Ecology, Economy and Society–the INSEE Journal 6(1): 109-122, January 2023 RESEARCH PAPER Attributing Vegetation Recovery During the Indian Summer Monsoon to Climate Drivers in Central India Vikram Chandel and Tejasvi Chauhan Abstract: Increasing droughts and heat waves as a result of global warming pose a major threat to forests and croplands in India. Monitoring the dynamics of vegetation during a drought and its recovery is essential for the Indian socio- economy and biodiversity. We investigate vegetation recovery from a stressed state in the pre-monsoon (May) period to the end of the monsoon period (September). We then attribute net change during the monsoon period to climate drivers such as temperature, precipitation, and soil moisture. To delineate non-linear interactions, we use an information-theoretic metric to understand the relative association of climate variables with vegetation productivity on a daily scale. We found that pre- monsoon vegetation stress is influenced by soil moisture (r = 0.8, p < 0.01), which is driven by variations in temperature and precipitation. During the monsoons, precipitation contributes to vegetation recovery from pre-monsoon stress through soil moisture recharge while inhibiting vegetation productivity by limiting the amount of radiation available for photosynthesis. Linear regression shows the significant negative dependence of vegetation recovery on precipitation (β = –0.7, p < 0.01) and positive dependence on soil moisture (β = 0.4, p < 0.1) indicating radiation limitation on photosynthesis. We also found that post-monsoon vegetation recovery is independent of pre-monsoon vegetation stress (p > 0.1). Mutual information showed the stronger, non-linear dependence of vegetation recovery on soil moisture than on precipitation, which is a contrasting result that highlights the importance of including non-linear measures in analyses of natural systems. Our results show that vegetation recovery in central India is driven by soil moisture during the Indian summer monsoon and is independent of pre-monsoon vegetation stress.  Interdisciplinary Programme in Climate Studies, Indian Institute of Technology Bombay, India. vikram.chandel63@gmail.com  Department of Civil Engineering, Indian Institute of Technology Bombay, India. chauhan.tejasvi9@gmail.com Copyright © Chandel and Chauhan 2023. Released under Creative Commons Attribution © NonCommercial 4.0 International licence (CC BY-NC 4.0) by the author. Published by Indian Society for Ecological Economics (INSEE), c/o Institute of Economic Growth, University Enclave, North Campus, Delhi 110007. ISSN: 2581–6152 (print); 2581–6101 (web). DOI: https://doi.org/10.37773/ees.v6i1.927 mailto:vikram.chandel63@gmail.com mailto:chauhan.tejasvi9@gmail.com https://doi.org/10.37773/ees.v6i1.927 Ecology, Economy and Society–the INSEE Journal [110] 1. INTRODUCTION The frequency, intensity, and spatial extent of droughts has been increasing in India due to global warming (Mishra, Shah, and Thrasher 2014; Gupta and Jain 2018). Droughts are often accompanied by heat waves, leading to increased mortality due to heat stress (Rohini, Rajeevan, and Srivastava 2016; Panda, AghaKouchak, and Ambast 2017; Mazdiyasni et al. 2017). Apart from affecting human life, droughts adversely affect vegetation by changing carbon fluxes, causing soil water storages, and disrupting ecosystem services. Consequently, food production (Zhang et al. 2017) in the country is threatened. Prolonged droughts can lead to irreversible shifts in vegetation phenology, thus threatening the biodiversity of a region (Clark et al. 2016; Wendling et al. 2019). Vegetation responds to drought conditions at varying time scales depending on its physiology, and this response that vegetation has to any stressor can be quantified using different metrics (Vicente-Serrano et al. 2013). Resilience is defined as the ability of a system to return to its original state after a disturbance (Scheffer et al. 2009). Resilience can be measured as the time taken by a system to return to its original state after a disturbance. Resistance is defined as the ability to persist and resist change during a disturbance (Nimmo et al. 2015). It can be quantified as the deviation of processes and variables from the mean state during a disturbance; the more the deviation, the less the resistance. Recovery is the process by which an ecosystem retreats toward the pre-disturbance state (Fraccascia, Giannoccaro, and Albino 2018). These metrics help us study the response of different vegetation ecosystems ranging from croplands to tropical rainforests. Forests serve as carbon sinks in the global carbon cycle (Soepadmo 1993; Whitehead 2011; Martin et al. 2001). Understanding the response of forests to drought is important for understanding carbon cycles and any potential adversities to biodiversity. Drought and heat stress cause large-scale tree mortality, which affects the forest phenology and the carbon cycle and turns forests into carbon sources (Ciais et al. 2005; Anderegg, Kane, and Anderegg 2013; Ma et al. 2016; 2015, Li et al. 2019; Jiao et al. 2020; Schuldt et al. 2020; Senf et al. 2020). The drivers of recovery in a forested ecosystem are climate variables, forest type, and forest stock volumes (Anderson-Teixeira et al. 2013, Luo et al. 2022). An ecosystem requires an optimum amount of soil moisture, soil nutrients, air temperature, and solar radiation for a speedy recovery. Soil moisture is recharged by precipitation and depleted by evapotranspiration driven by the air temperature and vapour pressure deficit in the [111] Chandel and Chauhan atmosphere. The optimum temperature for photosynthesis in an ecosystem depends on the biodiversity of the ecosystem (Bonan 2015). Sunlight is an essential ingredient in the process of photosynthesis, and cloudy days result in a decline in photosynthetic rates due to a lack of direct solar radiation. Here, we focus on the interplay of climate variables such as soil moisture, temperature, and precipitation in driving vegetation recovery. We use the leaf area index (LAI), to measure the state of vegetation. LAI is defined as the one-sided green leaf area per unit ground area in broadleaf forests and as half the total needle surface area per unit ground area in coniferous forests. The recovery rate of vegetation system can be measured by the rate of photosynthesis. Gross primary productivity (GPP), a proxy for the photosynthetic rate, is the amount of chemical energy produced by primary producers during photosynthesis. We use both LAI and GPP to study vegetation recovery and its climate drivers in a forested ecosystem. Figure 1: Study Region Lies in Central India, Bounded by 20°N, 77°E, 23°N, and 82°E Source: Authors Ecology, Economy and Society–the INSEE Journal [112] We selected a biodiversity-rich forested region in central India (Figure 1a), bounded by 20°N, 77°E, 23°N and 82°E. The region encompasses the Melghat Tiger Reserve, Satpura National Park, Pench National Park, and some agricultural areas. We investigate the ecosystem’s recovery from a stressed state in the pre-monsoon (May) period to the end of the monsoon period (September). We analyse deviations in the vegetation amount during the pre-monsoon and post-monsoon periods and attribute the net change during the monsoon period to climate drivers such as temperature, precipitation, and soil moisture. We also examined the dependence of post- monsoon vegetation stress to pre-monsoon vegetation stress. To delineate non-linear interactions, we use an information-theoretic metric to understand the relative association of climate variables with vegetation productivity on a daily scale. 2. DATA AND METHODS We use the leaf area index (LAI) obtained from the moderate resolution imaging spectroradiometer (MODIS). This product has a spatial resolution of 500 m and a temporal resolution of eight days. We used rainfall (P) (Pai et al. 2014) and temperature (T) gridded data from the Indian Meteorological Department (IMD). We used the Global Land Evaporation Amsterdam Model (GLEAM V3) dataset to determine soil moisture (SM) (Martens et al. 2017). We also used daily gross primary productivity (GPP) data from FluxSat v2.0, which is a combined dataset of the FLUXNET eddy covariance tower site data and satellite data from MODIS (Joiner and Yoshida 2020). 2.1 VEGETATION STRESS We use percent anomaly in the month of May (Δmay) as an indicator of stress in vegetation. We look at the percent anomaly after the monsoons and calculate the Δsep for September. Both Δmay and Δsep (Figure 2) can take positive and negative values. We use May and September, as during the monsoon period, satellite products like LAI are unreliable due to frequent cloud cover. [113] Chandel and Chauhan Figure 2: (a) Schematic Diagram Representing Vegetation Anomalies Proposed for Measuring Vegetation Recovery; (b) Climatology of the Leaf Area Index (LAI) over the Study Region Source: Authors’ analysis 2.2 INFORMATION THEORY Uncertainty (or variability) of a time series which can be classified into ‘m’ different states can be quantified using Shannon’s entropy as (1) Ecology, Economy and Society–the INSEE Journal [114] where is the probability (or relative frequency) of the state and H is bound as . H can be normalized using to form Hnorm. In dynamical systems, in any interaction between two variables uncertainty regarding one variable reduces when we have information about another. This reduction in uncertainty is called Mutual Information (MI) and can be computed as (2) Where is the joint entropy computed using the joint probabilities of X and Y in equation 1 or 2. MI is bound as . It can measure non-linear associations between variables and hence has been argued to be a measure of true statistical independence (Knuth et al. 2013). To compare the associations between various components inside a system, MI can be normalized as which indicates the percentage of variance of one variable that be explained using another. In this work, to compare the linear and non-linear dependencies of vegetation on precipitation, soil moisture, and temperature, we compute pairwise normalized mutual information ( of daily datasets of soil moisture (SM), temperature (T), and precipitation (P) with gross primary productivity (GPP) to understand the association of hydrometeorological variables with vegetation productivity for the various above-discussed transitions of vegetation stress from May to September for our study region. We partition each time series into 11 bins and compute pairwise using Equation 3 for each year separately and then compute the average for the years falling in the above-discussed transition types. 3. RESULTS Figure 3 shows the climatology of precipitation (P), temperature (T), soil moisture (SM), and gross primary productivity (GPP) in the study area in central India (figure 1). Central India receives rainfall during the Indian summer monsoons (ISMR) in June–September (JJAS). The pre-monsoon period sees high temperatures along with minimal rainfall, while the onset of ISMR provides respite from the heat of summer. The region experiences temperatures as high as 40°C during April and May, which reduce on the arrival of the monsoons and vary from 30–35°C during and after the monsoons. GPP (Figure 3), a measure of photosynthesis, remains low in the pre-monsoon period due to the unavailability of soil moisture and rises [115] Chandel and Chauhan with an increase in soil moisture after the onset of the monsoons. Since GPP is influenced by the air temperature and the amount of radiation and moisture available, if the condition of air temperature and radiation is fulfilled, it is mainly driven by water availability. We observe similar behavior in our region as the seasonal variation of GPP is similar to variation in SM. Figure 3: The Climatology of Precipitation, Temperature, and Soil Moisture and Gross Primary Productivity Source: Authors To understand the recovery of our system, we see the magnitude of stress in vegetation in different years (Figure 4). The negative values of percent anomaly (Δ) indicate stress in vegetation. Δmay varies from -20% to 25%. We performed Pearson correlation of Δmay with precipitation, temperature and Ecology, Economy and Society–the INSEE Journal [116] soil moisture during the pre-monsoon period of March, April and May. We found that temperature has a correlation of –0.7 (p<0.01) which signifies that hotter pre-monsoon period puts stress on vegetation. We found positive correlation of delta with soil-moisture (0.80; p<0.01) and precipitation (0.62; p<0.01), which implies that these variables reduce the stress in month of May. Figure 4: Scatter plot of percentage anomaly in May LAI (Δmay) and corresponding change in September (Δsep ) in same year Source: Authors Δsep varies from –30% to 25% across the years (figure 4). To understand the dominant hydro-meteorological drivers of vegetation productivity during monsoon, we performed a linear regression of Δsep with cumulative precipitation anomalies, mean temperature anomaly, and mean soil moisture anomaly during JJAS. We found that Δsep has a significant (p-value < 0.05) association with soil moisture and precipitation with coefficients –10.1 and 6.2 respectively. Negative coefficient between P and GPP is counter- intuitive and indicates a reduction in vegetation productivity due to radiation limitation imposed on photosynthesis by cloud cover. Hence, high [117] Chandel and Chauhan (and continuous) precipitation can lead to a loss of vegetation productivity. SM on the other hand has a positive coefficient with GPP as soil moisture recharged by precipitation takes time to deplete and can aid photosynthesis when solar radiation is available. The negative anomaly in September is present for both positive and negative values of Δmay (Figure 4). This demands further investigation to find out whether Δsep depends on Δmay or not. To investigate, we added Δmay as an additional predictor to the above regression. We did not find any statistically significant coefficient for Δmay and there was no change in the coefficients of climate other variables. This shows that variations in Δsep are insensitive to variations in Δmay . Since linear regression fails to capture non- linear associations, we employ information-theory based metrics to capture the non-linearities. Figure 5(a) shows average normalised Shannon’s entropy of P, T, SM, and GPP across all years. Shannon’s entropy quantifies the variability of a time series. P and SM are the variables with the lowest and highest entropies, respectively, with the entropy of P at around 40% and that of SM at 90% of the maximum possible value of log(m). Entropies of T is slightly lesser than SM with values between 80%–90% of Hmax. Entropy of GPP is between 60%–70% of Hmax. While Shannon’s entropy measures the variability, mutual information (MI) measures the amount of variability that can be explained between any two variables in the system. MI measures non-linear associations as well, and, hence, is a better measure of association between any two variables. To understand the dynamic association of hydrometeorological variables with GPP, we computed mutual information between these variables. Figure 5(b) shows the average normalised mutual information (MInorm) of P, T, and SM, with GPP, where MInorm is ratio of MI to maximum MI to enable comparison across multiple variable pairs (See methods). MInorm indicates the percentage variance of a variable that can be explained by another variable. SM has highest MI with GPP, varying from 20% to 25%, showing the strong association between variability in soil moisture and variability of vegetation productivity in the region. While linear regression showed the almost equal (and opposite) strength of association between GPP and P and SM, MI shows that GPP has a stronger association with SM—the association is almost twice as strong as than with P. Different relative strengths of association across linear regression and MI indicate strong non- linear associations that linear regression fails to capture. The extra strength of the association between SM and GPP, hence, comes from non-linear interactions. The overall positive association between SM and GPP offsets the negative coefficient of P seen in the linear regression and hence leads to Ecology, Economy and Society–the INSEE Journal [118] positive vegetation productivity. This shows that for our study region, recovery of vegetation from stressed pre-monsoon conditions has a strong dependence on soil moisture. Figure 5: (a) Average Normalised Shannon’s Entropy of P, T, SM, and GPP across All Years; (b) Average Normalised Mutual Information (MInorm) of P, T, and SM, with GPP Where MInorm Is the Ratio of MI to the Maximum MI to Make It Useful for Comparison across Multiple Variable Pairs Source: Authors (a) (b) [119] Chandel and Chauhan While our study provides novel insights on the vegetation dynamics of forest regions, it ignores agricultural areas that are heavily affected by human interventions such as irrigation. Dynamics of vegetation may differ in the presence of anthropogenic stresses, which needs further examination. The independence of September vegetation from May vegetation may not hold for extreme events such as forest fires, heatwaves, and prolonged droughts. During such events, vegetation may cross a tipping point and experience irreversible changes. Precipitation shows a negative correlation with vegetation productivity, which is counter-intuitive. Since precipitation in our study also acts as a proxy for solar radiation, including data such as photosynthetically active radiation (PAR) will provide more accurate inferences. 4. CONCLUSIONS Our study tries to measure the recovery of vegetation from pre-monsoon stress between the months of May and September using deviations from climatology. Since good quality observed datasets are not available for the monsoon period, our approach is a novel method to characterise vegetation recovery and attribute it to climate variables using datasets for the months of May and September, which are mostly available. We tried to attribute vegetation stress to climate variables and found that pre-monsoon monsoon vegetation stress depends on soil moisture, which is driven by variations in temperature and precipitation. Since pre-monsoon precipitation is sparse, heat waves during the pre-monsoon period can deplete soil moisture rapidly. During the monsoons, precipitation contributes to vegetation recovery from pre-monsoon stress through soil moisture recharge, while it also inhibits vegetation productivity by limiting the available radiation for photosynthesis. While linear methods show the strong negative dependence of vegetation on precipitation, indicating radiation-limited photosynthesis, metrics that include non-linear dependencies such as mutual information show stronger dependence on soil moisture. These contrasting results highlight the importance of including non-linear measures in analysis of natural systems. ACKNOWLEDGEMENTS This study was supported by Prime Minister’s Research Fellowship (PMRF), Ministry of Education, Government of India — RSPMRF0262. REFERENCES Anderegg, William R L, Jeffrey M Kane, and Leander D L Anderegg. 2013. “Consequences of Widespread Tree Mortality Triggered by Drought and Ecology, Economy and Society–the INSEE Journal [120] Temperature Stress.” Nature Climate Change, 3: 30–36. https://doi.org/10.1038/nclimate1635 Anderson-Teixeira, Kristina J, Adam D Miller, Jacqueline E Mohan, Tara W Hudiburg, Benjamin D Duval, and Evan H DeLucia. 2013. “Altered Dynamics of Forest Recovery under a Changing Climate.” Global Change Biology, 19(7): 2001–21. https://doi.org/10.1111/gcb.12194 Bonan, Gordon. 2015. Ecological Climatology. Cambridge University Press. https://doi.org/10.1017/cbo9781107339200 Ciais, Ph, M Reichstein, N Viovy, A Granier, J Ogée, V Allard, M Aubinet, et al. 2005. “Europe-Wide Reduction in Primary Productivity Caused by the Heat and Drought in 2003.” Nature, 437: 529–533. https://doi.org/10.1038/nature03972. Clark, James S, Louis Iverson, Christopher W Woodall, Craig D Allen, David M Bell, Don C Bragg, Anthony W D’Amato, et al. 2016. “The Impacts of Increasing Drought on Forest Dynamics, Structure, and Biodiversity in the United States.” Global Change Biology, 22(7): 2329–52. https://doi.org/10.1111/gcb.13160 Fraccascia, Luca, Ilaria Giannoccaro, and Vito Albino. 2018. “Resilience of Complex Systems: State of the Art and Directions for Future Research.” Complexity. https://doi.org/10.1155/2018/3421529 Gupta, Vivek, and Manoj Kumar Jain. 2018. “Investigation of Multi-Model Spatiotemporal Mesoscale Drought Projections over India under Climate Change Scenario.” Journal of Hydrology, 567: 489–509. https://doi.org/10.1016/j.jhydrol.2018.10.012 Jiao, Tong, Christopher A Williams, John Rogan, Martin G De Kauwe, and Belinda E Medlyn. 2020. “Drought Impacts on Australian Vegetation During the Millennium Drought Measured With Multisource Spaceborne Remote Sensing.” Journal of Geophysical Research: Biogeosciences, 125(2). https://doi.org/10.1029/2019JG005145 Joiner, Joanna, and Yasuko Yoshida. 2020. “Satellite-Based Reflectances Capture Large Fraction of Variability in Global Gross Primary Production (GPP) at Weekly Time Scales.” Agricultural and Forest Meteorology, 291. https://doi.org/10.1016/j.agrformet.2020.108092 Knuth, Kevin H, Anthony Gotera, Charles T Curry, Karen A Huyser, Kevin R Wheeler, and William B Rossow. 2013. “Revealing Relationships among Relevant Climate Variables with Information Theory,” November. https://doi.org/10.48550/arxiv.1311.4632 Li, Xiangyi, Yue Li, Anping Chen, Mengdi Gao, Ingrid J. Slette, and Shilong Piao. 2019. “The Impact of the 2009/2010 Drought on Vegetation Growth and Terrestrial Carbon Balance in Southwest China.” Agricultural and Forest Meteorology. https://doi.org/10.1016/j.agrformet.2019.01.036 Luo, Hui, Tao Zhou, Peixin Yu, Chuixiang Yi, Xia Liu, Yajie Zhang, Peifang Zhou, Jingzhou Zhang, and Yixin Xu. 2022. “The Forest Recovery Path after Drought Dependence on Forest Type and Stock Volume.” Environmental Research Letters, 17(5). https://doi.org/10.1088/1748-9326/ac57e5 https://doi.org/10.1038/nclimate1635 https://doi.org/10.1111/gcb.12194 https://doi.org/10.1017/cbo9781107339200 https://doi.org/10.1111/gcb.13160 https://doi.org/10.1155/2018/3421529 https://doi.org/10.1016/j.jhydrol.2018.10.012 https://doi.org/10.1029/2019JG005145 https://doi.org/10.1016/j.agrformet.2020.108092 https://doi.org/10.48550/arxiv.1311.4632 https://doi.org/10.1016/j.agrformet.2019.01.036 https://doi.org/10.1088/1748-9326/ac57e5 [121] Chandel and Chauhan Ma, Xuanlong, Alfredo Huete, James Cleverly, Derek Eamus, Frédéric Chevallier, Joanna Joiner, Benjamin Poulter, et al. 2016. “Drought Rapidly Diminishes the Large Net CO2 Uptake in 2011 over Semi-Arid Australia.” Scientific Reports, 6. https://doi.org/10.1038/srep37747 Ma, Xuanlong, Alfredo Huete, Susan Moran, Guillermo Ponce-Campos, and Derek Eamus. 2015. “Abrupt Shifts in Phenology and Vegetation Productivity under Climate Extremes.” Journal of Geophysical Research: Biogeosciences, 120(10): 2036–52. https://doi.org/10.1002/2015JG003144 Martens, Brecht, Diego G. Miralles, Hans Lievens, Robin Van Der Schalie, Richard A.M. De Jeu, Diego Fernández-Prieto, Hylke E. Beck, Wouter A. Dorigo, and Niko E.C. Verhoest. 2017. “GLEAM v3: Satellite-Based Land Evaporation and Root-Zone Soil Moisture.” Geoscientific Model Development, 10(5): 1903–25. https://doi.org/10.5194/gmd-10-1903-2017 Martin, Philippe H, Gert Jan Nabuurs, Marc Aubinet, Timo Karjalainen, Edward L Vine, John Kinsman, and Linda S Heath. 2001. “Carbon Sinks in Temperate Forests.” Annual Review of Energy and the Environment, 26: 435–65. https://doi.org/10.1146/annurev.energy.26.1.435 Mazdiyasni, Omid, Amir AghaKouchak, Steven J Davis, Shahrbanou Madadgar, Ali Mehran, Elisa Ragno, Mojtaba Sadegh, et al. 2017. “Increasing Probability of Mortality during Indian Heat Waves.” Science Advances, 3(6). https://doi.org/10.1126/sciadv.1700066 Mishra, Vimal, Reepal Shah, and Bridget Thrasher. 2014. “Soil Moisture Droughts under the Retrospective and Projected Climate in India.” Journal of Hydrometeorology, 15(6): 2267–92. https://doi.org/10.1175/JHM-D-13-0177.1 Nimmo, D G, R Mac Nally, S C Cunningham, A Haslem, and A F Bennett. 2015. “Vive La Résistance: Reviving Resistance for 21st Century Conservation.” Trends in Ecology and Evolution, 30(9): 516–23. https://doi.org/10.1016/j.tree.2015.07.008 Pai, D S, Latha Sridhar, M Rajeevan, O P Sreejith, N S Satbhai, and B Mukhopadhyay. 2014. “Development of a New High Spatial Resolution (0.25° × 0.25°) Long Period (1901-2010) Daily Gridded Rainfall Data Set over India and Its Comparison with Existing Data Sets over the Region.” Mausam, 65(1): https://mausamjournal.imd.gov.in/index.php/MAUSAM/article/view/851. Panda, Dileep Kumar, Amir AghaKouchak, and Sunil Kumar Ambast. 2017. “Increasing Heat Waves and Warm Spells in India, Observed from a Multiaspect Framework.” Journal of Geophysical Research, 12(7): 3837–58. https://doi.org/10.1002/2016JD026292. Rohini, P, M Rajeevan, and A K Srivastava. 2016. “On the Variability and Increasing Trends of Heat Waves over India.” Scientific Reports, 6. https://doi.org/10.1038/srep26153 Scheffer, Marten, Jordi Bascompte, William A Brock, Victor Brovkin, Stephen R Carpenter, Vasilis Dakos, Hermann Held, Egbert H Van Nes, Max Rietkerk, and George Sugihara. 2009. “Early-Warning Signals for Critical Transitions.” Nature, 461: 53–59. https://doi.org/10.1038/nature08227 https://doi.org/10.1038/srep37747 https://doi.org/10.1002/2015JG003144 https://doi.org/10.5194/gmd-10-1903-2017 https://doi.org/10.1146/annurev.energy.26.1.435 https://doi.org/10.1126/sciadv.1700066 https://doi.org/10.1175/JHM-D-13-0177.1 https://doi.org/10.1016/j.tree.2015.07.008 https://mausamjournal.imd.gov.in/index.php/MAUSAM/article/view/851 https://doi.org/10.1038/srep26153 https://doi.org/10.1038/nature08227 Ecology, Economy and Society–the INSEE Journal [122] Schuldt, Bernhard, Allan Buras, Matthias Arend, Yann Vitasse, Carl Beierkuhnlein, Alexander Damm, Mana Gharun, et al. 2020. “A First Assessment of the Impact of the Extreme 2018 Summer Drought on Central European Forests.” Basic and Applied Ecology, 45: 86–103. https://doi.org/10.1016/j.baae.2020.04.003 Senf, Cornelius, Allan Buras, Christian S Zang, Anja Rammig, and Rupert Seidl. 2020. “Excess Forest Mortality Is Consistently Linked to Drought across Europe.” Nature Communications, 11. https://doi.org/10.1038/s41467-020-19924-1 Soepadmo, E. 1993. “Tropical Rain Forests as Carbon Sinks.” Chemosphere, 110(1): 52–57. https://doi.org/10.1016/0045-6535(93)90066-E Vicente-Serrano, Sergio M, Célia Gouveia, Jesús Julio Camarero, Santiago Beguería, Ricardo Trigo, Juan I. López-Moreno, César Azorín-Molina, et al. 2013. “Response of Vegetation to Drought Time-Scales across Global Land Biomes.” Proceedings of the National Academy of Sciences of the United States of America. https://doi.org/10.1073/pnas.1207068110 Wendling, Valentin, Christophe Peugeot, Angeles G Mayor, Pierre Hiernaux, Eric Mougin, Manuela Grippa, Laurent Kergoat, Romain Walcker, Sylvie Galle, and Thierry Lebel. 2019. “Drought-Induced Regime Shift and Resilience of a Sahelian Ecohydrosystem.” Environmental Research Letters, 14(10). https://doi.org/10.1088/1748-9326/ab3dde Whitehead, David. 2011. “Forests as Carbon Sinks – Benefits and Consequences.” Tree Physiology, 31(9): 893–902. https://doi.org/10.1093/treephys/tpr063 Zhang, Xiang, Renee Obringer, Chehan Wei, Nengcheng Chen, and Dev Niyogi. 2017. “Droughts in India from 1981 to 2013 and Implications to Wheat Production.” Scientific Reports, 7. https://doi.org/10.1038/srep44552 https://doi.org/10.1016/j.baae.2020.04.003 https://doi.org/10.1038/s41467-020-19924-1 https://doi.org/10.1016/0045-6535(93)90066-E https://doi.org/10.1073/pnas.1207068110 https://doi.org/10.1088/1748-9326/ab3dde https://doi.org/10.1093/treephys/tpr063 https://doi.org/10.1038/srep44552