sink down, triggering a response in benthic activity. Benthic communities live on the sea floor of ocean margins. These communities in the Arctic Ocean are some of the most active in the world; however, they are only active during ice-free periods, limiting their annual carbon intake in comparison to other ocean margin ecosystems (Renaud et al., 2007). Ice algae, bacteria, viruses and protists live within the brine channels located inside the sea ice matrix. More specifically, organisms tend to be located in the bottom layers of the ice column, where temperatures are slightly lower (compared to the atmosphere) and where exchange with the underlying seawater is possible

Figure 5. Ice core with algae band (Krembs & Deming, 2011)

(Krembs et al., 2002). During the winter, when irradiance is low from lack of sunlight or snow-cover, photosynthesis does not occur and instead biological assemblages tend towards being net heterotrophic (Riedel et al., 2008). During the spring and summer, however, increased daylight enables photosynthesis to occur, leading to extensive algae blooms (largely driven by diatoms) within the sea ice and phytoplankton within the ocean water. Respiration by heterotrophs does occur, however, the rate of carbon influx by photosynthesis renders the carbon outflux rates minor in comparison. In the Arctic Ocean, primary production within sea ice is estimated to account for less than 25% of total primary production (Riedel et al., 2008). Sea ice algae are responsible for producing exopolymeric substances (EPS), a diverse group of polyssacharides and glycoproteins with gel-like characteristics that are high in carbon. During the sea algae bloom period, EPS are produced in high concentrations and make up 72% of the total particulate organic carbon produced during this time (Riedel et al., 2008). Kremb et al. (2011) demonstrate that sea ice algae produce EPS to improve the habitability, and survivability of sea ice as well as increasing the potential for primary production, and thus the potential for enhancing the uptake of CO2. By producing sufficient quantities of EPS, sea ice algae are able to alter the microstructure and desalinate growing sea ice giving an overall effect to reduce ice permeability and to retain more salt. EPS also increases the viscosity of the bulk fluid, and thus decreases the diffusion and advection of ions through the ice column, aiding to trap nutrients for the algae. Finally, EPS was found to act as antifreeze, suppressing the growth of ice crystals offering cryoprotection. This is

important in preventing the growth of ice crystals into the bodies of organisms during the winter and also helps increase the immediate temperature surrounding the algae. EPS also benefits other organisms; it is thought to be an important component supporting the food web within sea ice. Bacteria and viruses use it as a substrate for colonization, and it is also used for nutrient regeneration for heterotrophic communities (Riedel et al., 2008). The nature of EPS being rich in carbon also provides a direct method of inputting carbon into the ocean when it is released from melting ice during the summer.


Because the study of the sea-ice carbon pump is a relatively new field of research, little has been done to quantify its impact on the global carbon cycle. The majority of studies have so far only extrapolated their results for annual carbon uptake rates, but only for the specific study areas. The following are results from a few studies. Sejr et al. (2011) calculated a 32 g C m-2 yr-1 uptake for Young Sound, located on the coast of northeast Greenland. They took measurements during the summer and winter months and extrapolated results for an annual uptake estimate. Else et al. (2010) observed direct in situ enhanced gas exchange during sea ice formation within the Amundsen Gulf, in Northwest Canada. Using eddy covariance calculations for CO2 fluxes from November to January, they observed a mean uptake of 4.88 µmol m-2 s-1 and a max uptake of 27.95 µmol m-2 s-1of CO2 into the ocean of a polynya with dynamic winter sea ice coverage. They also observed an outgassing event with a mean flux of 2.10 µmol m-2 s-1of CO2 into the atmosphere. Converting these mean flux values into annual FIELD NOTES | VOL I | 67

carbon flux rates (assuming the means are constant) results in a 1.83 x 103 Tg C m-2 yr-1 flux into the ocean and a 17 Tg C m-2 yr-1 flux into the atmosphere. They hypothesize this enhanced gas exchange can be attributed to high water-side turbulence as well as physical and chemical modifications of the surface water by cooling and brine rejection may play a role. Rysgaard et al. (2007) made vertical field measurements of dissolved inorganic carbon, total alkalinity and salinity, in Franklin Bay, Canada; and Young Sound, Greenland and inputted these values into a model to calculate the CO2 flux. Calculating a 10 Tg C yr-1 uptake for Franklin Bay (they did not include estimates for Young Sound) and extrapolating results globally, they estimated the Arctic Ocean uptakes 50 Tg C yr-1 (assuming an ice cover of 7 x 106 km2). Comparing this value to the current estimate of global ocean uptake (2300 Tg C yr-1) shows this estimate accounts for roughly 2% of global ocean carbon uptake (Doney, 2010). Results obtained from direct measurement are extremely varied. This can be explained within the variability of sea ice itself. First the extent of the growth of sea ice (both vertical thickness and the horizontal cover) is dependent on temperature. If temperatures vary slightly from year to year, this will be reflected within the extent of the formation of sea ice. Temperature also affects the porosity of the sea ice—the greater the porosity, the greater the exchange of CO2 between the ocean and atmosphere. Secondly, the amount of brine solution that is trapped within the ice matrix changes every time ice forms. This then impacts the variability of the amount of ikaite that precipitates and can impact the extent of biological communities within the ice. Third, the difficulty of measuring the sea-ice carbon pump alone can vary the methods used and the mag68 | WARD

nitude of sources of error. Results from studies can differ from many orders of magnitude, particularly because errors become amplified when extrapolating results from a few study sites to a global estimate. This is quite common when measuring biogeochemical cycles on a global scale. Rysgaard et al. (2011) is the only study found that attempted to quantify the sea ice carbon pump not at the global scale but at an ocean scale (including ice covered and non-covered areas) for both the Arctic and Antarctic Oceans. They used a simple model to derive results which can be seen in Table 1, using specific sea ice and water column parameters, as indicated within the table that were determined based on average calculations (e.g. average ice thickness). The model calculated a gross transport of dissolved inorganic carbon (TCO2 in the table) of 324 Tg C yr-1 for both the Arctic and Antarctic Oceans. Factoring in ikaite (CaCO3) formation within ice resulted in a 50 Tg C yr-1 increase in the amount of carbon fluxing into the ocean, indicating that ikaite plays a large role in inputting carbon in the ocean. Comparing areas of the oceans that are ice covered to areas that are not found that the sea-ice carbon pump makes up 17-42% of the total net uptake of carbon into high latitude oceans. Although these are not overall global estimates, they do indicate that sea ice does not impede airsea gas exchange and should be included into global models. A rough comparison of the estimates from the model to global ocean carbon uptake rates (2300 Tg C yr-1) demonstrated that sea ice in high latitude oceans accounts for 4% of the global ocean carbon influx (Doney, 2010). This model, however, has limitations as pointed out by the authors. The model does not account for seasonal changes in the mixed layer depth of the water column, nor does it account

Table 1. Quantification of CO2 fluxes in ice covered oceans (Rysgaard et al., 2011). FIELD NOTES | VOL I | 69

for heat content or changing concentrations of salinity and dissolved inorganic carbon. Estimates of ice extent are based on overall changes and not total sea ice production, therefore underestimating ice extent because some areas, such as polynyas form ice continuously from autumn to late spring. Despite these limitations, the authors express that the model quantifies the expected perturbations of the air-sea CO2 flux resulting from the sea-ice carbon pump.


This paper reviewed the processes behind the sea-ice carbon pump, a mechanism that inputs carbon into the ocean that is unique to high latitude oceans. Although biological communities play a role, the sea-ice pump is driven by the physical and chemical dynamics of sea-ice brine (both brine drainage and the precipitation/dissolution of ikaite). During autumn, as sea ice forms, impurities get rejected from the ice forming front, and accumulate in pore spaces within the ice column. Within the brine is a concentration of dissolved CO2 that enters the ocean via gravity drainage and sinks into the intermediate and deep ocean waters. During the winter, ikaite precipitates within the brine solution, further uptaking CO2. When the sea ice melts, the water column becomes stratified with the less dense, less saline water from the melt staying at the surface, the lower salinity increases the solubility of CO2, and further enhancing the CO2 flux into the ocean. Sea-ice algae uptakes CO2 via photosynthesis during the spring and summer and by producing EPS during the winter that enter the water during the spring ice melt. Although global estimates of the sea-ice carbon pump are varied and limited, they show that sea ice does not simply impede the exchange of gases between the ocean 70 | WARD

and atmosphere but instead plays an active role that is complex and should be included in global carbon models. The field of research concerning the sea ice carbon pump is still relatively new with the bulk of the research being carried out within the last decade. There are still many unknowns that need to be studied before an accurate estimation of the sea ice carbon pump can be included in global models. First, carbon flux estimates need to be completed on all types of sea ice (e.g. compact ice, open and closed pack ice, and consolidated ice) because all ice types have different thickness and horizontal extents, thus changing the amount of CO2 in flux between the atmosphere and ocean (Rysgaard et al., 2011). Secondly, the model by Rysgaard et al. (2011; see table 1) demonstrates that ikaite plays a larger role in the uptake of carbon into the ocean than previously thought. Having a better understanding of the mechanisms of the precipitation and dissolution of ikaite minerals within sea ice will provide better estimates of its role within global carbon models. Climate change will greatly impact the magnitude of the carbon flux into the ocean by the sea-ice carbon pump, particularly in changing the annual extent of sea ice. Increasing climate temperatures are expected to decrease sea ice extent, thus decreasing the magnitude of carbon taken up by the sea ice carbon pump. This will be largely felt in the Antarctic, where the majority of sea ice is first year ice (NSIDC, 2012). The effects in the Arctic will not be felt right away since a greater share of ice is multiyear-ice in comparison to the Antarctic. This ice will become first-year ice so the magnitude of the seaice carbon pump will be maintained, but only until a threshold is reached and sea ice extent becomes reduced. Quantifying the differences of carbon uptake between

the sea-ice pump and the biological pump (primary production during the summer) and how they interact together will be important to understand in the context of climate change (Rysgaard et al., 2011). It is expected that the sea-ice carbon pump will play less of a role in the future and thus the biological pump will play a more important role, which will be significant in accurately quantifying the magnitude of carbon uptake within global models.

REFERENCES Comision, J.C. (2010). Variability and trends of the global sea ice cover. In: Sea Ice. 2nd Edition. (eds Thomas, D.N. and Dieckmann, G.S.) Wiley-Blackwell, Oxford, UK, pp. 205-246. Comisio, J.C. and F. Nishio (2008). Trends in the sea ice cover using enhanced and compatible AMSR-E, SSM/I, and SMMR data. Journal of Geophysical Research, 113(2). Doney, S. C (2010). The growing human footprint on coastal and open-ocean biogeochemistry. Science 328(1), 1512-1518. Dieckmann, G.S., G. Nehrke, S. Papadimitriou, J. Gottlicher, R. Steininger, H. Kennedy, D. Wolf-Gladrow, & D.N., Thomas. (2008). Calcium carbonate as ikaite crystals in Antartic sea ice. Geophysical Reasearch Letters, 35(1). Dieckmann, G.S., G. Nehrke, C., Uhlig, J. Gottlicher. S. Gerland, M.A. Granskog, & D.N. Thomas. (2010). Brief communication: Ikaite (CaCO3*6H2O) discovered in Arctic sea ice. The Cryosphere, 4(1), 227330. Else, B.G., T.N. Papakyriakou, R.J. Galley, W.M. Drennan, L.A. Miller & H. Thomas. (2010). Wintertime CO2 fluxes in an Arctic polynya using eddy covariance: evidence for enhanced air-sea gas transfer during ice formation. Journal of Geophysical Research, 116(1). Haubjerg Sogaard D., P.J. Hansen, S. Rysgaard, & R.N. Glud. (2011). Growth limitation of

three Arctic sea ice algal species: effects of salinity, pH, and inorganic carbon availability. Polar Biology, 34(1), 1157-1165. Geilfus, N-X., G. Carnat, T. Papakyriakou, J-L. Tison, B. Else, H. Thomas, E. Shadwick, & D. Delille. (2012). Dynamics of pCO2 and related air-ice CO2 fluxes in the Arctic coastal zone (Amundsen Gulf, Beaufort Sea). Journal of Geophysical Research, 117(1). Krembs, C., H. Eicken & J.W. Deming. (2011). Exopolymer alteration of physical properties of sea ice and implications for ice habitability and biogeochemistry in a warmer Arctic. PNAS, 108(9), 3653-3658. Krembs, C., H. Eocken, K. Junge, & J.W Deming. (2002). High concentrations of exopolymeric substances in Arctic winter sea ice: implications for the polar ocean carbon cycle and cryprotection of diatoms. DeepSea Research, 49(1), 2163-2181. Kwok, R., G.F. Cunningham, M. Wensnahan, I. Rigor, & H.J. Zwally. (2009). Thinning and volumne loss of the Arctic Ocean sea ice cover: 2003-2008. Journal of Geophysical Research, 114(1). Krembs, C. & J.Deming. (2011). Sea ice: a refuge for life in polar seas? NOAA Arctic Theme Page. Accessed at: http://www.arctic. noaa.gov/essay_krembsdeming.html. Maksym, T. (2012). Salinity and brine. National Snow & Ice Data Center. Accesses at: http://nsidc.org/cryosphere/seaice/characteristics/brine_salinity.html. McLaren, A.J., H.T. Banks, C.F. Durman, J.M. Gregory & T.C. Johns. (2006). Evaluation of the sea ice simulation in a new coupled atmospheric-ocean climate model (HadGEM1). Journal of Geophysical Research, 111(1). Merners, K., G. Dieckmann, E. Damm and D. Nomura (2012). Sea ice biogeochemistry. SIPEX2. Accessed at: http://seaice.acecrc. org.au/sipex2012/science-projects/sea-iceecology/sea-ice-biogeochemistry/. Miller, L.A., T.N. Papakyriakou, R.E. Collins,


J.W. Deming, J.K. Ehn, R.W., Macdonald, A. Mucci, O. Owens, M. Raudsepp & N, Sytherland. (2009). Carbon dynamics in sea ice: a winter flux time series. Journal of Geophysical Research, 116(1). National Snow and Data Center (NSIDC, 2012) All about ice: Arctic vs Antarctic. University of Colorado Boulder. Accessed at: http://nsidc.org/crypsphere/seaice/characteristics /differences.html Petrich, C. & H. Eicken. (2010). Growth, structure and properties of sea ice. In: Sea Ice. 2nd Edition. (eds Thomas, D.N. and Dieckmann, G.S.) Wiley-Blackwell, Oxford, UK, pp. 23-77. Renaud, P. E., A. Riedel, C. Michel, N. Morata, M. Gosselin, T. Juul-Pedersen & A. Chiuchiolo. (2007). Seasonal variation in benthic community oxygen demand: a response to an ice algal bloom in the Beaufort Sea, Canadian Arctic. Journal of Marine Systems, 67(1), 1-12. Riedel, A., C. Michel, M. Gosselin, & B. LeBlanc. (2008). Winter-spring dynamics in sea-ice carbon cycling in the coastal Arctic Ocean. Journal of Marine Systems, 74(1), 918-932. Rysgaard, S., J. Bengtsen, B. Delilli, G.S. Dieckmann, R.N. Glub, H. Kennedy, J. Mortensen, S. Papadimitriou, D.N. Thomas & J.L. Tison. (2011). Sea ice contribution to the air-sea CO2 exchange in the arctic and Southern oceans. Tellus, 63(B), 823-830. Rysgaard, S., R.N. Glud, M.K. Sejr, J. Bendtsen & P.B. Christensen. (2007). Inorganic carbon transport during sea ice growth and decay: a carbon pump in polar seas. Journal of Geophysical Research, 112(1). Sejr, M.K., D. Krause-Jensen, S. Rysgaard, L.L. Sorensen, P.B. Christensen & R.N. Glud. (2011): Air-sea flux of Co2 in arctic coastal waters influences by glacial melt water and sea ice. Tellus, 63(B), 815-822. Stedmon, C.A., D.N. Thomas, S. Papadimitriou, M.A. Granskog & G.S. Dieckmann. (2011). Using fluorescence to characterize

72 | WARD

dissolved organic matter in Antarctic sea ice brines. Journal of Geophysical Reasearch, 116(1). Thomas, D.N, S. Papadimitriou, & C. Michel. (2010). Biogeochemistry of sea ice. In: Sea Ice. 2nd Edition. (eds Thomas, D.N. and Dieckmann, G.S.) Wiley-Blackwell, Oxford, UK, pp. 425-467. Worby, A.P., C.A. Greiger, M.J. Paget, M.L. van Woert & S.F. Ackley. (2008). Thickness distribution of Antarctic sea ice. Journal of Geophysical Research, 113(1).

Total Watt Hours per Rooftop Building in Ville-Marie MALCOM ARAOS-EGAN & DANIEL HABERMAN This project created that identifies how much wattage Montreal borough of Ville-Marie, statement of the radiation potential the adoption of alternative energy PROJECT AIM.

METHODS A dataset was constructed to accurately reflect the built environment of Ville-Marie. Buildings and mountains, especially in a highly urbanized, hilly locale like Montreal, cast shadows on neighbouring roofs. These shadows decrease the amount of sun received by a particular rooftop, and the radiation potential of the roof is reduced. Thus it is essential to model the effect of shadows on the solar radiation potential of rooftops. To do this, a topographically correct dataset of Ville-Marie was created that incorporates land elevation values of a Digital Elevation Model (DEM) and the height values of buildings in the borough. This dataset was created by combining a vector dataset of building footprints containing the height attributes of buildings and a raster DEM of the borough. The solar radiation received throughout the borough was calculated using the local areal solar radiation tool in the Arc-


a geographic information system is received by rooftops in the in the hopes that an explicit of these locations will incentivize practices and urban agriculture.

GIS 10. This tool calculates the amount of kilowatt-hours received at each pixel during a certain time interval and adds it up for an entire year. The tool operates using changing values for each of the geophysical variables that affect the amount of insolation received, taking into account shadows by changing the viewshed of the sun as it moves through the sky. The variables used for calculating the location of the sun include the solar hour, zenith angle, altitude, and latitude. The solar hour angle is used to account for the uneven distribution of solar radiation received throughout the day due the tilt of the solar declination. The zenith angle is the angle the sun rays pass the upper atmosphere of the earth relative to a line perpendicular to the earth’s surface and varies according to the time of day and day of the year. The output of the tool was a map showing the amount of Kilowatt Hours received at each pixel in Ville-Marie.

Our Contributors AMANDA ALLNUTT Amanda graduated from McGill in May 2012 with a degree in Geography and Environmental Studies. She is interested in geography because she enjoys studying the ways in which humans interact with the physical environment and she also likes the diversity of the discipline. During her time at McGill, Amanda became particularly interested in the ecological determinants of health. Now that she has completed her undergraduate degree, Amanda plans on furthering her education in information studies. During her free time, she enjoys travelling, swimming, and cycling and her favourite travel destination is Ljubljana, Slovenia.

MALCOM ARAOS-EGAN Malcolm Araos Egan is a U3 Honours B.A. Geography (Urban Systems) and Political Science student interested in urban social and political geography, city planning, and open data. Specifically, Malcolm examines how politics are polarized and fragmented through space, and how cities could be redesigned to emphasize livability and social justice. Malcolm is also attracted to open data as a tool to highlight the importance of place in the realms of local government and economy. Malcolm likes geography because he thinks the discipline’s use of space is the most useful tool for analyzing and critiquing the city. Malcolm hopes to use the skills gained from his degree to plan better cities and empower vulnerable urban citizens. In his free time, Malcolm likes to walk around town, play music, visualize data, and fiddle with bicycles. Malcolm is most interested in exploring the innovative urban metropolises of South America.


ERICA CALHOON Erica is an Environmental Studies student in her third year at McGill. She enjoys geography because her mind works in a spatial manner and she likes learning about diverse populations and the ways in which they interact with their environments. Erica enjoys drawing, spending time in the outdoors, and going to music festivals. She hopes that her degree in environmental studies will allow her to work in interesting places, perhaps through environmental education. Erica’s loves to travel to the rugged national parks of the West.

JUSTINE DESMOND Justine is a recent graduate, having majored in International Development and minored in Geography and Environment. Despite having dabbled in numerous fields throughout her degree, Justine’s heart lies with the geographers of the world. Geography allows people with diverse interests and skill sets to come together to better understand and provide solutions for a variety of issues. She also loves the opportunities for travel it provides—during her undergraduate career, Justine had a chance to travel across East Africa as well as to Barbados through field courses. When not being a geo nerd, she likes to spend her time guiding canoe trips and living the camp life. She hopes to eventually be involved with resource and environmental management and environmental policy-making. Justine’s favourite travel destination is a toss up between Zanzibar and Barbados. Zanzibar has a vibrant history that can be seen all around you, not to mention some sweet scuba diving.

DANIEL HABERMAN Daniel Haberman is a U3 B.Sc. Geography and GIS student interested in physical geography, urban sustainability, and food security. Specifically, he is interested in developing methods to evaluate the feasibility of urban agriculture. He aims to develop comprehensive criteria for implementing farming in cities, so he analyzes yield, transportation, and business models of current urban agriculture projects. Dan likes geography because he feels the discipline provides the best tools for tackling current problems of global sustainability. With his degree Dan plans to continue developing his GIS skills, hopefully in the field of urban agriculture research. In his free time, Dan likes to play music, snowboard, play squash, jorkyball, and guessing people’s shoe size. Although Dan loves his native city of Montreal, he thinks the Caribbean would be his favourite travel the destination. Dan finds the laid-back vibe and warm weather very appealing. FIELD NOTES | VOL I | 77

COREY LESK Corey is a second year Earth System Science student broadly interested in the effects of human resource use on the Earth system. At the moment he is particularly interested in climate-agriculture interactions and food security in arid developing areas. He also follows global politics closely and is beginning to dabble in economics. Geography as a discipline appeals to Corey because it is diverse and pragmatic, applying scholarship to the pressing questions facing humanity, and it integrates many of his interests. Beyond school he take pleasure in music, cooking, homebrewing beer, wilderness tripping, among other hobbies. (Currently in fermenter: oatmeal rye stout.) Corey would be interested in travelling to virtually anywhere in the world, though by mid-semester he is usually fantasizing about the tranquility of a paddle in northern Quebec. After school he will likely pursue a research career, though the plan tends to change.

KIRYA MARCHAND Kirya is an English literature, Environmental Studies, and International Development Studies student in her fourth year at McGill. She is interested in studying geography because it is multi-disciplinary and highlights the humannature intersection in multifaceted ways. In her spare time, she enjoys poetry, baking, nature, and hula hooping. While she will not graduate with a degree in geography, she wants to pursue environmental philosophy. Kirya’s most recently enjoyed her trip to the Andes and would like to see the Grand Canyon soon.

DEREK LUBETKIN Derek is an Urban Systems student in his final year of study. He enjoys Geography because he feels the discipline bridges the gap between human and environmental needs. Upon graduation, he plans on using his degree to better understand the world around him. Outside of the classroom, Derek likes travelling and exploring with his dog (in his sidecar rig). His favourite place is his grandma’s house on the shores of Lake Nipissing, in Northern Ontario.


MARIE-LINE SARRAZIN Marie-Line is a fourth year undergraduate student completing a double major in Geography and International Development Studies. Passionate about sustainability, she focuses on biodiversity conservation and alternative development, as well as ways to promote them. Marie-Line is interested in Geography because of its attention to the interface between humans and their socio-physical environment. She plans on completing a Master degree and, hopefully, working in the non-governmental sector. Her favorite travel destination has been, up to now, Mexico.

MELISSA KARINE WARD With a major in Physical Geography and a minor in Geology, Melissa graduated from McGill with a Bachelor of Science in June 2012. She has now begun a Master’s of Science at McGill under the supervision of Professor Wayne Pollard. Melissa loves the field component of geography and wants to pursue a career in research. As such, her thesis focuses on the geomorphology and geochemistry of two perennial springs in the high Arctic on Axel Heiberg Island. Melissa loves to bake and cook and generally spends her weekends in the kitchen. She also enjoys travelling, reading novels, and spending time outside. Her favourite travel destinations are France and Australia.

JASON WONG Jason is a recent graduate of the International Development, Geography, and GIS programs at McGill. He enjoys Geography’s encompassing field of study and its emphasis on making connections—a reflection of today’s interdisciplinary reality. This winter, he will fulfill a long-time dream and volunteer in Nepal for six months. He is interested in exploring ways to integrate GIS into development projects with limited resources, particularly in microfinance. He is also particularly interested in using participatory GIS, Agent Based Modeling, and social network analysis to improve the services and reach of community development projects to rural communities. He is also excited to go whitewater rafting on the Himalayan-fed Kaligandaki River. Following these six months, Jason hopes to draw inspiration from his experiences abroad to pursue a Master’s degree or to work with the Aga Khan Foundation.


Profile for Field Notes

Field Notes - Volume I  

Field Notes: Journal of the McGill Undergraduate Geography Society

Field Notes - Volume I  

Field Notes: Journal of the McGill Undergraduate Geography Society