Issuu on Google+

View with images and charts Diagnosis and Monitoring Of Drought Using Regional Climate Model Over Bangladesh Introduction 1.1 Preamble Drought is a prolonged, continuous period of dry weather along with abnormal insufficient rainfall. It occurs when evaporation and transpiration exceed the amount of precipitation for a reasonable period. Drought causes the earth to parch and a considerable hydrologic (water) imbalance resulting water shortages, Wells to dry, depletion of groundwater and soil moisture, stream flow reduction, crops to wither leading to crop failure and scarcity in fodder for livestock. Drought is a major natural hazard faced by communities directly dependent on rainfall for drinking water, crop production, and rearing of animals. Since ancient times droughts have far-reaching effects on mankind. Large land areas often suffer damages from dust storms and fire. Drought could be the reason for migration of early human communities. It has long been considered to be a natural hazard responsible for us and downs of many civilisations of the world. It is the slow onset natural disaster, which commences before anybody notices it and by the time it is noticeable it is too late. Unlike other natural disasters, it starts unnoticed and develops cumulatively, thus its impact is not immediately observable by naked eye or ground data. It may be the most devastating, yet least understood of all weather phenomenon. Drought can erupt in a matter of months, or it can gradually creep up on an unsuspecting society over several seasons. Drought is rarely forecasted with any skill, and goes unobserved by the public until Impacts from the drought have already occurred. Inevitably, officials charged with mitigating those impacts want to know how a current drought measures up historically to other droughts in terms of intensity, areal coverage, and duration. Additionally, these three factors differ in relative time and space scales from drought to drought. Meteorological drought is a short lived, recurring natural disaster, which originate from the lack of precipitation and can bring significant economic losses [Pal et al., 2000]. It is not possible to avoid meteorological droughts, but it can be monitored, and their adverse impacts can be alleviated [Gommes, 1994]. The success of the drought prediction depends on how well it is defined and identified. Bangladesh is one of the most seriously affcteded countries suffering from meteorological disasters such as droughts in pre- and post-monsoon seasons and floods in the summer monsoon, tropical cyclones and meso-scale disturbances. Additionally, agriculture, power generation and industrial production subtantially depend upon precipitation (Devkota, 2006). In view of the critical influnence of large inter-annual variability of precipitation on agricultural and industrial production, seasonal prediction of drought becomes very important for policy making efforts [Giorgi et al., 1996]. Beside the other natural disasters, Bangladesh face drought situation. Now-a-days food security is an important issue in the World. Because drought is intimately related with food security, therefore, its diagnosis and monitoring is essential to carry out too. The diagnosis of drought is also important for the utilization of drought projection using climate modeling facilities for the stakeholders and planners of a country. 1.2 Drought: Definitions


Droughts have no universal definition. As drought definitions are region specific, reflecting differences in climate characteristics as well as incorporating different physical, biological and socio-economic variables, it is usually difficult to transfer definitions derived for one region to another. However some of the common definitions of drought can be noted as under: • The Director of Common Wealth Bureau of Meteorology in 1965 suggested a broad definition of drought as “severe water shortage”. • Definition given by Palmer states that “Drought is an interval of time, generally of the order of months of years in duration, during which the actual moisture supply at a given place rather consistently falls short of the climatically expected or climatically appropriate moisture supply (Palmer, 1965)”. • According to Mc Mohan and Diaz Arena (1982), “Drought is a period of abnormally dry weather sufficiently for the lack of precipitation to cause a serious hydrological imbalance and carries connotations of a moisture deficiency with respect to man’s usage of water”. • Another definition given by Flag is worth mentioning “ Drought is a period of rainfall deficiency, extending over months to year of such a nature that crops and pasturage for stock are seriously affected, if not completely burnt up and destroyed, water supplies are seriously depleted or dried up and sheep and cattle perish”. • According to Hangman (1984), “Drought is considered by many to be the most complex but least understood of all natural hazards affecting more people than any other hazard.” • According to National Drought Policy Commission, “A persistent and abnormal moisture deficiency having adverse impacts on vegetation, animals, and people”. • According to The Convention to Combat Desertification (CCD), “drought” means the naturally occurring phenomenon that exists when precipitation has been significantly below normal recorded levels, causing serious hydrological imbalances that adversely affect land resources production systems. A drought is a complex phenomenon that can be defined from several perspectives. Wilhite and Glantz ((2000) categorize drought definitions into conceptual (definitions formulated in general terms) and operational. Conceptual definitions formulated in general terms; help people understand the concept of drought but these normally do not provide quantitative answers. Operational definitions on the other hand help identify the drought beginning, end and degree of severity. By studying the above definitions it can be understood that drought is mainly concerned with the shortage of water which in turn affects availability of food and fodder thereby leading to displacement and loss to economies as a whole. 1.3 Classification of Drought Drought can be classified in three main ways: • Meteorological drought: related to rainfall amounts • Hydrological drought: determined by water levels in reservoir • Agricultural drought: related to the availability of water for crops. 1.3.1 Meteorological Drought Meteorological drought is generally defined by comparing the rainfall in a particular place and at a particular time with the average rainfall for that place. The definition is, therefore, specific to a particular location. Meteorological drought leads to a depletion of soil moisture and this almost always has an impact on crop production. When we define drought this way,


we only consider the reduction in rainfall amounts and don't take into account the effects of the lack of water on water resevoirs, human needs or on agriculture. Meteorologically drought can be classified into three types: permanent drought characterised by arid climate; seasonal drought - caused by irregularities in recognised rainy and dry seasons; and contingent drought - caused by irregular rainfall. In Bangladesh, the last two types are more prevalent. 1.3.2 Hydrological Drought Hydrological drought is associated with the effect of low rainfall on water levels in rivers, reservoirs, lakes and aquifers. Hydrological droughts usually are noticed some time after meteorological droughts. First precipitation decreases and, some time after that, water levels in rivers and lakes drops. Hydrological drought affects uses which depend on the water levels. Changes in water levels affect ecosystems, hydro electrical power production and recreational, industrial and urban water use. 1.3.3 Agricultural drought Agricultural drought occurs when there is not enough water available for a particular crop to grow at a particular time. This drought doesn’t depend only in the amount of rainfall, but also on the correct use of that water. Imagine a period of low rainfall where water is used carelessly for irrigation and other purposes. Under these circumstances, the effect of the drought becomes more pronounced than it was before. Agricultural drought is typically seen after meteorological drought (when rainfall decreases) but before a hydrological drought (when the water level in rivers, lakes and reservoirs decreases). It is important to mention that the effects of droughts are different in irrigated and nonirrigated agriculture. In regions which rely on irrigation, the impacts of short lived agricultural droughts are usually lower than in regions where crops are not irrigated. Irrigated agriculture relies on stocks of water so if it doesn't rain, these crops still get the water they need (until the reservoirs run dry). However, in non-irrigated agriculture crops depend directly on the rain as their water source. If it doesn't rain, the crops don't get the water they need to survive. 1.4 Drought in Bangladesh Bangladesh extends from 20o34'N to 26o38'N latitude and from 88001'E to 92041'E longitude. Climatically, the country belongs to sub-tropical region where monsoon weather prevails throughout the year. The average temperature of the country ranges from 17 to 20.6 oC during winter and 26.9 to 3l.1oC during summer. Four distinct seasons can be recognized in Bangladesh from climatic point of view: (i) the winter season from December to February, (ii) the pre-monsoon season from March to May, (iii) the monsoon season which lasts from June to September and (iv) Post-Monsoon season lasting from October and November. The average annual rainfall of the country varies from 1329 mm in the northwest to 4338 mm in the northeast (Shahid et a1., 2005). The gradient of rainfall from west to east is approximately 9 mm/km. The western part of Bangladesh experiences an average areal rainfall of approximately 2044 mm, which is much lower than other parts of the country. The rainfall is also very much seasonal, almost 66% of rainfall occurs during monsoon. In summer, the hottest days experience temperatures of about 45oC in the North-Western region. Again in the winter the temperature even falls at 5oC in some places.


On the global level, impact of natural hazards and disasters are staggering. In Bangladesh, the major natural hazards are also in line with global patterns. In the context of global warming, most of the climate models project a decrease in precipitation in dry season and an increase during monsoon in south Asia (Christensen et a1., 2007). This will cause a cruel combination of more extreme floods and droughts in this region. According to the report of National Drought

Fig 2.1.1 Frequency of drought occurrences all over the World by country during 1974 to 2003 shows the drought situation all over the world. Source: CRED International Disaster Database EM-DAT. Mitigation Center (2006) Bangladesh has already shown an increased frequency of droughts in recent years. Concern among scientists has grown on changes of precipitation and frequent occurrence of droughts both in Bangladesh. Therefore, study on drought hazards especially drought monitoring and projection are essential for implementing mitigation to reduce drought impact in Bangladesh. After 1971 Bangladesh has experienced droughts of major magnitude in 1973, 1978, 1979, 1981, 1982, 1989, 1992, 1994, and 1995. Although droughts are not always continuous in any area, they do occur sometimes in the low rainfall zones of the country. As listed above, Bangladesh experienced consecutive droughts in 1978-1979, 1981- 1982, and 1994-1995. The droughts of 1994-95 in the northwestern districts of Bangladesh led to a shortfall of rice production of 3.5 million tons (Paul, 1995). Literature Review 2.1 Review of the previous work In Bangladesh, a number of studies have been carried out on the impact of droughts on agriculture (Jabbar et al., 1982; Karim et al., 1990; Jabbar, 1990; Saleh et a1., 2000; Mazid et al., 2005), food production (Ahmed and Bemard, 1989; Ericksen et al.,1993), land degradation (Rasheed,1998; Karim and lqbal, 2001; Government of Bangladesh, 2005), economy (Erickson et al., 1993; World Bank Bangladesh, 2000), and society (Erickson et al., 1993; Paul, 1998) in Bangladesh. WARPO-EGIC (1996) prepared maps of winter and premonsoon drought prone areas of Bangladesh using the agro ecological zones database and land resources inventory map. Up to date so far, No standard drought index method has been used for the assessment of projection of droughts in Bangladesh using Regional Climate Model simulated output data.


Jabbar et al. (1982), emphasized the combined role of drought hazard and vulnerability in defining risk. Standardized precipitation index method in a GIS environment was used to map the spatial extents of drought hazards in different time steps. The key social and physical factors that define drought vulnerability in the context of Bangladesh were identified and corresponding thematic maps in district level are prepared. Composite drought vulnerability map was developed through the integration of those thematic maps. The risk was computed as the product of the hazard and vulnerability. The result showed that droughts pose highest risk to the northern and north-western districts of Bangladesh. Ministry of environment and Forests, Government of the Peoples Republic of Bangladesh, Dhaka (2001), produced a notional report on “Implementation of United Nations Convention to Combat Desertification”. On this report drought was calculated considering the distribution of rainfall and evapotranspiration regimes and the drought condition in the country, it was proposed that the regions fulfilling the following conditions may comprise dry regions in Bangladesh. The conditions are: (i) annual rainfall ® should be less than 2000 mm; (ii) dry season (November – May) Excess Evapotranspiration (ETo-R) should be more than 400 mm and (iii) dry season R/ ETo ratio value should be less than 0.65. With this assumption made and applied on the rainfall and evapotranspiration data available for the agro ecological zones of Bangladesh, the Northwest, Southwest and North central zones can be considered as dry region of the country. It was further proposed that the dry regions may be divided into two sub regions on the basis of severity of dryness as below: Table 2.1.1 Severity of Dry Regions Severity Moderate

Conditions/ Criteria Defined Annual Rainfall ® less than 1600 mm Dry season (November-May) Excess Evapotranspiration (ETo- R) - more than 400 mm Dry season (November-May) R/ ETo ratio value less than 0.4 Slight Annual Rainfall ® 1600-2000 mm Dry season (November-May) Excess Evapotranspiration (ETo- R ) - 200-400 mm Dry season (November-May) R/ ETo ratio value 0.4 - 0.65 Non dry Annual Rainfall ® more than2000 mm Dry season (November-May) Excess Evapotranspiration (ETo- R ) – less than 200 mm Dry season (November-May) R/ ETo ratio value more than 0.65 Considering the rainfall and evapotranspiration data available in WARPO a map was prepared to show the severity of dryness in the dry regions. A map of the dry regions of Bangladesh had been prepared. The Rivers and Estuary hydrological region; the coastal region and the Sunderbans are considered as non-dry. The dry zones are extended over an area of 6.442 M ha. The extents of the dry zones are given in the following table: Table 2.1.2 Extent of Dry Zones of Bangladesh Dry Zones Moderate Slight Nondry

No. of Thanas occupied 64 163 263

Area covered (M ha) Total land (%) 2.015 14.37 4.427 31.56 7.585 54.07


Two critical dry periods were distinguished (Karim et al., 1990), Rabi and pre-Kharif drought (January - May), due to: (i) the cumulative effect of dry days; (ii) higher temperatures during pre-Kharif (> 40 degrees Celsius in March-May); and (iii) low soil moisture availability. This drought affects all the Rabi crops, such as HYV Boro, Aus, wheat, pulses and potatoes especially where irrigation possibilities are limited. It also affects sugarcane production. Kharif droughts in the period June/July to October, created by sub-humid and dry conditions in the highland and medium highland areas of the country (in addition to the west/northwest also the Madhupur tract is drought prone). Shortage of rainfall affects the critical reproductive stages of transplanted Aman crops in December, reducing its yield, particularly in those areas with low soil moisture holding capacity. Considering the Agro ecological Zones (AEZ) database and land resources inventory map at 1:1,000,000 scale, BARC has identified and mapped drought prone areas of Bangladesh for Rabi and Pre-Kharif seasons (WARPO- EGIS, 1996). Recently BARC has reviewed this concept and produced three different maps for Rabi, Pre-Kharif and Kharif seasons (BARC, 2001). The drought maps had been revised by BARC to produce three maps for Rabi, PreKharif and Kharif seasons. The drought severity classes were defined; Moderate, severe and Very severe related to the yield losses of 15-20%, 20-35%, 35-45%, and 45-70% respectively for different crops (Karim and Iqbal, 2001). The northwestern part is prone to drought mainly because of rainfall variability in the premonsoon and the postmonsoon periods. Inadequate pre-monsoon showers, a delay in the onset of the rainy season or an early departure of the monsoon may create drought conditions in Bangladesh, and adversely affect crop output. Since it puts severe strain on the land potential. It acts as a catalyst of land degradation through reduced soil moisture and water retention, increased soil erosion, decline in soil organic contents and overexploitation of sparse vegetation. Human interventions in the form of land abuse and mismanagement have exacerbated these actions during the spells of periodic droughts. An analysis of the relative effects of flood and drought on rice production between 1969-70 and 1983-84 shows that drought is more devastating than floods to aggregate production (World Bank, 2000). 2.2 Meteorological Drought indices The impact of drought on society and agriculture is a real issue but it is not easily quantified. Reliable indices to detect the spatial and temporal dimensions of drought occurrences and its intensity are necessary to assess the impact and also for decision-making and crop research priorities for alleviation. Several indices have been developed for quantification of drought based on the type of drought. Palmer Drought Severity Index developed by Palmer (1965), is the most widely used drought index based on the demand and supply concept of water balance equation. Palmer (1968) derived another index, the Crop Moisture Index (CMI) by modifying PDSI to find out the severity of agricultural drought. Hydrological droughts characterized by low precipitation, lowering groundwater level. The National Drought Mitigation Center is using a newer index, the Standardized Precipitation Index, to monitor moisture supply conditions. Distinguishing traits of this index are that it identifies emerging droughts months sooner than the Palmer Index and that it is computed on various time scales. Most water supply planners find it useful to consult one or more indices before making a decision. A brief review of these indices is given below:


2.2 .1 Percent of Normal The percent of normal is a simple calculation well suited to the needs of TV weathercasters and general audiences. It is Quite effective for comparing a single region or season though, it could be misunderstood, as normal is a mathematical construct that does not necessarily correspond with what we expect the weather to be. The percent of normal precipitation is one of the simplest measurements of rainfall for a location. Analyses using the percent of normal are very effective when used for a single region or a single season. Percent of normal is also easily misunderstood and gives different indications of conditions, depending on the location and season. It is calculated by dividing actual precipitation by normal precipitation—typically considered to be a 30-year mean—and multiplying by 100%. This can be calculated for a variety of time scales. Usually these time scales range from a single month to a group of months representing a particular season, to an annual or water year. Normal precipitation for a specific location is considered to be 100%. One of the disadvantages of using the percent of normal precipitation is that the mean, or average, precipitation is often not the same as the median precipitation, which is the value exceeded by 50% of the precipitation occurrences in a long-term climate record. The reason for this is that precipitation on monthly or seasonal scales does not have a normal distribution. Use of the percent of normal comparison implies a normal distribution where the mean and median are considered to be the same. An example of the confusion this could create can be illustrated by the long-term precipitation record in Melbourne, Australia, for the month of January. The median January precipitation is 36.0 mm (1.4 in.), meaning that in half the years less than 36.0 mm is recorded, and in half the years more than 36.0 mm is recorded. However, a monthly January total of 36.0 mm would be only 75% of normal when compared to the mean, which is often considered to be quite dry. Because of the variety in the precipitation records over time and location, there is no way to determine the frequency of the departures from normal or compare different locations. This makes it difficult to link a value of a departure with a specific impact occurring as a result of the departure, inhibiting attempts to mitigate the risks of drought based on the departures from normal and form a plan of response. 2.2 .2 Standardized Precipitation Index (SPI) The SPI is an index based on the probability of precipitation for any time scale. Many drought planners appreciate the SPI’s versatility. The SPI can be computed for different time scales, can provide early warning of drought and help assess drought severity, and is less complex than the Palmer. Developed by T.B. McKee, N.J. Doesken, and J. Kleist, Colorado State University, 1993. The understanding that a deficit of precipitation has different impacts on groundwater, reservoir storage, soil moisture, snowpack, and streamflow led McKee, Doesken, and Kleist to develop the Standardized Precipitation Index (SPI) in 1993. The SPI was designed to quantify the precipitation deficit for multiple time scales. These time scales reflect the impact of drought on the availability of the different water resources. Soil moisture conditions respond to precipitation anomalies on a relatively short scale. Groundwater, streamflow, and reservoir storage reflect the longer-term precipitation anomalies. For these reasons, McKee et al. (1993) originally calculated the SPI for 3–, 6–,12–, 24–, and 48–month time scales. The SPI calculation for any location is based on the long-term precipitation record for a desired period. This long-term record is fitted to a probability distribution, which is then transformed into a normal distribution so that the mean SPI for the location and desired period is zero (Edwards and McKee, 1997). Positive SPI values indicate greater than median precipitation, and negative values indicate less than median precipitation. Because the SPI is


normalized, wetter and drier climates can be represented in the same way, and wet periods can also be monitored using the SPI. McKee et al. (1993) used the classification system shown in the SPI values table to define drought intensities resulting from the SPI. McKee et al. (1993) also defined the criteria for a drought event for any of the time scales. A drought event occurs any time the SPI is continuously negative and reaches an intensity of -1.0 or less. The event ends when the SPI becomes positive. Each drought event, therefore, has a duration defined by its beginning and end, and an intensity for each month that the event continues. The positive sum of the SPI for all the months within a drought event can be termed the drought’s “magnitude”. Mathematical derivation of SPI is discussed in the chapter 4. 2.2 .3 Palmer Drought Severity Index (PDSI) The Palmer is a soil moisture algorithm calibrated for relatively homogeneous regions. Many U.S. government agencies and states rely on the Palmer to trigger drought relief programs. It is the first comprehensive drought index developed in the United States. Palmer values may lag emerging droughts by several months; less well suited for mountainous land or areas of frequent climatic extremes; complex—has an unspecified, built-in time scale that can be misleading. W.C. Palmer developed an index to measure the departure of the moisture supply (Palmer, 1965). Palmer based his index on the supply-and-demand concept of the water balance equation, taking into account more than just the precipitation deficit at specific locations. The objective of the Palmer Drought Severity Index (PDSI), as this index is now called, was to provide measurements of moisture conditions that were standardized so that comparisons using the index could be made between locations and between months (Palmer 1965). The PDSI is a meteorological drought index, and it responds to weather conditions that have been abnormally dry or abnormally wet. When conditions change from dry to normal or wet, for example, the drought measured by the PDSI ends without taking into account streamflow, lake and reservoir levels, and other longer-term hydrologic impacts (Karl and Knight, 1985). The PDSI is calculated based on precipitation and temperature data, as well as the local Available Water Content (AWC) of the soil. From the inputs, all the basic terms of the water balance equation can be determined, including evapotranspiration, soil recharge, runoff, and moisture loss from the surface layer. Human impacts on the water balance, such as irrigation, are not considered. Complete descriptions of the equations can be found in the original study by Palmer (1965) and in the more recent analysis by Alley (1984). 2.2 .4 Crop Moisture Index (CMI) The Crop Moisture Index (CMI) uses a meteorological approach to monitor week-to-week crop conditions. It was developed by Palmer (1968) from procedures within the calculation of the PDSI. Whereas the PDSI monitors long-term meteorological wet and dry spells, the CMI was designed to evaluate short-term moisture conditions across major crop-producing regions. It is based on the mean temperature and total precipitation for each week within a climate division, as well as the CMI value from the previous week. 2.2 .5 Surface Water Supply Index The Surface Water Supply Index (SWSI) was developed by Shafer and Dezman (1982) to complement the Palmer Index for moisture conditions across the state of Colorado. The Palmer Index is basically a soil moisture algorithm calibrated for relatively homogeneous regions, but it is not designed for large topographic variations across a region and it does not account for snow accumulation and subsequent runoff. Shafer and Dezman designed the


SWSI to be an indicator of surface water conditions and described the index as “mountain water dependent”, in which mountain snowpack is a major component. The objective of the SWSI was to incorporate both hydrological and climatological features into a single index value resembling the Palmer Index for each major river basin in the state of Colorado (Shafer and Dezman 1982). These values would be standardized to allow comparisons between basins. Four inputs are required within the SWSI: snowpack, streamflow, precipitation, and reservoir storage. Because it is dependent on the season, the SWSI is computed with only snowpack, precipitation, and reservoir storage in the winter. During the summer months, streamflow replaces snowpack as a component within the SWSI equation. 2.2 .6 Reclamation Drought Index The Reclamation Drought Index (RDI) was recently developed as a tool for defining drought severity and duration, and for predicting the onset and end of periods of drought. The impetus to devise the RDI came from the Reclamation States Drought Assistance Act of 1988, which allows states to seek assistance from the Bureau of Reclamation to mitigate the effects of drought. Like the SWSI, the RDI is calculated at a river basin level, and it incorporates the supply components of precipitation, snowpack, streamflow, and reservoir levels. The RDI differs from the SWSI in that it builds a temperature-based demand component and duration into the index. The RDI is adaptable to each particular region and its main strength is its ability to account for both climate and water supply factors. Oklahoma has developed its own version of the RDI and plans to use the index as one tool within the monitoring system designated in the state’s drought plan. The RDI values and severity designations are similar to the SPI, PDSI, and SWSI. 2.3 Objectives of the present study The aim of the present study is to characterize the spatial and temporal pattern of drought hazards, identify the possibility of utilizing climate model outputs in drought diagnosis and monitoring in Bangladesh. Regional Climate Model (RegCM) successfully forecasts climatic parameters with some discrepancy between real & model data. It is essential to do some validation of the RegCM model outputs with the ground based data; rain-gauge and surface air temperature to adopt the RegCM for this region [Liu et al., 1996]. The studies on prediction of drought over Bangladesh exhibits a large spatial and temporal variability [Islam, M. N, 2005]. This study will allow us to identify the drought conditions for a large coverage using observed data in Bangladesh. The same will be obtained from RegCM outputs. This will allow to obtain the model efficiency in detection of drought over the country. Once the model data is found useful in drought diagnosis in Bangladesh then it may be useful in projection of drought events which are helpful for planners and decision makers. Regional Climate Model (RegCM) 3.1 Introduction Climate models are numerical representations of the fundamental equations that describe the behaviour of the climate system and the interactions across its components: Atmosphere, ocean, cryosphere, biosphere and chemosphere. It is a huge computer codes based on fundamental mathematical equations of motion, thermodynamics and radiative transfer. These equations govern: • Flow of air and water • Exchange of heat, water & momentum between atmosphere and earth.


• Release of latent heat by condensation during the formulation of clouds and raindrops. •Absorption of sunshine and emission of thermal radiation (infra-red) The equations of climate model are: a. Horizontal equation of motion 

  1    dU = −2Ω× u − ∇. P + g + Fr ; (Conservation of momentum) dt ρ

b. Hydrostatic assumption dP = −ρg (Conservation of water) dz

c. Continuity equation

1 dρ   + ∇.u = 0 (Conservation of mass) ρ dt

d. Conservation of energy dQ = dU + dW (1st law of thermodynamics) e. Equation of state for gas P = ρRT The idea that limited area models (LAMs) could be used for regional studies was originally proposed by Dickinson et al. (1989) and Giorgi (1990). This idea was based on the concept of one-way nesting, in which large scale meteorological fields from General Circulation Model (GCM) runs provide initial and time dependent meteorological lateral boundary conditions (LBCs) for high resolution Regional Climate Model (RCM) simulations, with no feedback from the RCM to the driving GCM. The first generation NCAR RegCM was built upon the NCAR-Pennsylvania State University (PSU) Mesoscale Model version 4 (MM4) in the late 1980s (Dickinson et al., 1989; Giorgi, 1989). For application of the MM4 to climate studies, a number of physics parameterizations were replaced, mostly in the areas of radiative transfer and land surface physics, which led to the first generation RegCM (Dickinson et al., 1989; Giorgi, 1990). The first generation RegCM included the Biosphere-Atmosphere Transfer Scheme, BATS (Dickinson et al., 1986) for surface process representation, the radiative transfer scheme of the Community Climate Model version 1 (CCM1), a medium resolution local planetary boundary layer scheme, the Kuo-type cumulus convection scheme of (Anthes, 1977) , and the explicit moisture scheme of (Hsie et al., 1984). Changes in the model physics include a large-scale cloud and precipitation scheme which accounts for the subgrid-scale variability of clouds (Pal et al., 2000), new parameterizations for ocean surface fluxes (Zeng et al) 1998), and a cumulus convection scheme (Emanuel, 1991; Emanuel and Zivkovic-Rothman, 1999). Also new in the model is a mosaic-type parameterization of subgrid-scale heterogeneity in topography and land use (Giorgi et al., 2003b). Other improvements in RegCM3 involve the input data. Lastly, improvements in the user-friendliness of the model have been made. New scripts have been included which make running the programs easier. Also, a new website has been developed where users can freely download the entire RegCM system, as well as all of the input data necessary for a simulation. The RegCM modeling system has four components: Terrain, ICBC, RegCM, and Postprocessor. Terrain and ICBC are the two components of RegCM preprocessor. Terrestrial variables (including elevation, land use and sea surface temperature) and three-dimensional isobaric meteorological data are horizontally interpolated from a latitude-longitude mesh to a


high-resolution domain on either a Rotated (and Normal) Mercator, Lambert Conformal, or Polar Stereographic projection. Vertical interpolation from pressure levels to the coordinate system of RegCM is also performed. 3.2 Physics 3.2.1 Radiation Scheme RegCM3 uses the radiation scheme of the NCAR CCM3, which is described in Kiehl et al. (1996). Briefly, the solar component, which accounts for the effect of O3, H2O, CO2, and O2, follows the d-Eddington approximation of Kiehl et al. (1996). It includes 18 spectral intervals from 0.2 to 5 μm. The cloud scattering and absorption parameterization follow that of Slingo (1989), whereby the optical properties of the cloud droplets (extinction optical depth, single scattering albedo, and asymmetry parameter) are expressed in terms of the cloud liquid water content and an effective droplet radius. 3.2.2 Land Surface Model The surface physics are performed using Biosphere-Atmosphere Transfer Scheme version 1e (BATS1e) which is described in detail by Dickinson et al. (1993). BATS is a surface package designed to describe the role of vegetation and interactive soil moisture in modifying the surface-atmosphere exchanges of momentum, energy, and water vapour. The model has a vegetation layer, a snow layer, a surface soil layer 10 cm thick, or root zone layer 1-2 m thick, and a third deep soil layer 3 m thick. Prognostic equations are solved for the soil layer temperatures using a generalization of the force-restore method of Deardoff (1978). 3.2.3 Planetary Boundary Layer Scheme The planetary boundary layer scheme, developed by Holtslag et al. (1990), is based on a nonlocal diffusion concept that takes into account countergradient fluxes resulting from large-scale eddies in an unstable, well-mixed atmosphere. 3.2.4 Convective Precipitation Schemes Convective precipitation is computed using one of three schemes: (1) Modified-Kuo scheme (Anthes, 1977); (2) Grell scheme (Grell, 1993); and (3) MIT-Emanuel scheme (Emanuel, 1991; Emanuel and Zivkovic-Rothman, 1999). In addition, the Grell parameterization is implemented using one of two closure assumptions: (1) the Arakawa and Schubert closure Grell et al. (1994) and (2) the Fritsch and Chappell closure (Fritsch and Chappell, 1980) hereafter referred to as AS74 and FC80, respectively. 3.3 Pre-Processing and simulation Before performing a regional climate simulation there are two pre-processing steps that need to be completed. The first step involves defining the domain and grid interval, and interpolating the landuse and elevation data to the model grid. The second step is to generate the files used for the initial and boundary conditions during the simulation. The input data necessary to run the model can be downloaded from the PWC website at the following URL: http://www.ictp.trieste.it/pubregcm/RegCM3. The regcm.x script will compile and execute the model. Compile the source code and start the simulation. Running the model generates the following monthly output files, Atmospheric model output (see Table 3.1.1.): ATM.YYYYMMDDHH Land surface model output (see Table 3.1.2.): SRF.YYYYMMDDHH Radiation model output : RAD.YYYYMMDDHH


3.4 Post-processing The model generates three output files every month • ATM.YYYYMMDDHH from the atmospheric model (see Table 3.1.1 for list of variables) • SRF.YYYYMMDDHH from the land surface model (see Table 3.1.2 for list of variables) • RAD.YYYYMMDDHH from the radiation model. The RegCM postprocessor converts these model output files to new output files of averaged variables in commonly used formats GrADS (Grid Analysis Display System). This will need to modify the postproc.in file in working directory to specify how to average the variables (daily, monthly, etc.) and the file format. Then run the postproc.x script which will compile and execute the program. Table 3.1.1. List of output variables from atmosphere. Variables u w qv

Description Eastward wind (m s−1) Omega (hPa) p-velocity Water vapour Mixing ratio (g kg1 ) v Northward wind (m s-1) tpr Total precipitation (mm) t Temperature (K) tgb Lower soil layer temp (K) qc Cloud water mixing ratio (g kg−1) smt Total soil water (mm) psa Surface pressure (Pa) rno Base flow (mm day−1) Table 3.1.2. List of output variables from surface model. Variables u10m

Description Anemometer eastward wind (ms-1)

v10m uvdrag

Anemometer northward wind (ms-1) Surface drag stress

tgb

Ground temperature (K)

tlef t2m q2m ssw

Foliage temperature (K) Anemometer temperature (K) Anemometer specific humidity kg kg-1 Top layer soil moisture (mm)

rsw tpr evp

Root layer soil moisture (mm) Total precipitation (mm day-1) Evapotranspiration (mm day-1)

runoff

Surface runoff (mm day-1)

scv

Snow water equivalent (mm)


sena

Sensible heat (W m-2)

flw

Net longwave (W m-2)

fsw

Net solar absorbed (W m-2)

flwd

Downward longwave (W m-2)

sina

Solar incident (W m-2)

prcv

Convective precipitation (mm day-1)

psb

Surface pressure (Pa)

zpbl

PBL height (m)

tgmax

maximum ground temperature (K)

tgmin

minimum ground temperature (K)

t2max

maximum 2m temperature (K)

t2min

minimum 2m temperature (K)

w10max

maximum 10m wind speed (m s-1)

psmin

minimum surface pressure (hPa)

Standardized Precipitation Index McKee et al. (1993) developed the Standardized Precipitation Index (SPI) for the purpose of defining and monitoring drought. Among others, the Colorado Climate Center, the Western Regional Climate Center, and the National Drought Mitigation Center use the SPI to monitor current states of drought in the United States. The nature of the SPI allows an analyst to determine the rarity of a drought or an anomalously wet event at a particular time scale for any location in the world that has a precipitation record. Thom (1966) found the gamma distribution to fit climatological precipitation time series well. The gamma distribution is defined by its frequency or probability density function: g ( x) =

1 x α−1e −x / β βΓ(α)

for x>0

(4.1)

Where: α >0 β >0 x >0

α is a shape parameter β is a scale parameter x is a precipitation amount

(4.2) (4.3) (4.4)

Γ(α) = ∫ y α−1e −y dy

Г(α) is the gamma function

(4.5)

0

Computation of the SPI involves fitting a gamma probability density function to a given frequency distribution of precipitation totals for a station. The alpha and beta parameters of the gamma probability density function are estimated for each station, for each time scale of interest (3 months, 12 months, 48 months, etc.), and for each month of the year. From Thom (1966), the maximum likelihood solutions are used to optimally estimate α and β: αˆ =

1  4 A  1 + 1 + 4 A  3 

x βˆ = αˆ

(4.6) (4.7)


Where: A = ln( x ) −

∑ln( x)

(4.8)

n

n = number of precipitation observations (4.9) The resulting parameters are then used to find the cumulative probability of an observed precipitation event for the given month and time scale for the station in question. The cumulative probability is given by: x

G ( x) = ∫ g ( x) dx = 0

Letting

ˆ, t =x / β

1

x

ˆ) ∫ βˆ αˆ Γ(α 0

ˆ

x αˆ −1e −x / β dx

(4.10)

this equation becomes the incomplete gamma function :

x

G ( x) =

1 t αˆ −1e −t dt ˆ) ∫ Γ(α 0

(4.11)

Since the gamma function is undefined for x=0 and a precipitation distribution may contain zeros, the cumulative probability becomes: H ( x) = q + (1 − q )G ( x)

(4.12) where q is the probability of a zero. If m is the number of zeros in a precipitation time series, Thom (1966) states that q can be estimated by m/n. Thom (1966) uses tables of the incomplete gamma function to determine the cumulative probability G(x). McKee et al. (1993) use an analytic method along with suggested software code from Press et al. (1988) to determine the cumulative probability. The cumulative probability, H(x), is then transformed to the standard normal random variable Z with mean zero and variance of one, which is the value of the SPI. This is an equiprobability transformation which Panofsky and Brier (1958) state has the essential feature of transforming a variate from one distribution (ie. gamma) to a variate with a distribution of prescribed form (ie. standard normal) such that the probability of being less than a given value of the variate shall be the same as the probability of being less than the corresponding value of the transformed variate. This method is illustrated in fig. 4.1. In this Fig, a 3 month precipitation amount (January through March) is converted to a SPI value with mean of zero and variance of one. The left side of fig 4.1. contains a broken line with horizontal hash marks that designate actual values of 3 month precipitation amounts (x-axis) for Fort Collins, Colorado for the months of January through March for the period 1911


Fig.4.1. Example of equiprobability transformation from fitted gamma distribution to the standard normal distribution. through 1995. The broken line also denotes the empirical cumulative probability distribution (y-axis) for the period of record. The empirical cumulative probabilities were found optimally as suggested by Panofsky and Brier (1958) where the precipitation data is sorted in increasing order of magnitude so that the kth value is k-1 values from the lowest and where n is the sample size: empirical cumulative probability =

k n +1

(4.13)

The smooth curve on the left hand side of Fig 4.1 denotes the cumulative probability distribution of the fitted gamma distribution to the precipitation data. The smooth curve on the right hand side of Fig 4.2 denotes the cumulative probability distribution of the standard normal random variable Z using the same cumulative probability scale of the empirical distribution and the fitted gamma distribution on the left hand side of the Fig. The standard normal variable Z (or the SPI value) is denoted on the x-axis on the right hand side of the Fig. Hence, this Fig can be used to transform a given 3 month (January through March) precipitation observation from Fort Collins, Colorado to a SPI value. For example, to find the SPI value for a 2 inch precipitation observation, simply go vertically upwards from the 2 inch mark on the x-axis on the left hand side of Fig 4.2 until the fitted gamma cumulative probability distribution curve is intersected. Then go horizontally (maintaining equal cumulative probability) to the right until the curve of the standard normal cumulative probability distribution is intersected. Then proceed vertically downward to the x-axis on the right hand side of Fig 3.2 in order to determine the SPI value. In this case, the SPI value is approximately +0.3. Since it would be cumbersome to produce these types of Figs for all stations at all time scales and for each month of the year, the Z or SPI value is more easily obtained computationally using an approximation provided by Abramowitz and Stegun (1965) that converts cumulative probability to the standard normal random variable Z:

 c0 + c1t + c2 t 2   Z = SPI = − t − for 0 < H ( x ) ≤ 0.5 2 3 1 + d t + d t + d t 1 2 3   2   c0 + c1t + c2 t  0.5 < H ( x ) ≤ 1.0 Z = SPI = +  t − 2 3  for 1 + d t + d t + d t 1 2 3   Where:  1 t = ln  ( H ( x)) 2 

   

 1 t = ln 1.0 − ( H ( x)) 2 

c 0 =2.515517

c1 =0.802853 c 2 =0.010328 d1 =1.432788

for 0 < H ( x ) ≤ 0.5    

for 0.5 < H ( x ) ≤ 1.0

(4.14) (4.14)

(3.17)

(3.18)


d 2 =0.189269 d 3 =0.001308

Conceptually, the SPI represents a z-score, or the number of standard deviations above or below that an event is from the mean. However, this is not exactly true for short time scales since the original precipitation distribution is skewed. Nevertheless, fig. 4.2 shows that during the base period for which the gamma parameters are estimated, the SPI will have a standard normal distribution with an expected value of zero and a variance of one. Katz and Glantz (1986) state that requiring an index to have a fixed expected value and variance is desirable in order to make comparisons of index values among different stations and regions meaningful. Additionally, no matter the location or time scale, the SPI represents a cumulative probability in relation to the base period for which the gamma parameters were estimated. Table 4.1 is a table of SPI and its corresponding cumulative probability. An analyst with a time series of monthly precipitation data for a location can calculate the SPI for any month in the record for the previous i months where i=1, 2, 3, ..., 12, ..., 24, ..., 48, ... depending upon the time scale of interest. Hence, the SPI can be computed for an observation of a 3 month total of precipitation as well as a 48 month total of precipitation. For this study, a 1, 3 and 6 month SPI are used for a short-term or seasonal drought index, a 9, 12 and 24 month SPI are used for an longe-term drought index. Therefore, the SPI for a month/year in the period of record is dependent upon the time scale.

Fig. 4.2. Standard normal distribution with the SPI having a mean of zero and a variance of one. Table 4.1. : SPI and Corresponding Cumulative Probability in Relation to the Base Period. SPI Cumulative Probability -3.0 0.0014 -2.5 0.0062 -2.0 0.0228 -1.5 0.0668 -1.0 0.1587 -.05 0.3085 0.0 0.5000 +0.5 0.6915


+1.0 +1.5 +2.0 +2.5 +3.0

0.8413 0.9332 0.9772 0.9938 0.9986

Data used and Methodology 5.1 Data used for analysis In order to diagnosis and monitor drought in Bangladesh, rainfall collected by Bangladesh Meteorological Department (BMD) for 27 stations over Bangladesh is utilized during 1961 to 1990. Regional Climate Model (RegCM) with Grell Arakawa and Schubert Scheme (GAS) output rainfall are extracted at rain-gauge locations of BMD during 1961 to 1990.

Fig. 5.1.1. Map of Bangladesh with the name of observation sites, observation location, elevation (in m) and detailed analyzed stations (circle). Record of drought event of Bangladesh obtained from â&#x20AC;&#x2DC;Agricultural Statistics Year book of Bangladeshâ&#x20AC;&#x2122; published by Bangladesh Bureau of Statistics (BBS) archive from 1975 to 1990. Chronology of major drought events in Bangladesh is also used from International Disaster database (EM-DAT) archive during 1971-1990. There was no information regarding drought in Bangladesh before the year 1971. Station names over Bangladesh are shown above the station location (plus mark, Fig.5.1.1) with the elevation (below plus mark, Fig. 5.1.1). The symbol circle is used with plus marks to indicate the stations which are used in detail study. These stations are namely Rangpur, Dhaka, Khulna, Cox's Bazar and Sylhet.


To analyze the drought condition of whole country, 27 BMD stations are divided into four regions according to topography and the record of rainfall by BMD (Fig. 5.2.1.) named (a) Central Region- Chandpur, Comilla, Dhaka, Faridpur, Feni, Mymensingh, Tangail, (b) North Region- Bogra, Dinajpur, Rajshahi, Rangpur, (c) South-West Region- Barisal, Hitiya, Jessore, Khepupara, Khulna, Mcourt, Patuakhali and (d) Eastern Region- Chittagong, Coxâ&#x20AC;&#x2122;s bazar, Kutubdia, Rangamati, Sandip, Sitakundu, Srimangal, Sylhet, and Teknaf. 4

Rainfall Deviation from 50 years average during 1950-2006 50 years average= 6.049 mm/d

2

-3

Mcourt

Patuakhali

Khulna

Khepupara

Hitiya

Jessor

Barisal

Sylhet

Taknaf

Srimangal

Sandip

Sitakunda

Kutubdia

Rangamati

Coxbazar

Rangpur

Chittagong

Rajshahi

Bogra

Dinajpur

Tangail

Feni

Mymensingh

-2

Dhaka

-1

Faridpur

0

Comilla

1 Chandpur

mm/day

3

BMD stations

Fig. 5.1.2. Deviation of rainfall (mm/day) for 27 observed BMD station from 50 years average (6.049 mm/d) of all over Bangladesh. 5.2 Analysis procedure There are several indices that measure how much precipitation for a given period of time has deviated from historically established norms. Although none of the major indices is inherently superior to the rest in all circumstances, some indices are better suited than others for certain uses. For example, the Palmer Drought Severity Index (PDSI) has been widely used by the U.S. Department of Agriculture to determine when to grant emergency drought assistance, but the PDSI is better when working with large areas of uniform topography. This procedure may not much effective for the small area like the country Bangladesh. The National Drought Mitigation Center of USA is using the Standardized Precipitation Index (SPI) to monitor moisture supply conditions. Distinguishing traits of this index are that it identifies emerging droughts months sooner than the Palmer Index and that it is computed on various time scales. In order to investigate the spatial and temporal extents and severity of drought occurrence in the study area, Standardized Precipitation Index (Mckee et al., 1993) is used. SPI is a widely used drought index based on the of precipitation on multiple time scales. It has been demonstrated by several researches (McKee et al., 1995; Guttman, 1998, 1999; Hayes et al., 1999) that the SPI is a good tool for detecting and monitoring the drought events. This study computes SPI during 1961-1990 from the observation (Rain-gauge) monthly rainfall data and RegCM model simulated rainfall data. The Standardized Precipitation Index (SPI) based on the probability of precipitation, computed the characteristics of drought in Bangladesh for multiple time scales; 1, 3, 6, 9, 12, 24 months time steps. These time scales reflect the impact of drought on the availability of the different water resources. Particular drought year is detected from the time series analysis. The range of SPI in different drought categories is shown in Table 5.2.1. The negative values of SPI are considered as dry and positive values for wet periods. The SPI is calculated for the same observed station locations and duration using RegCM output rainfall data. Statistical scores are also calculated to obtain the efficiency-index of RegCM in drought diagnosis. SPI detected drought cases in the particular region of Bangladesh for both observed and model data are also verified using historical record of drought from BBS and EM-DAT data archive.


Table 5.2.1. Drought categories defined for SPI values. SPI value

Drought category

0 to -0.99 -1.00 to -1.49 -1.50 to -1.99 -2.00 and less

Near normal or mild drought Moderate drought Severe drought Extreme drought

Historical records of Drought in Bangladesh 6.1 Bangladesh Bureau of Statistics Archive Record of drought event of Bangladesh obtained from â&#x20AC;&#x2DC;Agricultural Statistics Year book of Bangladeshâ&#x20AC;&#x2122; published by Bangladesh Bureau of Statistics (BBS) from 1975 to 1990. Drought affected area, time period, length of drought and damage has given below by table in detail. According to BBS data, in the year 1984, 1985 and 1986 most of the places of the country experienced moderate drought at pre-monsoon season (February to April). These prolong drought hindered crop production of all over the country. There was severe drought occurred in 1980, 1981 and 1982 in North, Central and western part of Bangladesh. In 1979 drought, 43 per cent of area and 48.93 per cent of population affected in Bangladesh. Table 6.1.1 Drought affected area, drought spell obtained from BBS for year 1975-1976. Year

Affected Area

1975-76

Dinajpur Rangpur Rajshahi Bogra Kishoregonj Mymensingh Tangail Dhaka Faridpur Comilla Pabna Kushtia Khulna

Drought time Period Dec-May Dec-April Nov-May Oct-April Dec-April Dec-April Dec-April Oct -Feb Oct-April March-May Dec-April March-May March-May

Length (Month) 6 5 7 7 5 5 5 5 7 3 5 3 3

Table 6.1.2. Drought affected area, drought spell obtained from BBS for year1976-77. Year

Affected Area

1976-77

Dinajpur Rangpur Rajshahi Kishoregonj Mymensingh Tangail Dhaka

Drought time Period Oct-May Oct-May Nov-Feb March-May March-May Nov-Feb Sep -Feb

Length (Month) 8 8 4 3 3 4 6


Faridpur Chittagong Jessore Pabna Kushtia Khulna

Oct -May March-May March-May March-May Oct May March-May

8 3 3 3 8 3

Table 6.1.3. Drought affected area, drought spell obtained from BBS for year 1980-1981. Year

Affected Area

1980-81

Dinajpur Rangpur Rajshahi Bogra Kishoregonj Dhaka Noakhali Jessore Khulna

Drought time Period May-August mid Oct-Dec mid Oct- Dec mid Oct-Jan mid Oct-Dec Nov-Dec Dec- Jan May -August Nov to Dec Nov to Jan

Length (Month) 4 3 3 4 3 2 2 4, 2 3

Table 6.1.4. Drought affected area, drought spell obtained from BBS for year 1981-1982. Year

Affected Area

198182

Dinajpur Rangpur Rajshahi Affected Area Bogra Kishoregonj Mymensingh Tangail Jamalpur Dhaka Chittagong Sylhet

Drought Duration Jan to June Jan to April, Nov to January Dec to July, September, Oct to Dec Drought Duration Jan to April April –July Dec – July, Oct – Feb Dec to July, Nov to January Dec to April, Nov to January April to July

Length (Month) 7 5, 3

Crop damage due to drought in 1982

7, 1, 3

In September, 40% area of Aman paddy, 62% area of jute. Crop damage due to drought in 1982

Dec to July Nov to Feb April to July

8, 4

Length (Month) 5 3 8, 5 8, 3 5, 3 4

4, 4

51% area of Aman paddy damage/ Loss of production is 1803 tons.


Nov to Feb Dec to June Feb to July Oct to January Oct to Dec

Noakhali Jessore Pabna

7 6, 4 3

Kushtia

Oct to March September

6, 1

patuakhali

Jan to April

5

17% area of Aman paddy, 41% area of Aus paddy damaged/ Loss of production is 28136 tons. 56% area of Aman paddy, 16.84% area of Sugarcane, 18% area of Jute.

Table 6.1.5. Drought affected area, drought spell obtained from BBS for year 1983-1984. Year

Affected Area Dinajpur

1983-84 The month March and April, 1984 all over the Rangpur country was dry. Rajshahi Bogra Kishoregonj Mymensingh Tangail Jamalpur Dhaka Affected Area Chittagong Sylhet Noakhali Jessore Pabna Kushtia Khulna

Drought time Length (Month) Period March-July, 5, 3 Oct-Dec Oct-Dec, 3 March-July 5 March-July 5 Oct-Dec 3 March-July 4 Jan-March 3 March-July 5 Jan-March 3 Drought time Length (Month) Period May-July 3 Oct-Dec, 3, 5 March-July Oct-Dec 3 March-July 5 Jan-July 8 Jan- August, 9, 3 Oct-Dec April-June, 3, 3 Oct-Dec,

Table 6.1.6. Drought affected area, drought spell obtained from BBS for year 1984-1985. Year

Affected Area

1984-85

Dinajpur Rangpur

Drought Period Dec-April Dec-April, Sep-Nov

time Length (Month) 5 5, 3


Rajshahi Bogra Tangail Dhaka Faridpur Chittagong Sylhet Jessore Pabna Kushtia Khulna

Sep-Dec Aug-Sep, Nov-Jan Aug-Sep, Nov-Jan Dec-April Oct-Dec Oct- April Early season of tea, Dec-April Aug-Sep Oct-Dec Oct-April Sep- Dec

4 2, 3 3, 3 5 3 7 5 2 3 7 4

Table 6.1.7. Drought affected area, drought spell obtained from BBS for year 1986-1986. Year

Affected Area

Dinajpur 1985-86 November to January Rangpur all over the country. Rajshahi Bogra Kishoregonj Mymensingh Tangail Jamalpur Dhaka Comilla Faridpur Sylhet Noakhali Mymensingh Barisal Jessore Pabna Kushtia Khulna Patuakhali

Drought time Period Nov- Feb, Oct-April Sep- Feb March -June, Nov-Feb Sep - Dec Dec-Feb Sep-Dec Aug-Sep Sep-Nov March - Sep Oct -Feb Nov-Dec Oct- Dec March-July Nov-Dec Oct-Dec Oct-Feb, Aug-Sep Nov- Feb April-June, Oct- April Oct- Feb, Sep-Nov Dec-Feb

Length (Month) 4, 7 6 4, 4 4 3 4 2 3 7 5 2 3 5 2 3 5, 2 4 3, 7 5, 3 3

Table 6.1.8. Drought affected area, drought spell from BBS for year 1987-1988. Year

Affected Area

Drought Period

time Length (Month)


1987-88

Dinajpur Rajshahi Jamalpur Dhaka Comilla khagrachari Sylhet Pabna Khulna Barisal

April-August, OctDec March -August, Oct-Dec Feb - April March -August March-May Feb- April Oct- Dec Feb-April Feb-May March-May

5, 3 6, 3 3 6 3 3 3 3 4 3

Table 6.1.9. Summary of drought event in Bangladesh during 1975-1990 obtained from BBS record. Year 1975 1976 1980

Drought Affected Area of Bangladesh Length (Month) North, Central, West 3 to 6 North, West 3 to 6 North, West 2 to 4

1981 1982 1983 1984 1985 1986 1987 1988

North, Central, West, East North, Central, West, East North, Central, West, East North, Central, West, East North, Central, West North, Central, West, North, West North, West

of

drought

3 to 8 3 to 8 3 to 6 3 to 6 2 to 6 2 to 6 3 to 6 3 to 6

6.2 International Disaster database (EM-DAT) archive Since 1988 the WHO Collaborating Centre for Research on the Epidemiology of Disasters (CRED) has been maintaining an Emergency Events Database EM-DAT. EM-DAT was created with the initial support of the WHO and the Belgian Government. Record of drought events of Bangladesh during 1971-1990 obtained from EM-DAT database. Source:"EM-DAT: The OFDA/CRED International Disaster Database, www.emdat.be UniversitĂŠ catholique de Louvain - Brussels - Belgium". Chronology of major drought events and its impact in Bangladesh by year details: 1973 One of the severest in the present century and was responsible for the 1974 famine in northern Bangladesh. 1975 This drought affected 47% of the entire country and caused sufferings to about 53% of the total population.


1978-79 Severe drought causing widespread damage to crops. Reduced rice production by about 2 million tons and directly affected about 42% of the cultivated land and 44% of the population. It was one of the severest in recent times. 1981 Severe drought adversely affected crop production. 1982 Caused a total loss of rice production amounting to about 53,000 tons. In the same year flood damaged about 36,000 tons of rice. 1989 Most of the rivers in North West Bangladesh dried up and several districts, such as Naogaon, Nawabganj, Nilpahamari and Thakurgaon; dust syndrome occurred for a prolonged period due to drying up the topsoil. Table 6.2.1. Summary of drought event in Bangladesh during 1970-1990 obtained from EMDAT database. Year

Length (Month) 3 to 6 month 10

1982-83

Location of affected area of Bangladesh north Northwest 47% area of the entire country Chittagong, Rangpur, Mymensingh All over country

1989

Northwest

1973 1974 1975 1978-79

Tot. Affected people

6

53% of total population 2000

7

20000000

4

5000000

Diagnosis of drought using SPI 7.1 Observed data analysis The SPI calculation has done for all 27 observational sites over Bangladesh with intervals of three hours from 1961 to 1990. First, 5 well separated stations named Rangpur, Dhaka, Khulna, Coxâ&#x20AC;&#x2122;s Bazar, and Sylhet has been considered for temporal analysis. To observe the drought condition of whole Country, 27 BMD stations separated in four regions named (a) Central Region, (b) North Region, (c) South-West Region, and (d) Eastern Region. 7.1.1 Observed Station Data The SPI is calculated in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) with Observed data for the station Rangpur is shown in Figs. 7.1.1 (a) and (b). It has seen that Rangpur experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.1.1. (a).


4 Obs_Rangpur_short

3 2 obs SPI

1 0 -1 -2 -3 1989

1987

M6 1985

1983

1981

M3

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1 -4

Fig. 7.1.1. (a). SPI calculated using Observed data in different short month-length at Rangpur. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.1.1. (b). From these two Figures it is clear that it is difficult to differentiate among the short-length months i.e. among M1, M3, M6. And also it is difficult to differentiate for all long-length months i.e. among M9, M12, M24. This is because the patterns are almost same with a small shift with increase of month. 4 Obs_Rangpur_long 3

obs SPI

2 1 0 -1 -2

1989

1987

M24 1985

1981

M12

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1983

-3

Fig.7.1.1. (b). SPI calculated using Observed data in different long month-length at Rangpur. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years has been detected by the analysis. It is seen from the figs. 7.1.1.(a) and (b) that there were moderate to severe droughts (SPI ≤ -1) in the year of 1962, 1963, 1966, 1967, 1970, 1974, 1976, 1980, 1982, 1989. And Extreme drought (SPI ≤ -2) years were detected in the year 1968, 1972, 1973, 1976, 1982 in Rangpur station according to Observed SPI calculation. For the station Dhaka, the SPI is calculated in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) using Observed data as shown in Fig. 7.1.2 (a) and (b). It has seen that Dhaka also experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.1.2. (a). 4

Obs_Dhaka_short

3

1 0 -1 -2 -3 1989

1987

M6 1985

1983

M3 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1

-4

1961

obs SPI

2


Fig.7.1.2. (a). SPI calculated using Observed data in different short month-length at Dhaka Station. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.1.2. (b). from these two Figure it can be said that it is difficult to differentiate among the short-lengths (among M1, M3, M6) and also among the long-lengths (among M9, M12, M24) because the patterns are almost same with a small shift. 4 Obs_Dhaka_long

3 2 obs SPI

1 0 -1 -2

1989

M24

1987

1983

M12

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9

1961

-4

1985

-3

Fig.7.1.2.(b). SPI calculated using Observed data in different long month-length at Station Dhaka. Magnitudes also vary in different month-length. Extreme drought (SPI ≤-2) in some years is detected by the analysis. It is seen from the Fig 7.1.2.(a) and (b) that there were moderate to severe droughts (SPI ≤-1) in the year of 1960, 1962, 1964, 1966, 1969, 1973, 1978, 1980, 1982, 1987, 1990. And Extreme drought (SPI ≤-2) year were detected in the year 1962, 1970, 1973, 1979, 1990 in Dhaka station according to Observed SPI calculation. From the observation it is clear that the calculated patterns of Dkaka station are quite similar with Rangpur but the magnitudes and detection of drought months/years are different between 2 stations. Similarly, drought detected by SPI using Observed data for Khulna, Cox’s Bazar and Sylhet are shown in both short month-length and long month-length in Fig.7.1.3, Fig.7.1.4 and Fig.7.1.5 respectively. 4 Khulna_short

3

obs SPI

2 1 0 -1 -2

1989

1987

M6 1985

M3 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1 1961

-4

1983

-3

Fig. 7.1.3. (a). SPI calculated using Observed data in different short month-length at Station Khulna.


4

Khulna_long

3

obs SPI

2 1 0 -1 -2

1989

M24 1987

1983

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1985

-3

Fig. 7.1.3. (b). SPI calculated using Observed data in different long month-length at Station Khulna. In Khulna station according to Observed SPI calculation there were moderate to severe droughts (SPI ≤ -1) in the year of 1962, 1966, 1971, 1976, 1978, 1979, 1987, 1988 and 1990. And Extreme drought (SPI ≤ -2) year were detected in the year 1969, 1972, 1978, 1982, 1989. 4 Obs_Coxsbazar_short

3

obs SPI

2 1 0 -1 -2

M6 1989

1985

M3 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1 1961

-4

1987

-3

Fig.7.1.4. (a). SPI calculated using Observed data in different Short month-length at Station Cox’s Bazar. 4 Obs_Coxsbazar_long

3

obs SPI

2 1 0 -1 -2 -3 1989

M24 1987

1985

M12 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M9 -4

Fig. 7.1.4. (b). SPI calculated using Observed data in different long month-length at Station Cox’s Bazar. In Cox’s Bazar station according to Observed SPI calculation there were moderate to severe droughts (SPI ≤ -1) in the year of 1966, 1970, 1972, 1978, 1979, 1980, 1984, 1986 and 1990. And Extreme drought (SPI ≤ -2) year were detected in the year 1965, 1969, 1973, 1977, 1989.


4 Obs_Sylhet_short

3

obs SPI

2 1 0 -1 -2 M6 1987

1983

1989

M3 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1 1961

-4

1985

-3

Fig. 7.1.5. (a). SPI calculated using Observed data in different short month-length at Station Sylhet. 4 Obs_Sylhet_long

3

obs SPI

2 1 0 -1 -2 -3 1989

1987

M24 1985

1983

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M9 -4

Fig. 7.1.5. (b). SPI calculated using Observed data in different long month-length at Station Sylhet. In the Sylhet station according to Observed SPI calculation there were moderate to severe droughts (SPI â&#x2030;¤-1) in the year of 1962, 1970, 1972, 1973, 1978, 1979, 1981, 1982, 1983 and 1988, 1989. And Extreme drought (SPI â&#x2030;¤-2) years were detected in the year 1962, 1979, 1981, 1986. Overall, detection of drought in different stations, different month lengths are not exactly same. This may happen due to many reasons such as chosen stations are well separated, though Bangladesh is small country, there is a difference in geographical and environmental from region to region. So, drought detection may vary from station to station. Moreover, the magnitude of drought also varies between different month-lengths. It observed that short month length i.e. M1, M3 can detect more peak then the long month-length i.e. M12, M24 etc. Due to station point information used in calculation of SPI, the exact same drought years are not possible to detect by the data values. Therefore, regional average of neighbour stations may provides better performance of using the SPI for drought analysis. Also the statistical analysis is needed for understanding that how much reliability is there to use RegCM outputs for drought diagnosis. 7.1.2 Observed Regional Average Data To observe the drought condition of whole Country, 27 station data utilized into four regions for detail study which are describe below. (a) Central Region


Data from the 6 stations of Bangladesh Meteorological Department (BMD) named Chandpur, Comilla, Dhaka, Faridpur, Mymensingh and Tangail has been consider for Central region. The average value of Observed Regional data used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the Central Region are shown in Fig. 7.1.6 (a) and (b). It is seen that this Region experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar in Fig 7.1.6. (a). 4 obs_central_short

3

obs SPI

2 1 0 -1 -2

1989

1987

M6

1985

1983

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

M3

1981

M1

-3

Fig. 7.1.6. (a). SPI calculated using Observed data in different short month-length at Central Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.1.6. (b). from these two Figuress it has observed that it is difficult to differentiate among the short-length months (among M1, M3, M6) and also among the long-length months (among M9, M12, M24) because the patterns are almost same with a small shift with increase of month. In the short time period SPI can detect more value then the long time period. 4 obs_Central_long

3

obs SPI

2 1 0 -1 -2

1990

1988

1986

M24

1984

M12

1982

1978

1976

1974

1972

1970

1968

1966

1964

1962

1960

-4

1980

M9

-3

Fig.7.1.6. (b). SPI calculated using Observed data in different long month-length at Central Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.1.6.(a) and (b) that there were moderate to severe droughts (SPI ≤ -1) in the year of 1964, 1970, 1973, 1974, 1970, 1979, 1982, 1986, 1989. And Extreme drought (SPI ≤ -2) year were detected in the year 1973 and 1979 in Central Region according to Observed SPI calculation. (b) North Region The average value of Observed data for 4 stations of Bangladesh Meteorological Department (BMD) named Bogra, Dinajpur, Rajshai, and Rangpur are consider for North region of Bangladesh. This data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the North Region are shown in Fig. 7.1.7 (a) and (b). It is seen that this Region experienced moderate to severe droughts for


many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.1.7. (a). 4 obs_north_short

obs SPI

3 2 1 0

-1 -2

1989

1987

M1 1985

M6 1983

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

1981

M3

-3

Fig. 7.1.7. (a). SPI calculated using Observed data in different long month-length at North Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.1.7. (b). from these two Figuress we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4

obs_north_long

3

obs SPI

2 1 0 -1

1991

1989

M24

1987

1985

M12

1983

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-3

1981

M9

-2

Fig. 7.1.7. (b). SPI calculated using Observed data in different long month-length at North Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.1.7.(a) and (b) that there were moderate to severe droughts (SPI ≤ -1) in the year of 1961, 1968, 1971, 1974, 1976, 1979, 1982, 1989. And Extreme drought (SPI ≤ -2) year were detected in the year 1973 and 1976 in North Region according to Observed SPI calculation. (c) South-West Region The average value of Observed data for 7 stations of Bangladesh Meteorological Department (BMD) named Barisal, Hatiya, Jessore, Khepupara, Khulna, M. Court, and Patuakhali are consider for the South-West Region of Bangladesh. This data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the South-West Region are shown in Fig. 7.1.8 (a) and (b). It is seen that this Region experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.1.8. (a).


4 obs_South-West_short

3

obs SPI

2 1 0

-1 -2 M3

M6

M1

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

-4

1961

-3

Fig.7.1.8. (a). SPI calculated using Observed data in different Short month-length at SouthWest Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.1.8. (b). from these two Figuress we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4 obs_South-West_long

3

obs SPI

2 1 0 -1 -2 M9

M12

M24

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

-4

1961

-3

Fig. 7.1.8. (b). SPI calculated using Observed data in different long month-length at SouthWest Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤-2) in some years is detected by the analysis. It is seen from the Figs 7.1.8.(a) and (b) that there were moderate to severe droughts (SPI ≤-1) in the year of 1962, 1963, 1966, 1967, 1974, 1981, 1982, 1983, 1984, 1989. And Extreme drought (SPI ≤-2) years were detected in the year 1972, 1973 in South-West Region according to Observed SPI calculation. (d) Eastern Region Data from the 10 stations of Bangladesh Meteorological Department (BMD) named Chittagong, Cox’s Bazar, Feni, Kutubdia, Rangamati, Sandwip, Sitakunda, Srimangal, Shlhet, and Teknaf has been consider for the Eastern Region. The average value of Observed Regional data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the Eastern Region are shown in Fig. 7.1.9 (a) and (b). It is seen that this Region experienced moderate droughts for some years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.1.9. (a).


4 obs_south-East_short

3

obs SPI

2 1 0 -1 -2

1989

1987

M6

1985

1983

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

M3

1981

M1

-3

Fig.7.1.9. (a). SPI calculated using Observed data in different Short month-length at Eastern Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are little different from short month length in Fig 7.1.9. (b). It may happen because in this region experiences heavy rainfall in the Monsoon season, [Islam M. N. and Uyedo H., 2007]. 4 obs_south-East_long

obs SPI

3 2 1 0 -1 -2

1989

1987

M24 1985

1983

M12 1981

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

1979

M9

-3

Fig. 7.1.9. (b). SPI calculated using Observed data in different long month-length at Eastern Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.1.9.(a) and (b) that there were moderate droughts (SPI ≤ -1) in the year of 1963, 1971, 1972, 1982, 1984, 1990. And Extreme drought (SPI ≤ -2) year were detected in the year 1973 in a short month length in Eastern Region according to Observed SPI calculation. 7.2 Model data analysis The SPI is also calculated for the same observed station locations and the duration using the Regional Climate Model (RegCM) output rainfall data. First consider 5 well separated stations named Rangpur, Dhaka, Khulna, Cox’s Bazar, and Sylhet for temporal analysis. To analyze the drought condition of whole Country, 27 BMD station locations model output data separated into four regions named (a) Central Region, (b) North Region, (c) South-West Region, (d) Eastern Region. 7.2.1 Model Station Data The SPI calculated in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) with Model data for the station Rangpur is shown in Fig. 7.2.1 (a) and (b). It is seen that Rangpur experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.2.1. (a).


4 Model_Rangpur_short

3

Model SPI

2 1 0 -1 -2

1989

1987

M6

1983

M3

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1

1961

-4

1985

-3

Fig. 7.2.1. (a). SPI calculated using Model data in different short month-length at Rangpur. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.2.1. (b). from these two Figuress we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4 model_Rangpur_long

model SPI

3 2 1 0 -1 -2

1989

M24

1987

1983

M12

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9

1961

-4

1985

-3

Fig.7.2.1. (b). SPI calculated using Model data in different long month-length at Rangpur. Magnitudes also vary in different month-length. Extreme drought (SPI â&#x2030;¤ -2) in some years is detected by the analysis. It is seen from the Figs 7.2.1.(a) and (b) that there were moderate to severe droughts (SPI â&#x2030;¤ -1) in the years of 1962, 1963, 1966, 1967, 1970, 1972, 1974, 1975, 1976, 1980, 1982, 1985, 1989, 1990. And Extreme drought (SPI â&#x2030;¤ -2) years were detected in the year 1969, 1971, 1979, 1986 in Rangpur station according to Model data SPI calculation. For the station Dhaka, The SPI is calculated in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) using Model data is shown in Fig. 7.2.2 (a) and (b). It is seen that Dhaka also experienced moderate to severe droughts for many years during 1961-1990. Patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.2.2. (a). 4

Model_dhaka_short

3

Model SPI

2 1 0 -1 -2

1989

M6

1987

M3

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1

1961

-4

1985

-3

Fig. 7.2.2. (a). SPI calculated using Model data in different short month-length at Dhaka Station.


The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.2.2. (b). from these two Figuress we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4 model_dhaka_long

model SPI

3 2 1 0 -1 -2

1989

1987

M24

1983

M12

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9

1961

-4

1985

-3

Fig. 7.2.2. (b). SPI calculated using Model data in different long month-length at Station Dhaka. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.2.2.(a) and (b) that there were moderate to severe droughts (SPI ≤ -1) in the year of 1963, 1966, 1969, 1970, 1971, 1974, 1978, 1980, 1982, 1987, 1990. And Extreme drought (SPI ≤ -2) years were detected in the year 1967, 1970 and 1979 in Dhaka station according to Model data SPI calculation. Similarly, drought detected by SPI using Model data for Khulna, Cox’s Bazar and Sylhet are shown in both short month-length and long month-length in Fig.7.2.3, Fig.7.2.4 and Fig.7.2.5 respectively. 4 Model_Khulna_short

3

Model SPI

2 1 0 -1 -2 M6

1989

1987

1983

M3

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1

1961

-4

1985

-3

Fig. 7.2.3. (a). SPI calculated using Model data in different Short month-length at Station Khulna. 4

model_Khulna_long

3

model SPI

2 1 0 -1 -2

1989

1987

M24 1985

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1983

-3

Fig. 7.2.3. (b). SPI calculated using Model data in different long month-length at Station Khulna. In Khulna station according to Model SPI calculation there were moderate to severe droughts (SPI ≤-1) in the year of 1961, 1966, 1969, 1970, 1971, 1974, 1975, 1987, 1979, 1980 and 1990. And Extreme drought (SPI ≤-2) years were detected in the year 1974, 1979 and 1987.


4

Model_Coxsbazar_short

3

Model SPI

2 1 0 -1 -2 M6 1989

1987

1983

M3 1981

1979

1977

1975

1973

1971

1969

1967

1965

1961

1963

M1

-4

1985

-3

Fig.7.2.4. (a). SPI calculated using Model data in different Short month-length at Station Cox’s Bazar. 4

model_Coxsbazar_long

model SPI

3 2 1 0 -1 -2

1989

M24 1987

1983

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1985

-3

Fig. 7.2.4. (b). SPI calculated using Model data in different long month-length at Station Cox’s Bazar. In Cox’s Bazar station according to Model SPI calculation there were moderate to severe droughts (SPI ≤-1) in the year of 1965, 1966, 1970, 1972, 1975, 1978, 1979, 1982, 1984, 1988 and 1990. And Extreme drought (SPI ≤-2) years were detected in the year 1963, 1966, 1989 and 1990. Fig. 7.2.5. (a). SPI calculated using Model data in different Short month-length at Station Sylhet. 4

model_Sylhet_long

3

model SPI

2 1 0 -1 -2

1989

1987

M24 1985

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1983

-3

Fig. 7.2.5. (b). SPI calculated using Model data in different long month-length at Station Sylhet. In the Sylhet station according to Model SPI calculation there were moderate to severe droughts (SPI ≤-1) in the year of 1963, 1967, 1969, 1970, 1971, 1975, 1979, 1982, 1985 and 1990. And no Extreme drought (SPI ≤-2) years were detected this region. Overall, detection of drought in different stations, different month lengths are not exactly same. This may happen due to many reasons such as chosen stations are well separated, though Bangladesh is small country, there is a difference in geographical and environmental from region to region. So, drought detection may vary from station to station.


Moreover, the magnitude of drought also varies between different month-lengths. It is seen that short month length i.e. M1, M3 can detect more peak than the long month-length i.e. M12, M24 etc. Due to station point information used in calculation of SPI, the exact same drought years are not possible to detect by the data values. Therefore, regional average of neighbour stations may provide better performance of using the SPI for drought analysis. Also the statistical analysis is needed for understanding that how much reliability is there to use RegCM outputs for drought diagnosis. 7.2.2 Regional Average of Model Data (a) Central Region Model Data from the 6 station locations of Bangladesh Meteorological Department (BMD) named Chandpur, Comilla, Dhaka, Faridpur, Mymensingh and Tangail has been consider for Central region of Bangladesh. The average value of Model output Regional data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the Central Region are shown in Fig. 7.2.6 (a) and (b). It is seen that this Region experienced moderate to severe droughts for many years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar in Fig 7.2.6. (a). 4

Model_central_short

3

Model SPI

2 1 0 -1 -2

1989

M6 1987

1983

M3 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M1 1961

-4

1985

-3

Fig. 7.2.6. (a). SPI calculated using Model output data in different short month-length at Central Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.2.6. (b). from these two Figuress we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift with increase of month. In the short time period SPI can detect more peak value then the long month-length period. 4

model_Central_long

3

1 0 -1 -2

1989

1987

M24 1985

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

M9 1963

-4

1983

-3 1961

model SPI

2


Fig. 7.2.6. (b). SPI calculated using Model output data in different long month-length at Central Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.2.6.(a) and (b) that there were moderate to severe droughts (SPI ≤ -1) in the year of 1963, 1966, 1970, 1971, 1974, 1978, 1979, 1982, 1988, 1989. And Extreme drought (SPI ≤ -2) year were detected in the year 1967 and 1979 in Central Region according to Model output SPI calculation. (b) North Region The average value of Model output data for 4 station locations of Bangladesh Meteorological Department (BMD) named Bogra, Dinajpur, Rajshai, and Rangpur has been consider for North region of Bangladesh. This data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the North Region are shown in Fig. 7.2.7 (a) and (b). It is seen that this Region experienced moderate to severe droughts for many years during 1961-1990. Patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.2.7. (a). 3 model_north_short 2

model SPI

1 0 -1 -2

1989

M1 1987

1983

M6 1981

1979

1977

1975

1973

1971

1969

1967

1963

1965

M3 1961

-4

1985

-3

Fig. 7.2.7. (a). SPI calculated using Model output data in different long month-length at North Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.2.7. (b). from these two Figures we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4

model_North_long

3

model SPI

2 1 0 -1 -2

1989

1987

M24 1985

M12 1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9 1961

-4

1983

-3

Fig. 7.2.7. (b). SPI calculated using Model output data in different long month-length at North Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤ -2) in some years is detected by the analysis. It is seen from the Figs 7.2.7.(a) and (b) that there were moderate to


severe droughts (SPI ≤ -1) in the year of 1961, 1967, 1970, 1974, 1975, 1979, 1982, 1986, 1990. And Extreme drought (SPI ≤ -2) years were detected in the year 1963, 1975 and 1979 in North Region according to Model output data SPI calculation. (c) South-West Region The average value of Model output data for 7 station locations of Bangladesh Meteorological Department (BMD) named Barisal, Hatiya, Jessore, Khepupara, Khulna, Mcourt, and Patuakhali are consider for the South-West Region of Bangladesh. This data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the South-West Region are shown in Fig. 7.2.8 (a) and (b). It is seen that this Region experienced moderate to severe droughts for many years during 19611990. Patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.2.8. (a). Fig. 7.2.8. (a). SPI calculated using Model output data in different Short month-length at South-West Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are also similar with a small shift with increase of month in Fig 7.2.8. (b). from these two Figures we can say that it is difficult to differentiate among the short-length (among M1, M3, M6) and also among the long-length (among M9, M12, M24) because the patterns are almost same with a small shift. 4 model_south-West_long

model SPI

3 2 1 0

-1 -2

1989

1987

M24

1985

M12

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M9

1961

-4

1983

-3

Fig. 7.2.8. (b). SPI calculated using Model output data in different long month-length at South-West Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤-2) in some years is detected by the analysis. It is seen from the Figs 7.2.8.(a) and (b) that there were moderate to severe droughts (SPI ≤-1) in the year of 1962, 1963, 1966, 1967, 1974, 1980, 1986, 1987, 1989 and 1990. And Extreme drought (SPI ≤-2) years were detected in the year 1966 and 1979 in South-West Region according to Model output data SPI calculation. (d) Eastern Region Model Data from the 10 station locations of Bangladesh Meteorological Department (BMD) named Chittagong, Cox’s Bazar, Feni, Kutubdia, Rangamati, Sandwip, Sitakunda, Srimangal, Shlhet, and Teknaf has been consider for the Eastern Region. The average value of Model output Regional data is used for the SPI calculation in month 1 (M1), month 3 (M3), month 6 (M6), month 9 (M9), month 12 (M12) and month 24 (M24) for the Eastern Region are shown in Fig. 7.2.9 (a) and (b). It is seen that this Region experienced moderate droughts for some years during 1961-1990. patterns for month 1 (M1), month 3 (M3) and month 6 (M6) are almost similar with a small shift with increase of month in Fig 7.2.9. (a).


4 model_South-East_short

3

model SPI

2 1 0 -1 -2

1989

1987

M6

1985

1983

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

M3

1981

M1

-3

Fig. 7.2.9. (a). SPI calculated using Model output data in different Short month-length at Eastern Region. The patterns for month 9 (M9), month 12 (M12) and month 24 (M24) are little different from short month length in Fig 7.2.9. (b). It may happen because this region experiences heavy rainfall in the Monsoon season [Islam M. N. and Uyedo H. 2007]. 4 model_South-East_long

3

model SPI

2 1 0 -1 -2

1989

1987

M24

1985

1983

M12

1981

1977

1975

1973

1971

1969

1967

1965

1963

1961

-4

1979

M9

-3

Fig. 7.2.9. (b). SPI calculated using Model output data in different long month-length at Eastern Region. Magnitudes also vary in different month-length. Extreme drought (SPI ≤-2) in some years is detected by the analysis. It is seen from the Figs 7.2.9.(a) and (b) that there were moderate droughts (SPI ≤-1) in the year of 1963, 1967, 1971, 1978, 1982 and 1990. And Extreme drought (SPI ≤-2) years were detected in the year 1963, 1966, 1979 and 1990 in a short month length in Eastern Region according to Model output data SPI calculation. Comparison of SPI in Bangladesh 8.1 Comparison of Observed and model SPI value The SPI calculated result for month 1 (M1) using data from Observed (Obs) and RegCM model output data at stations Rangpur, Dhaka, Khulna, Cox’s Bazar and Sylhet is shown in Fig 8.1.1. Model data is well detected drought in the year 1964, 1965, 1968, 1974, 1978, 1980, 1982 and 1989 with reference to Obs data at Rangpur. Similarly, 1962, 1964, 1966, 1969, 1978, 1980, 1982 and 1990 at Dhaka; 1962, 1966, 1971, 1978, 1979, 1987, 1988 at Khulna; 1966, 1968, 1972, 1979, 1980, and 1989 at Cox’s Bazar and 1970, 1973 and 1989 at Sylhet station. However, false detection of drought year by the Model output data with respect to Observed data also occurred in these five stations. In the year of 1971, 1973, 1989 the model data could not agree with the observed data in almost all stations.


-3

2

comparison M 1 Sylhet 1989

1987

comparison M1 Khulna

1989

1987

1989

1987

1985

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

SPI value 2

comparison M1 Dhaka

2

comparison M 1 Cox's Bazar

1989

1987

M1 Obs

1985

1983

1981

1979

1977

1975

M1 Obs

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

SPI value

M1 Obs

1985

M1 Obs

1983

1981

1979

1977

1975

M1 Obs

1983

1981

1979

1977

1975

1973

3

1973

1971

1969

1967

1965

1963

SPI value 3

1973

1971

1969

1967

1965

3

1971

1969

1967

1965

-3

1963

1961

-3

1961

SPI value -3

1963

SPI value -3

1961

3

comparison M1 Rangpur

(a)

1

0

-1

-2

M1 Model

2

b

1

0

-1

-2

M1 Model

2

c

1

0

-1

-2

M1 Model

d

1

0

-1

-2

M1 Model

3

e

1

0

-1

-2

M1 Model


Fig 8.1.1. The SPI calculated for month 1 (M1) using data from observation (OBS) and RegCM output at station (a) Rangpur, (b) Dhaka, (c) Khulna, (d) Coxâ&#x20AC;&#x2122;s Bazar and (e) Sylhet. Overall, it is seen that in some years drought is well detected by two types of data. On the other hand, in some years false detection of drought by RegCM model output indicate limitation of utilizing climate model outputs in drought projection. Important to note that, there are missing of Obs data in some station of some years during 1961-1990 period. Therefore, it can not simply disagree to the detection of drought with model data. Moreover, the historical records of drought are very important to investigate the possibility of model outputs in calculation of drought indices. Figs 8.1.2.,8.1.3 and Fig 8.1.4. shows the SPI calculated at stations Rangpur, Dhaka, Khulna, Coxâ&#x20AC;&#x2122;s Bazar and Sylhet using data from observation (Obs) and RegCM output for month 6 (M6), month 12 (M12) and month 24 (M24) respectively. 3

com parison M6 Rangpur

a

SPI value

2

1

0

-1

-2

3

comparison M6 Dhaka

b

2 SPI value

1989

1987

M6 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M6 Obs -3

1

0

-1

-2

3

com parison M6 Khulna

c

2 SPI value

1989

1987

M6 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M6 Obs -3

1

0

-1

-2

1989

1987

M6 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M6 Obs -3


3

comparison M6 Cox's Bazar

d

SPI value

2

1

0

-1

-2

3

com parison M6 Sylhet

1989

e

2 SPI value

1987

M6 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M6 Obs -3

1

0

-1

-2

1989

1987

M6 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M6 Obs -3

Fig 8.1.2. The SPI calculated for month 6 (M6) using data from observation (OBS) and RegCM output at station (a) Rangpur, (b) Dhaka, (c) Khulna, (d) Coxâ&#x20AC;&#x2122;s Bazar and (e) Sylhet. 3

comparison M12 Rangpur

a

2

SPI value

1 0

-1 -2

3

comparison M12 Dhaka

b

2 SPI value

1989

1987

M12 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1963

1965

M12 Obs 1961

-3

1

0

-1

-2

3

1989

1987

M12 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1961

1963

M12 Obs -3

comparison M12 Khulna

c

2

SPI value

1 0

-1 -2

1989

1987

M12 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12 Obs -3


3

d

comparison M12 Cox's Bazar 2

SPI value

1 0

-1 -2

3

1989

1987

M12 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12 Obs -3

e

comparison M12 Sylhet 2

SPI value

1 0

-1 -2

1989

1987

M12 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12 Obs -3

Fig 8.1.3. The SPI calculated for month 12 (M12) using data from observation (OBS) and RegCM output at station (a) Rangpur, (b) Dhaka, (c) Khulna, (d) Coxâ&#x20AC;&#x2122;s Bazar and (e) Sylhet. 3

comparison M24 Rangpur

a

SPI value

2

1

0

-1

-2

3

Comparison M24 Dhaka

b

2

SPI value

1989

M24 model 1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M24 Obs -3

1

0

-1

-2

3

com parison M24 Khulna

c

2

1

0

-1

-2

1989

M24 model

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

M24 Obs

-3

1961

SPI value

1989

1987

M24 model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M24 Obs -3


3

comparison M24 Cox's Bazar

d

2

SPI value

1

0

-1

3

comparison M24 Sylhet

1989

M24 model

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

-3

M24 Obs

1987

-2

e

2

SPI value

1

0

-1

-2

1989

1987

M24 model

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M24 Obs

-3

Fig 8.1.4. The SPI calculated for month 24 (M24) using data from observation (OBS) and RegCM output at station (a) Rangpur, (b) Dhaka, (c) Khulna, (d) Coxâ&#x20AC;&#x2122;s Bazar and (e) Sylhet. Similar to month 1 false detection of drought using model data are found in case of month 6, 12 and 24 references to Observed data. Above Fig indicates some discrepancies among the result of two data types. Important thing is to note that as the length of month increased from month 1 to month 24, discrepancies among the result of two data types also increased. In Some cases results obtained from model data just opposite to observed. Therefore, long-spell drought may not be well detected from point information of model data output. Use of regional average as the input of SPI calculation may bear the message of longâ&#x20AC;&#x201C;scale drought that is a regional phenomenon. Fig 8.1.5. and Fig 8.1.6. shows The SPI calculated at Central, North, South-West and Eastern region of Bangladesh using data from observation (OBS) and RegCM output for month 1 (M1) and month 12 (M12) respectively. 3 Central_M1

a

2

SPI value

1 0

-1 -2

1989

1987

1985

M1_Model

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1_Obs

-3


3 North_M1

b

2

SPI value

1 0 -1 -2

3

1989

1987

1985

M1_model 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1_Obs

-3

c

South-West_M1

SPI value

2 1 0

-1 -2

3

1989

1987

M1_model 1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1_Obs -3

d

South-East_M1 2

SPI value

1 0

-1 -2

1989

1987

1985

M1model 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1Obs -3

Fig 8.1.5. The SPI calculated for month 1 using data from observation (OBS) and RegCM output at (a) Central, (b) North, (c) South-West and (d) Eastern region of Bangladesh. 3 Central_M12

a

SPI value

2 1 0

-1 -2 1989

1987

1985

M12_Model

1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12_ObS

-3


3 North_M12

b

SPI value

2 1 0 -1 -2

1989

1987

1985

M12_Model 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12_Obs

-3

3 South-West_M12

c

2 SPI value

1 0 -1 -2 1989

1987

1985

M12_Model 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M12_Obs -3

3 South-East_M12

d

SPI value

2 1 0

-1 -2 1989

1987

1985

M12_model 1983

1981

1979

1977

1975

1973

1971

1969

1967

1965

1963

1961

M1Obs -3

Fig 8.1.6. The SPI calculated for month 12 using data from observation (OBS) and RegCM output at (a) Central, (b) North, (c) South-West and (d) Eastern region of Bangladesh. For month 1, detection of drought event at South-West region is much better than other regions and other month-length. From two types of drought spell it is clear that, short scale has detected more drought event compare to the long scale in all over Bangladesh. However, there are many mismatching events found during the analysis period. In 1966 there is a false detection of drought event by model data at Eastern and South-West region. At the year of 1973, in Central and South-West region results are nearly opposite among the model and observe data. Therefore, the historical records of drought event in Bangladesh region are very important for utilizing model output in drought detection of Bangladesh. 8.2 Comparison of observed and model SPI value with Historical records In this study, two different data sources were used for historical record of drought events in Bangladesh named; Bangladesh Bureau of Statistics (BBS) and International Disaster database (EM-DAT) which was discussed in chapter 6. In both two data sources there was no information available about drought before 1973 in Bangladesh. From the table 6.1.9. and 6.2.1. mention at chapter 6, it is seen that in the year 1975 and 1981 to 1984 drought affect almost the entire country. The SPI value for observe and model data matches with this results with a small deviation of time. Moreover, drought event in 1974,


1976, 1980, 1988 and 1989 detected by the model and observe SPI calculation also agree with historical records at North-West region. However, historical record confirms Drought event in North, West and Central region of Bangladesh at 1973, 1976, 1986 and 1987 where these events is not well detected by model data output of SPI calculation. On the other hand, drought in North, West and Central region of Bangladesh is well detected by model data in 1975 and 1985. Overall detection of drought using RegCM has been given below as a summary. Table 8.2.1. Drought detection by Model data and observed data with respect to both historical records in all over Bangladesh during 1973 to 1990. Drought Year Historical record 1973 1974 1975 1976 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989

fromDrought Affected BMD RF data Region of Bangladesh North Detected North, West Detected North, Central, West Detected North, West Detected North, Central Detected North, West Detected All over country Detected All over country Detected All over country Detected All over country Not detected North, Central, West Not detected North, Central, West, Detected North, West Detected North, West Detected North, West Detected

Model RF data Not detected Detected Detected Not detected Detected Detected Detected Detected Detected Detected Detected Not detected Not detected Detected Detected

Discussion It is clear from the analysis using station data from observed rainfall that sometimes single station information may not be sufficient for drought diagnosis. The result obtained from regional information of rainfall data is better in drought diagnosis with reference to station rainfall data. The historical record of drought confirms the drought situation detected by regional data. Table 9.1. Drought number at Central region of Bangladesh for month 1, Month 3 and Month 6 during 1961-1990 using SPI. Year Central region of Bangladesh M1 M3 M6 -1≥ -2≥ -1≥ -2≥ -1≥ -2≥ Ob Mod Ob Mod Ob Mod Ob Mode Ob Mod Ob Mo s el s el s el s l s el s del 196 1 2 0 0 4 1 0 0 3 0 0 0 1 196 3 0 0 0 2 0 0 0 6 0 0 0 2 196 3 3 1 0 7 3 2 0 11 7 0 0


3 196 4 196 5 196 6 196 7 196 8 196 9 197 0 197 1 197 2 197 3 197 4 197 5 197 6 197 7 197 8 197 9 198 0 198 1 198 2 198 3 198 4 198 5 198 6 198 7

0

1

0

0

0

1

0

0

2

0

0

0

2

3

0

0

2

5

0

0

2

4

0

0

3

3

0

1

3

3

0

0

3

2

0

0

0

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

2

0

0

0

0

0

0

0

2

3

0

0

1

3

0

0

0

3

0

0

0

3

0

0

0

2

0

0

1

1

0

0

4

0

0

0

2

0

3

0

2

1

3

0

3

1

1

0

3

0

2

0

4

0

0

0

3

1

0

0

1

0

0

0

1

0

2

0

4

4

0

1

2

8

0

0

1

3

0

1

3

1

0

0

4

2

0

0

10

3

0

0

1

3

0

0

1

1

0

0

0

1

0

0

2

2

0

1

3

5

0

0

1

3

0

0

5

3

0

0

4

3

3

2

3

3

2

0

3

4

0

2

3

4

0

1

6

9

1

0

1

1

0

1

1

0

0

0

0

1

0

0

1

2

0

0

2

2

0

0

0

1

0

0

1

2

0

0

0

0

0

0

1

0

0

0

0

1

0

0

0

1

0

0

0

0

0

0

1

1

0

0

0

1

0

0

0

0

0

0

2

3

0

0

1

5

0

0

0

5

0

0

1

0

0

0

0

1

0

0

1

4

0

0

0

2

0

0

0

0

0

0

0

0

0

0


198 8 198 9 199 0 total

0

2

0

0

0

2

0

0

0

2

0

0

3

2

0

1

0

0

0

1

0

0

0

0

0

2

0

0

1

4

0

0

0

3

0

0

52

56

2

7

49

57

10

4

58

56

8

1

Table 9.2. Drought number at North region of Bangladesh for month 1, Month 3 and Month 6 during 1961-1990 using SPI. Year North region of Bangladesh M1 M3 -1≥ -2≥ -1≥ Ob Mod Ob Mod Ob s el s el s 196 3 2 0 0 3 1 196 3 1 0 0 4 2 196 0 4 0 0 2 3 196 0 0 0 0 2 4 196 0 2 0 0 1 5 196 2 0 0 1 1 6 196 0 0 0 0 0 7 196 2 0 0 0 5 8 196 1 1 0 0 0 9 197 1 1 0 0 3 0 197 2 1 1 0 2 1 197 2 1 0 0 5 2 197 2 1 0 0 0 3 197 2 3 1 1 5 4 197 1 2 1 0 3 5 197 0 0 0 0 5 6 197 2 4 0 0 4

Mod el 1

-2≥ Ob Mod s el 0 0

M6 -1≥ Ob Mod s el 8 1

-2≥ Ob Mo s del 0 0

0

0

0

4

0

0

0

3

0

2

2

4

0

3

0

0

0

0

0

0

0

1

0

0

0

1

0

0

0

0

0

2

2

0

0

0

0

0

0

0

0

0

0

0

0

2

0

0

0

1

0

0

2

0

0

0

0

0

0

2

0

0

0

0

1

0

5

0

0

0

1

1

0

8

0

0

0

0

0

0

2

0

1

0

9

1

0

7

8

0

0

1

0

0

7

2

0

1

0

0

0

1

0

0

0

3

0

0

2

3

0

0


7 197 8 197 9 198 0 198 1 198 2 198 3 198 4 198 5 198 6 198 7 198 8 198 9 199 0 total

2

3

0

0

0

2

0

2

1

10

0

0

1

0

1

2

1

3

1

1

3

3

0

0

0

5

0

0

0

2

0

0

0

1

0

0

2

2

0

0

1

1

0

1

0

2

0

0

0

1

0

0

0

0

0

0

0

0

0

0

1

2

0

0

0

0

0

0

0

1

0

0

0

3

0

0

0

0

0

0

0

0

0

0

0

3

0

1

1

7

0

1

0

12

0

0

0

1

0

0

1

1

0

0

0

2

0

0

0

1

0

0

0

1

0

0

0

0

0

0

0

3

0

0

0

1

0

0

0

2

0

0

3

1

0

0

1

1

0

0

0

1

0

0

0

1

0

1

0

1

0

2

0

3

0

0

32

49

4

6

50

40

4

9

58

58

1

4

This indicates that regional phenomena like drought may not be possible using only point data. Therefore, regional average data is recommended for drought diagnosis and monitoring. In long-term plan drought projection has the priority for planners. Therefore, applicability of model data is invaluable even though uncertainties are there. So, statistical score is prepared to understand the overall performance of utilization of model outputs. The counts of drought month are calculated for Observed and RegCM data for Central region (Table 9.1), North region (Table 9.2), Eastern region (Table 9.3) and South-West region of Bangladesh (Table 9.4) for month 1, month 3 and month 6. Table 9.3. Drought number at Eastern region of Bangladesh for month 1, Month 3 and Month 6 during 1961-1990 using SPI. Year Eastern region of Bangladesh M1 M3 -1≥ -2≥ -1≥ Ob Mod Ob Mod Ob Mod s el s el s el 196 3 3 0 0 6 1 1

-2≥ Ob Mod s el 0 0

M6 -1≥ Ob Mod s el 5 0

-2≥ Ob Mo s del 0 0


196 2 196 3 196 4 196 5 196 6 196 7 196 8 196 9 197 0 197 1 197 2 197 3 197 4 197 5 197 6 197 7 197 8 197 9 198 0 198 1 198 2 198 3 198 4 198 5 198

3

0

0

0

5

0

0

0

7

0

0

0

2

3

0

1

5

4

1

1

8

3

0

0

0

0

0

0

0

0

0

0

2

0

0

0

1

3

0

0

1

4

0

0

1

2

0

0

0

3

0

0

0

9

0

0

0

11

0

1

4

1

0

0

3

2

0

0

3

3

0

0

1

0

0

0

3

0

0

0

0

0

0

0

2

2

0

0

2

1

0

0

2

1

0

0

0

2

0

0

0

3

0

0

2

3

0

0

2

0

0

0

4

2

0

0

2

6

1

0

1

1

1

0

1

0

2

0

1

0

0

0

2

1

0

0

4

0

0

0

2

0

1

0

1

5

0

0

0

7

0

0

0

3

0

0

3

1

0

0

2

3

0

0

1

5

0

0

0

3

0

0

0

3

0

0

0

2

0

0

0

2

0

1

0

2

0

0

0

1

0

0

2

4

0

0

2

4

0

1

0

2

0

0

0

3

0

1

2

2

0

2

2

7

0

0

1

2

0

0

0

0

0

0

0

0

0

0

1

3

0

0

1

1

0

1

0

4

0

0

1

2

0

0

0

0

0

0

1

0

0

0

0

1

0

0

0

0

0

0

0

0

0

0

1

1

0

0

1

0

0

0

0

0

0

0

0

2

0

0

0

0

0

0

0

1

0

0

1

1

0

0

1

2

0

0

0

2

0

0


6 198 7 198 8 198 9 199 0 total

1

1

0

1

0

3

0

0

0

0

0

0

0

1

0

1

0

0

0

0

0

0

0

0

2

0

0

1

1

0

0

2

0

2

0

0

0

4

0

0

1

4

0

1

0

4

0

2

35

55

1

6

45

57

3

8

39

62

2

3

On the other hand, for extreme drought case (-2≥ SPI) using SPI calculation model data may not count efficiently with reference to observed data over Bangladesh. One reason for it that there was missing of observed data of some stations for few months during 1961-1990. Over all, RegCM data over-counted for moderate and severe drought cases (-1≥ SPI) 33.12%, 5.7% and 6.00% for month 1, month 3 and month 6 respectively. For extreme drought count, model data over estimated 166.66% and 38.00% for M1 and M6 with reference to observe data. On the other hand, model data under estimated 28.57% for 6 month spell extreme drought case. Table 9.4. Drought number at South-West region of Bangladesh for month 1, Month 3 and Month 6 during 1961-1990 using SPI. Year South-West region of Bangladesh M1 M3 -1≥ -2≥ -1≥ Ob Mod Ob Mod Ob Mod s el s el s el 196 2 2 0 1 6 3 1 196 3 2 0 0 5 1 2 196 4 1 0 1 4 3 3 196 4 0 0 0 4 0 4 196 1 3 0 1 4 4 5 196 3 3 0 0 2 2 6 196 3 0 0 0 3 1 7 196 1 0 0 0 2 0 8 196 1 2 0 0 1 0 9 197 1 2 0 0 0 1 0

-2≥ Ob Mod s el 0 0

M6 -1≥ Ob Mod s el 8 1

-2≥ Ob Mo s del 0 0

0

1

11

4

0

0

0

0

7

5

0

0

0

0

7

0

0

0

0

1

7

3

0

0

0

0

5

3

0

0

0

0

3

2

0

0

1

0

0

0

0

0

0

0

2

0

0

0

0

0

1

0

0

0


197 1 197 2 197 3 197 4 197 5 197 6 197 7 197 8 197 9 198 0 198 1 198 2 198 3 198 4 198 5 198 6 198 7 198 8 198 9 199 0 total

2

1

0

0

4

0

0

0

2

0

0

0

3

2

2

0

5

2

3

0

6

0

0

0

0

2

0

0

1

0

0

0

0

0

3

0

1

5

0

0

0

6

0

1

0

7

0

0

2

2

0

0

3

1

0

0

1

3

0

0

0

5

0

0

0

3

0

1

0

2

0

1

1

5

0

0

0

0

0

0

0

3

0

0

0

2

0

0

0

2

0

2

0

2

0

0

2

3

0

1

1

2

0

2

1

5

0

1

1

1

0

0

0

0

0

0

0

0

0

0

2

3

0

0

2

2

0

0

0

0

0

0

2

2

0

0

1

0

0

0

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

2

0

0

0

0

0

0

0

0

0

0

0

0

2

0

0

0

3

0

0

0

0

0

0

1

1

0

0

0

3

0

0

0

7

0

0

1

1

0

0

0

2

0

0

0

1

0

0

0

0

0

0

0

2

0

0

0

1

0

0

1

3

0

1

0

1

0

0

0

1

0

0

0

2

0

0

1

6

0

0

0

5

0

0

44

57

2

5

49

50

4

8

62

55

3

2

Therefore, RegCM model can be utilized in detection of drought cases but not exactly the period when drought would occur. Even though, detection of drought case or frequency may be helpful for long term planning purposes with some limitation regarding detection of extreme drought case. Table 9.5. Drought number of entire Bangladesh for month 1, Month 3 and Month 6 during 1961-1990 using SPI as a summary.


Regio n

M1

M3

-1≥ Ob s

Centra l North Easter n SouthWest Bangla desh In Perce nt (%)

-2≥

M6

-1≥

-2≥

-1≥

-2≥

Ob s

Mo del

Ob s

Mo del

Ob s

Mo del

Ob s

Mo del

Obs

Mo del

52

M o del 56

3

7

49

57

10

4

58

56

8

1

32 35

49 55

4 1

6 6

50 45

40 57

4 3

9 8

58 39

58 62

1 2

4 3

44

57

2

5

49

50

4

8

62

55

3

2

9

24

19 204 3 Over estimated 6

21

29

21 231 7 Over estimated 6

14

10

16 21 3 7 Model Over estimate d 33

Over estimated 140

Over estimated 38

Under estimated 28

Finally, drought event record of Bangladesh obtained from Bangladesh Bureau of Statistics (BBS) and International Disaster database (EM-DAT) confirms drought events in the year 1974, 1975, 1979, 1980-1985 1988, 1989 in different region of Bangladesh. The Regional Climate Model data could also detect this event on that particular region of Bangladesh. Though there is a limitation for utilizing model data in drought detection in the year 1973, 1976, 1986 and 1987 with references to historical record, it may be concluded that drought event may be projected by using Regional Climate Model with GAS scheme. Conclusions In this study drought situation over Bangladesh is analyzed using data from Observation and Regional Climate Model (RegCM). To see the drought condition at different places of Bangladesh these data has been calculated using a drought index called Standardized Precipitation Index (SPI) during 1961 to 1990. To count the drought spell, a statistical score is prepared. For the confirmation of drought event detected by the model, historical record about drought occurrence has also been utilized. In this investigation it is evident that, • • •

SPI calculation over a region provides better consistency of drought situation instead of single station information. Frequency of drought detection is much better at North and Central regions than the other regions of Bangladesh for all month length. Count of the frequency of drought events indicate that model can be utilized to detect moderate and severe drought cases (-1≥ SPI) than the detection of extreme cases (-2≥ SPI), in all region of Bangladesh. According to statistical score of SPI calculation it is found that model is over detected 33.12% for M1, 5.70% for M3 and 6.00% for M6 month length. However, for extreme drought cases (-2≥ SPI) model over estimated 166.66% and 38% for M1 and M3 month length. And under estimated 28.57% for 6 month length. Hence, model may be useful to detect 3 to 6 month length moderate


â&#x20AC;˘

and severe drought cases over a region but not exactly the period when drought would occur. Drought detection by the RegCM output using drought index also agrees with the historical record of drought in all over Bangladesh with some exceptions.

Finally, it is concluded that Regional Climate Model outputs may be useful for projecting drought frequency but not for particular drought event.


diagnosis and monitoring of drought