Possible climate interaction mechanism between terrestrial climatic parameters and solar activity

Page 1

Possible climate interaction mechanism between terrestrial climatic parameters and solar activity J. Osorio R.∗

B. Mendoza O.

Instituto de Geof´ısica Universidad Nacional Aut´ onoma de M´exico M´exico D.F., C.P. 04510.

Instituto de Geof´ısica Universidad Nacional Aut´onoma de M´exico M´exico D.F., C.P. 04510.

jaime@geofisica.unam.mx

blanca@geofisica.unam.mx

February 22, 2012

Abstract The solar activity has been proposed as one of the main factors of Earths climate variability, however biological processes have been also proposed. The Dimethylsulfide (DMS) is the main biogenic sulfur compound in the atmosphere. The DMS is mainly produced by the marine biosphere and plays an important role in the atmospheric sulfur cycle. Currently it is accepted that terrestrial biota adapts and influences to environmental conditions through regulations on the chemical composition of the atmosphere In the present study we used different methods of analysis to investigate the relationship between the DMS, Low Clouds, Ultraviolet Radiation A (UVA) and Sea Surface Temperature (SST) in the Southern Hemisphere. We found that the series analyzed have different periodicities which can be associated with climatic and solar phenomena such as ENSO, The Quasi-Biennial Oscillation and changes in solar activity mainly in the cycle of the 11 years. Also, we found an anti-correlation between DMS and UVA, the relation between DMS and low clouds is mainly non-linear and there is a correlation between DMS and SST. The results suggest a positive feedback interaction among DMS, solar radiation and low clouds at time-scales shorter (2 years) than the solar cycle.

Keywords: Dimethylsulfide, Solar Activity, Sun-Earth Relations, Wavelet Analysis, Vector Autoregressive Analysis.

Earth Sciences Graduate Programme - UNAM

J. Osorio & B. Mendoza (2012),

1


1

Introduction

The solar activity has been proposed as an external factor of Earth’s climate change, solar phenomena such as total and spectral solar irradiance could change the Earth’s radiation balance and hence climate (Solanki, 2002 ). However, biological processes have also been proposed as another factor of climate change through its impact on cloud albedo. One of the most important issues regarding the Earth function system is whether the biota in the ocean responds to changes in climate (Charlson et al., 1987, Miller et al., 2003, Sarmiento et al., 2004 ). According to several authors, the major source of cloud condensation nuclei (CCN) over the oceans is dimethylsulfide (DMS) (e.g., Andreae and Crutzen, 1997; Vallina et al., 2007 ). Dimethylsulponiopropionate (DMSP) in phytoplankton cells is released into the water column where it is transformed into dimethylsulfide. The DMS diffuses through the sea surface to the atmosphere, where it oxidizes to form a range of products, among them SO2 . This compound oxidizes to from H2 SO4 , which forms sulphate particles that act as CCN. The DMS concentration is controlled by the phytoplankton biomass and by a web of ecological and biogeochemical processes driving by the geophysical context (Sim´ o, 2001 ). Solar radiation is the primary driving mechanism of the geophysical context and is responsible for the growth of the phytoplankton communities. Clouds have a major impact on the amount of heat and radiation budget of the atmosphere. Clouds modify both albedo (short-wave) and long-wave radiation. In particular, for low clouds over oceans, the albedo effect is the most important result of cloud radiation interaction and has a net cooling effect on the climate(Chen et al., 1999; Rossow, 1999 ). The DMS, solar radiation and cloud albedo are hypothesized to have a feedback interaction (Charlson et al., 1987; Shaw et al., 1998; Gunson et al., 2006 ). This feedback can be either negative or positive. A negative feedback process requires a positive correlation between solar irradiance and DMS: increases in solar irradiance reaching the sea surface increase the DMS, augmenting the CCN and the albedo. An overall increase in the albedo produces a decrease in the irradiance reaching the sea surface and thus cooling occurs. A positive feedback requires an anti-correlation between solar irradiance and the DMS: increases in solar irradiance produce a decrease of the DMS, the CCN and the albedo; a net reduced albedo allows more radiation to be absorbed, producing heating. Using seasonal and annual time scales, a first attempt to find correlations between a global database of DMS concentration and several geophysical parameters was unsuccessful (Kettle et al., 1999 ). Quantitative studies that use data bases spanning roughly three decades (Sim´o and Dachs, 2002; Sim´o and Vallina, 2007; Vallina et al., 2007 ) have concluded that the DMS and solar radiation have a high positive correlation on seasonal time scales for most of the ocean, favouring a negative feedback on climate. Another study (Larsen, 2005 ), based on a conceptual model, proposed a positive feedback. A positive feedback would also imply

J. Osorio & B. Mendoza (2012),

2


an anti-correlation between Total Solar Irradiance (TSI) and cloud cover. Indeed, a global anti-correlation between TSI and oceanic low cloud cover has been found (Kristj´ansson et al., 2002; Lockwood, 2005 ). Furthermore, case studies do not allow conclusive results regarding the sign of the solar radiation and DMS correlation (Kniventon et al., 2003; Toole and Siegel, 2004; Toole et al., 2006 ). A more recent study on decadal relation between north and south high latitude concentrations of methane sulphonic acid (MSA), associated exclusively to DMS and TSI, found that at the time scales coincident with the 22 years magnetic solar cycle, the northMSA and TSI follow each other, favoring a negative feedback on local climate; but at the time scales of the 11 years solar cycle for north and south, the DMS-TSI presents an anti-correlation that has increased since the 1940’s favoring a positive feedback on local climate (Mendoza and Velasco, 2009 ), this indicates that the relations between the variables of interest may change with time and location. The purpose of the present study is to examine the relationship between DMS and climatic and solar phenomena, through low cloud cover anomaly (LCC), sea surface temperature (SST) and the solar ultraviolet radiation A (UVA) in a selected location and at time-scales shorter than the solar cycle.

2

Region of Study and Data

The data analysis was performed for the Southern Hemisphere between 40o − 60o latitude for the entire strip length. Considering the abundance of chlorophyll by regions, this area includes the biogeochemical provinces SSTC, FKLD, SANT, ANTA, NEWS, CHIL and TASM (Longhurst, 1995 ). We are interested in this area because it is the least polluted in the world, the so-called pristine zone, then solar effects on biota and climate should be more evident. Over 90% of the area is ocean, the remaining corresponds to the southern parts of Chile, Argentina, Tasmania and New Zealand (South Island). The studied period is 19832008, containing almost 25 year of data. The DMS data set was obtained from the site http://saga.pmel.noaa.gov/dms. The original data series in this web site contains the DMS measurements collected in the global oceans during 1972-2010 and are given as raw data samples (Fig.1a). The sequence of data presents important data gaps in space and time, more details are given in Kettle et al. (1999 ). We also use the SST time series (Fig. 1b), obtained from Climatic Research Unit http://www.cru.uea.ac.uk/cru/data/temperature/ hadsst2sh.txt. Another time series we use is the low cloud cover anomaly data (LCC) from International Satellite Cloud Climatology Project (ISCCP) http://isccp.giss.nasa.gov. Two series of low cloud cover anomaly were obtained: Visible-Infrared (VI-IR)and Infrared (IR) (Fig. 1c and 1d). Finally, we work with solar ultra violet radiation A data (UVA), which comprises wavelengths from 320 to 400 nm, because the 95% of wavelengths longer than 310 nm reach the surface (Lean et al., 1997 ) and has a large impact J. Osorio & B. Mendoza (2012),

3


on marine ecosystems (H¨ader et al., 2003; Toole et al., 2006; Hefu et al. 1997; Slezak et al., 2003; Kniventon et al., 2003; H¨ader et al., 2011 ). We use the UVA composite series from the Nimbus-7 (1978-1985), NOAA-9 (1985-1989), NOAA-11 (1989-1992) and SUSIM satellites between 1992 and 2008 (Fig. 1e)(DeLand et al., 2008 ). The atmospheric attenuated UVA series was calculated at sea surface level using the Santa Barbara Disorted Atmospheric Radiative Transfer (SBDART) program; it was obtained from the site http://www.icess.ucsb.edu/esrg/SBDART.html (Ricchiazzi et al., 1998 ). The average attenuation was estimated at about 5% less than the top of the atmosphere series.

3

The Method

Some of the previous efforts on elucidating a plausible contribution of DMS on the Earth’s climate have been mostly based on correlation analysis models. Such analysis suggest certain relation between the time series, however, they are of global nature and do not provide a precise information about when such a relation occurs. Moreover, the fact that two series have similar periodicities does not necessarily imply that one is the cause and another is the effect. Besides, even if the correlation coefficient is very low, there is the possibility that an existing relation could be of non-linear nature, or that there is a strong phase shift between both phenomena. Here we apply three different non-linear analysis to study the time series: the Wavelet Method, the Fractal Analysis and the Vector Autoregressive Analysis. They are briefly described below.

3.1

Wavelet Analysis

In order to analyze local variations of power within a single non-stationary time series at multiple periodicities we apply the wavelet method using the Morlet function (Torrence and Compo, 1998; Grinsted et al., 2004 ). Meaningful periodicities (confidence level greater than 95%) must be inside the cone of influence (COI), which is the region of the wavelet spectrum outside which the edge effects become important (Torrence and Compo, 1998 ). An appropriate background spectrum is either the white noise (with a flat power spectrum) or the red noise (increasing power with decreasing frequency). We calculate the significance levels in the global wavelet spectra with a simple red noise model (Gilman et al., 1963 ). We only take into account those periodicities above the red noise level. Furthermore, we find the wavelet coherence spectra (Torrence and Compo, 1998; Grinsted et al., 2004 ). It is especially useful in highlighting the time and frequency intervals when the two phenomena have a strong interaction. If the coherence of two series is high, the arrows in the coherence spectra show the phase between the phenomena. Arrows at 0o (horizontal right) indicate that both phenomena are in phase, and arrows at 180o (horizontal left) indicate that they are in antiphase. It is very important to point out that these two cases imply a linear relation between the considered phenomena. Arrows at 90o and 270o (vertical up and down, respectively) indicate an out of phase situation, which means that the two phenomena have a non-linear relation. J. Osorio & B. Mendoza (2012),

4


We also include the global spectra in the wavelet and coherence plots, which is the average power at each period over the whole time span considered within the COI (e.g. Velasco and Mendoza, 2008 ).The uncertainties of the peaks of both global wavelet and coherence spectra are obtained from the peak full-width at the half-maximum of the peak.

3.2

Fractal Analysis

The time series can be characterized by a non-integer dimension (fractal dimension) when treated as random walks or self-affine profiles. Self-affine systems are often characterized by the roughness, which is defined as the fluctuation of the height over a length scale. For self-affine profiles, the roughness scales with the linear size of the surface by an exponent called the roughness or Hurst exponent (H ). Self-affine fractals are generally treated quantitatively using spectral techniques (Turcotte, 1992 ). Also, his referred to as the index of dependence, and is the relative tendency of a time series either to regress strongly to the mean or to cluster in a direction (Kleinow, 2002 ). Most natural systems follow a biased random walk or trend with statistical noise that can be approximated as Gaussian. The strength of the trend and the noise level can be measured by analyzing the ratio R/S, where R indicates the range and S the standard deviation of the observed values, on a time scale in term of H, that is, when H is above 0.50 (Hurst, 1951 ). A value of 1 < H < 1.50 results in afractal dimension(Ds ) nearby to a line. A persistent time series results in a smooth line, with fewer peaks than a random walk. Antipersistent series with 0 < H < 0.50 have a higher Ds . The H exponent and Ds are related as Ds = 2−H (Peters, 1997, Voss, 1988 ).

3.3

Vector Autoregressive Analysis (VAR)

The VAR is a tool for multivariate time series. In the VAR all variables are considered endogenous, as each is expressed as a linear function of its own lagged values and the lagged values of the other considered variables. An univariate auto regression is a singleequation, single-variable linear model in which the current value of a variable is explained by its own lagged values. The VAR is an n-equation, n-variable linear model in which each variable is in turn explained by its own lagged values, plus current and past values of the remaining n-1 variables. This simple framework provides a systematic way to capture rich dynamics in multiple time series with a coherent and credible approach to data description, forecasting, structural inference and analysis (Sims, 1980 ). The VAR is also a powerful technique for generating reliable forecasts in the short term, it has certain limitations, such as omitting the possibility to consider non-linear relationships between variables and it does not take into account conditional heteroscedasticity problems (that is, the observations sample error variances are different from each other) or structural change in the estimated parameters. The VAR finds the relations among

J. Osorio & B. Mendoza (2012),

5


a set of variables and is appropriate for forecasting variables in systems of interrelated time series, where each variable helps to predict the other variables. The VAR is also frequently used in analyzing the dynamic impact of different types of disturbances in systems of variables (L¨ utkepohl, 2005 ). The estimated VAR model uses the method of ordinary least squares. In the methodology of autoregressive models the modeling depends on the behavior of a variable (Yt ) according to its own past values. The general expression of the VAR model is given by: Yt = A1 Yt+1 + . . . + Ap Yt−p + Bxt + t

(1)

Where Yt is a vector of endogenous variables. To estimate the VAR model you select a number of variables that make up the model (L¨ utkepohl, 2005 ). A basic part of the VAR is the use of an impulse-response function that shows the response of current and future values of each of the variables, in other words, this function traces the response of the endogenous variables in the model to an impulse (Lin, 2006 ), we apply here this particular aspect of the VAR.

4

Results and Discussion

The results obtained with the previous analysis will be presented and discussed in this section.

4.1

The Main Frequencies

The results of the wavelet method are shown in Fig. 1, a summary of the main periodicities found for all the time series is shown in the left column of Table 1. Fig. 1e for the UVA shows that the 11 years periodicity is the most significant, although it is partially outside the COI due to the short time interval. The LCCIR and LCCVI-IR coincide in two (0.9 and 5 years) out of three periodicities. To corroborate the frequencies obtained with the wavelet analysis we apply fractal analysis. The results of this method are in Fig. 2, also, a summary appears in the right column of Table 1. With few exceptions, we notice that most of the wavelet periodicities, considering the uncertainties, coincide with the fractal periodicities. Also, according to the H coefficient all the time series can be predicted. The exception is the UVA; we consider that this is due to the shortness of the series compared to the 11 years dominant periodicity.

4.2

The Wavelet Coherence Spectra

The coherence wavelet spectra in Fig. 3 present the coherence analysis between DMS vs SST, DMS vs LCCIR, DMS vs LCCVI-IR and DMS vs UVA respectively along 19922008. We choose this time interval because the DMS time series has the largest quantity of data. For each panel, the time series appears at the top, the wavelet coherence spectrum appears at the middle and the global wavelet coherence spectra is at the right. Table 2 J. Osorio & B. Mendoza (2012),

6


summarizes the results. Fig. 3a, shows that the DMS and SST time series have the most persistent and prominent coherence ∼ 4 years and tend to be in phase. The DMS and LCCIR time series in Fig. 3b present the most prominent and persistent coherence ∼ 5 years, and tend to be in anti-phase. The DMS and LCCVI-IR time series show persistent coherences ∼ 3 and 5 years, they tend to be in phase and anti-phase respectively. The DMS and UVA time series show persistent coherence at ∼ 3 years but it is not very prominent, in fact the prominent coherences are between ∼ 0.4 and 1.2 years, they are very localized in time and tend to be in anti-phase. There is predominance in the periodicity between 3 and 5 years. Peaks shorter than 1 year may be due to seasonal climatic phenomena. The ∼ 2 years period can be associated with the Quasi-Biennial Oscillation (QBO) in the stratosphere(Holton et al., 1972; Dunkerton, 1997; Baldwin et al., 2001; Naujokat, 1986; Holton et al., 1980 )and with the solar activity (Kane, 2005 ).The periodicities ∼ 3 and 4 years could be related to the El Ni˜ no-Southern Oscillation (ENSO) (Nuzhdina, 2002; Njau, 2006: Enfield, 1992; Trenberth et al., 1997; Philander, 1990 ) that is a large-scale climatic phenomenon. The periodicities ∼ 5 years can be a harmonic of the 11 years solar cycle (Djurovic and P´aquet, 1996 ). From Fig. 3 and Table 2, we notice a consistent correlation between DMS and SST and an anti-correlation between DMS and UVA, the relation between DMS and clouds is mainly non-linear. The anti-correlation between UVA and DMS suggest a positive feedback, as discussed in other works (Larsen, 2005 ) or as implied by the findings of other papers (Mendoza and Velasco, 2009; Lockwood, 2005; Kristj´ansson et al., 2002 ).

4.3

The VAR Impulse-Response Function

The impulse-response function appears in Fig. 4. In most cases we observe an impulseresponse of 10 ± 2 months. In the present study the impulse-response function shows statistically how the Earth’s climate has a recovery period of approximately 10 month. From top to bottom in Fig. 4 we notice the strongest meaningful responses of: the DMS to the impulse of SST, the LCC to DMS and the SST to the LCC. This indicates relations among clouds, DMS and SST and between the SST and DMS, supporting the existence of relations among the elements of the feedback process hypothesized in Section 1.

5

Conclusions

Here we study relationship between DMS and the SST, LCCIR, LCCVI-IR and UVA using different methods of analysis. The DMS, SST, LCCIR and LCCVI-IR series show persistence and therefore have the possibility of a future estimation. For the UVA, the results are not realistic and this is due to the shortness of the series that have prominent periodicities for 11 years. The fractal analysis corroborated the wavelet periodicities. Using the wavelet method of spectral analysis, we found a predominance of periodicity between 3 and 5 years. The periodicities ∼ 3 and 4 years could be related to the ENSO. The periodicities ∼ 5 years are associated with solar activity. We found a correlation J. Osorio & B. Mendoza (2012),

7


between DMS and SST and an anticorrelation between DMS and UVA, the relation between DMS and clouds is mainly non-linear. Then, our results suggest a positive feedback interaction among DMS, solar radiation and clouds. The VAR impulse-response function model indicates statistically that Earth’s climate has a recovery period of approximately 10 months and the existence of strong relations among low clouds, DMS and SST and between the SST and DMS.

6

References

Andreae, M.O., Crutzen, P.J. Atmospheric aerosols: biogeochemical sources and role in atmospheric chemistry, Science 276, 1052-1058, 1997. Baldwin, M. P., Gray, L. J., Dunkerton, T. J., Hamilton, K., Haynes, P. H., Randel, W. J., Holton, J. R., Alexander, M. J., Hirota, I., Horinouchi, T., Jones, D. B., Kinnersley, J. S., Marquardt, C., Sato, K., Takahashi, M., The Quasi-Biennial Oscillation, Reviews of Geophysics, Vol. 39, Num. 2, 179-229, 2001. Charlson, R.J., Lovelock, J.E., Andreae, M.O., Warren, S.G. Oceanic phytoplankton, atmospheric sulfur, cloud albedo and climate: a geophysiological feedback. Nature 326, 655-661, 1987. Chen, T., Rossow, B. W., Zhang, Y., Radiative effects of Cloud-Type variations, Journal of climate Vol. 13, 1999. DeLand, M. T., Cebula, R. P., Creation of a composite solar ultraviolet irradiance data set. J. Geophys. Res 113, A11103, 2008. Djurovic, D., Pquet, P., The common oscillations of solar activity, the geomagnetic field, and the earth’s rotation, Solar Physics, Vol. 167, 427-439, 1996. Dunkerton, T. J., The role of gravity waves in the quasi-biennial oscillation. J. Geophys. Res., 102, 26,053-26,076, 1997. Enfield, D. B., Historical and prehistorical overview of El Nio/Southern Oscillation. Cambridge, Cambridge University Press, 1, 95-117, 1992. Grinsted, A., Moore, J., Jevrejera, S., Application of the cross wavelet transform and wavelet coherence to geophysical time series. Nonlinear Processes in Geophysics 11, 561566, 2004. Gunson, J.R., Spall, S.A., Anderson, T.R., Jones, A., Totterdell, I.J., Woodage, M.J., Climate sensitivity to ocean dimethylsulphide emissions. Geophys. Res. Lett. 33, L07701, 2006. J. Osorio & B. Mendoza (2012),

8


Hder, D. P., Helbling, E. W., Williamson, C. E., Worrest, R. C., Effects of UV radiation on aquatic ecosystems and interactions with climate change, Photochem. Photobiol. Sci. 10, 242-260, 2011. Hder, D. P., Kumar, R. C., Smith, R. C., Worrest, R. C., Aquatic ecosystems: effects of solar ultraviolet radiation and interactions with other climatic change factors, Journal of the Royal Society of Chemistry, 2, 39-50, 2003. Hefu, Y., Kirst, G. O., Effects of UV radiation on DMS content and DMS formation of Phaeocystis Antartica, Polar Biol. 18, 402-409, 1997. Holton, J.R., Lindzen, R. S., An updated theory for the Quasi-Biennial cycle of the tropical stratosphere. Journal of the Atmospheric Sciences Vol. 29, 1076-1080. 1972. Holton, J.R., Tan, H. C., The influence of the equatorial Quasi-Biennial Oscillation on the global atmospheric circulation at 50mb, Journal of Atmospheric Science., Vol. 37, 2200-2208, 1980. Hurts, H.E., Long-term storage capacity of reservoirs, Trans. Am. Soc. Civ. Eng., 116, 770-779, 1951. Kettle, A.J., Andreae, M. O., Amouroux, D., et al. A global data base of sea surface Dimethylsulfide (DMS) measurements and a procedure to predict sea surface DMS as a function of latitude, longitude and month, Global Biogeochemical Cycles 13, 399-444, 1999. Kleinow, T., Testing Continuous Time Models in Financial Markets, Berlin HumboldtUniv., Diss. 1, 2002 Kniventon, D. R., Todd, M. C., Sciare, J., Mihalopoulos, N., Variability of atmospheric dimethylsulphide over the southern Indian Ocean due to changes in ultraviolet radiation, Global Biogeochem. Cycles 17, 1096, 2003. Kristjnsson, J. E., Staple, A., Kristiansen, J., Kaas, E., A new look at possible connection between solar activity, clouds and climate, Geophys. Res. Lett., 29, 2107, 2002. Larsen, S.H., Solar variability, dimethylsulphide, clouds and climate, Global Biogeochem.Cycles, 19, GB1014, 2005. Lean, J. R., Lee, G. J., Woods, H., Hickey, T. N., Puga, J., Detection and parameterization of variations in solar mid- and near-ultraviolet radiation (200-400 nm). Geophys. Res. Lett. 102, 939-956, 1997.

J. Osorio & B. Mendoza (2012),

9


Lin, J. L., Teaching notes on impulse response function and structural VAR, Institute of Economics National Chengchi University, 2006. Lockwood, M., Solar outputs, their variations and their effects on Earth, in The Sun, Solar Analogs and the Climate, Saas-Fee Advanced Course 34, 109-306, Springer, the Netherlands, 2005. Longhurst, A., Sathyendranath, S., Platt, T., Caverhill, C., An estimate of global primary production in the ocean from satellite radiometer data, J. Plankton Res., 17, 1245-1271, 1995. Ltkepohl, H., New introduction to multiple time series analysis, Springer-Verlag, 2005. Mendoza, B., Velasco, V., High-Latitude Methane Sulphonic Acid Variability and Solar Activity. J. Atm. and Solar-Terr Phys. 71, 33-40, 2009. Miller, A. J., Alexander, M.A., et al. Potential feedbacks between Pacific Ocean ecosystems and interdecadal climate variations. Bull. Am. Meteorol. Soc., 84, 617-633, 2003. Naujokat, B., An update of the observed Quasi-Biennial Oscillation of stratospheric winds over the tropics, Journal of the Atmospheric Sciences, Vol. 43, 1873-1877, 1986. Njau, E. C., Solar activity, El Nio-Southern oscillation and rainfall in India, Pakistan Journal of Meteorology, Vol. 3, 2006. Nuzhdina, M. A., Connection between ENSO phenomena and solar and geomagnetic activity, Natural Hazards and Earth System Sciences, 2, 83-89, 2002. Peters, E.E., Chaos and order in the capital markets. John Wiley and sons, 1997. Philander, S. G., El Nio, La Nia and the Southern Oscillation, San Diego: Academic Press, 1990. Ricchiazzi, P., Yang, S., Gautier, C., Sowle, D., SBDART:A Research and teaching software tool for plane-parallel radiative transfer in the Earth’s atmosphere. Bull. Amer. Meteor. Soc. 79, 2101-2114, 1998. Rossow, W. B., Schiffer, R. A., Advances in understanding clouds from ISCCP, Bull. Am. Meteor. Soc. 80, 2261- 2287, 1999. Sarmiento, J. L., Gruber, N., Brzezinski, M. A., Dunne, J. P., High-latitude controls of thermocline nutrients and low latitude biological productivity. Nature, 427 (6969),

J. Osorio & B. Mendoza (2012),

10


56-60, 2004. Shaw, G.E., Benner, R.L., Cantrell, W., Veazey, D., The regulation of climate: A sulfate particle feedback loop involving deep convection-An editorial essay. Climate Change 39, 23-33, 1998. Sim´o, R., Production of atmospheric sulfur by oceanic plankton: biogeochemical, ecological and evolutionary links, Trends in Ecology and Evolution, 16, 287-294, 2001. Sim´o, R., Dachs J., Global ocean emission of dymethylsulfide predicted from biogeophysical data, Global Biogeochem. Cycles, 16, 2002. Sim´o, R., Vallina, S. M., Strong relationship between DMS and the solar radiation dose over the global surface ocean, Science, 315, 506-508, 2007. Sims, C.A., Macroeconomics and Reality, Econometrica 48, 1-48, 1980. Slezak, D., Herndl, G. J., Effects of ultraviolet and visible radiation on the cellular concentration of dimethylsulfoniopropionate (DMSP) in Emilianiahuxleyi (strain L), Mar. Ecol. Prog. Ser. 246, 61-71, 2003. Solanki, S. K., Solar variability and climate change: is there a link?, Astronomy and Geophysics, Vol. 43,509-513, 2002. Toole, D. A., Slezak, D., Kiene, R. P., Kieber, D. J., Effects of solar radiation on dimethylsulfide cycling in the western Atlantic Ocean, Deep-Sea Res. I, 53, 136-153, 2006. Toole, D.A., Siegel D. A., Light-driven cycling of Dimethylsulfide (DMS) in the Sargasso Sea: Closing the Loop, Geophys. Res. Lett., Vol. 31, 2004. Torrence, C., Compo, G. B., A practical guide to wavelet analysis, Amer. Meteor.Soc. 79, 61-78, 1998. Trenberth, K. E., The definition of El Nio, Bull. Am. Meteor. Soc. 78 (12): 27717, 1997. Turcotte, D.L., Fractals and Chaos in Geology and Geophysics, Cambridge University Press, 1992. Vallina, S. M., Sim, R., Gass, S., DeBoyer-Montgut, C. D., Jurado, E., Dachs, J., Analysis of a potential ”solar radiation dose-dimethylsulfide-cloud condensation nuclei” link from globally mapped seasonal correlations, Global Biogeochemical Cycles 21, GB2004, 2007.

J. Osorio & B. Mendoza (2012),

11


Velasco, V., Mendoza, B., 2008. Assessing the relationship between solar activity and some large scale climatic phenomena, Advances in Space Research 42, 866-878, 2008. Voss, R.J., Fractals in Nature: from characterization to simulation, The Science of fractal images, Heinz-Otto Peitgen and Dietmar Saupe, Eds., New York: Springer-Verlag, 21-70, 1988.

7

Figure Captions

Fig. 1 Wavelet analysis. For each panel at the top there is the time series, at the middle the wavelet spectra and at the right the global wavelet spectra. (a) Dimethylsulfide (DMS). (b) Sea Surface Temperature (SST). (c) Low Cloud Cover Infrared Anomaly (LCCIR). (d) Low Cloud Cover Visible-Infrared Anomaly (LCCVI-IR). (e) Ultraviolet Radiation A (UVA). The gray code indicating the statistical significance level for the spectral plots appears at the bottom of the figure; in particular the 95% level is inside the black contours. The dashed lines in the global wavelet spectra represent the red noise level. Fig. 2 Power Spectrum plot for the time lapse 1983-2008, the linear trend has been removed. The power-spectral density is given as a function of frequency for time scales of months. The fitted line is used to estimate the fractal dimension (Ds). (a) Dimethylsulfide (DMS).(b) Sea Surface Temperature (SST). (c) Low Cloud Cover Anomaly (LCCIR). (d) Low Cloud Cover Anomaly (LCCVI-IR). (e) Ultraviolet Radiation A (UVA). Fig. 3 Wavelet coherence analysis. For each panel at the top there is the time series, at the middle the wavelet coherence spectra and at the right the global wavelet coherence spectra. (a) DMS (pointed line) and SST (dashed line). (b) DMS (pointed line) and LCCIR (dashed line). (c) DMS (pointed line) and LCCVI-IR. (d) DMS (pointed line) and UVA. The gray code indicating the statistical significance level for the spectral plots appears at the bottom of the figure; in particular the 95% level is inside the black contours. Fig. 4 Estimate of the VAR impulse-response function of the various series studied here for 10 month in the autoregressive model. The dashed lines are the uncertainties of the prediction.

J. Osorio & B. Mendoza (2012),

12


8

Figures

Figure 1: Wavelet Analysis.

J. Osorio & B. Mendoza (2012),

13


Figure 2: Power Spectrum Analysis.

J. Osorio & B. Mendoza (2012),

14


Figure 3: Wavelet Coherence Analysis.

J. Osorio & B. Mendoza (2012),

15


Figure 4: Impulse-Response Function.

J. Osorio & B. Mendoza (2012),

16


Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.