Top Navigation Bar

Article Request Page ASABE Journal Article

Synthetic Scenarios from CMIP5 Model Simulations for Climate Change Impact Assessments in Managed Ecosystems and Water Resources: Case Study in South Asian Countries

A. Anandhi, N. Omani, I. Chaubey, R. Horton, D. A. Bader, R. S. Nanjundiah


Published in Transactions of the ASABE 59(6): 1715-1731 (doi: 10.13031/trans.59.11585). Copyright 2016 American Society of Agricultural and Biological Engineers.


Submitted for review in October 2016 as manuscript number NRES 11585; approved for publication as part of the Climate Change collection by the Natural Resources & Environmental Systems Community of ASABE in December 2016.

The authors are Aavudai Anandhi, Assistant Professor, Department of Biological Systems Engineering, Florida Agricultural and Mechanical University, Tallahassee, Florida; Nina Omani, ASABE Member, and Indrajeet Chaubey ASABE Member, Professor and Head, Department of Earth, Atmospheric, and Planetary Sciences, Purdue University, West Lafayette, Indiana; Radley Horton, Associate Research Scientist, and Daniel A. Bader, Programmer, Center for Climate Systems Research, Columbia University, New York, New York; Ravi S. Nanjundiah, Professor and Chair, Centre for Atmospheric and Oceanic Sciences and Divecha Centre for Climate Change, Indian Institute of Science, Bangalore, India. Corresponding author: Dr. Aavudai Anandhi, PAIG, Florida Agricultural and Mechanical University, Tallahassee, FL 32307; phone: 785-340-5873; e-mail: anandhi@famu.edu.

Abstract. Increasing population, urbanization, and associated demand for food production compounded by climate change and variability have important implications for the managed ecosystems and water resources of a region. This is particularly true for south Asia, which supports one quarter of the global population, half of whom live below the poverty line. This region is largely dependent on monsoon precipitation for water. Given the limited resources of the developing countries in this region, the objective of our study was to empirically explore climate change in south Asia up to the year 2099 using monthly simulations from 35 global climate models (GCMs) participating in the fifth phase of the Climate Model Intercomparison Project (CMIP5) for two future emission scenarios (representative concentration pathways RCP4.5 and RCP8.5) and provide a wide range of potential climate change outcomes. This was carried out using a three-step procedure: calculating the mean annual, monsoon, and non-monsoon precipitation and temperatures; estimating the percent change from historical conditions; and developing scenario funnels and synthetic scenarios. This methodology was applied for the entire south Asia region; however, the percent change information generated at 1.5° grid scale can be used to generate scenarios at finer spatial scales. Our results showed a high variability in the future change in precipitation (-23% to 52%, maximum in the non-monsoon season) and temperature (0.8% to 2.1%) in the region. Temperatures in the region consistently increased, especially in the Himalayan region, which could have impacts including a faster retreat of glaciers and increased floods. It could also change rivers from perennial to seasonal, leading to significant challenges in water management. Increasing temperatures could further stress groundwater reservoirs, leading to withdrawal rates that become even more unsustainable. The high precipitation variability (with higher propensity for localized intense rainfall events) observed in the region can be a key factor for managed ecosystems and water management and could also lead to more incidence of severe urban flooding. The results could be used to assess both mitigation and adaptation alternatives to reduce vulnerabilities in managed ecosystems (agricultural and urban) and water resources.

Keywords.Adaptation, Agriculture, Agro-ecosystems, Climate change, Mitigation, Urban ecosystems, Water supply.

The emerging scientific consensus suggests that the global population is likely to exceed 9 billion people by 2050 and is unlikely to stabilize in the 21st century (Gerland et al., 2014). Most of the increasing population will be in cities. The unprecedented rates of urban population growth over the past century have occurred on <3% of the global terrestrial surface. However, the impacts have been global, with increases of 78% for carbon emissions, 60% for residential water use, and 76% for wood used for industrial purposes attributed to cities (Grimm et al., 2008). Urbanization, while providing additional living space and infrastructure, may diminish agricultural output due to farmland loss and may change ecological conditions due to conversion and fragmentation of forests and other natural landscapes (Alig et al., 2004). More than 95% of the net increase in the global population will be in cities of the developing world (Grimm et al., 2008). The increase in population will require 70% to 100% more food (Tscharntke et al., 2012). Even the most optimistic scenarios require at least a 50% increase in food production (Horlings and Marsden, 2011). Considering the shift of populations to urban areas, as well as climate change and variability, the need to increase food production while conserving our natural resources with limited land and water resources is challenging. In addition, the climate change related impacts on natural and human systems at various time scales can cause disruptions to economic activities, including agriculture, industry, and services, with a range of consequences.

The impacts of climate change and variability on managed ecosystems and water resources are often studied using simulations from global climate model (GCM) scenarios combined with ecosystem and hydrological models (Anandhi et al., 2011a). Future climate is uncertain and unknown. GCMs are among the most advanced tools, which simulate climatic conditions on earth hundreds of years into the future (Anandhi et al., 2008). The scenarios are often used in investigating the potential consequences of anthropogenic climate change and natural climate variability (Anandhi et al., 2009). A scenario is a coherent, internally consistent, and plausible description of a possible future state of the world (Berkhout et al., 2002). In the past, the IPCC scenarios (IS92) (Leggett et al., 1992) and the scenarios from the Special Report on Emission Scenarios (SRES) (Nakicenovic et al., 2000) were commonly used in climate change impact assessments. In recent impact assessments, the four future scenarios’ representative concentration pathways (RCPs) (Van Vuuren et al., 2011) have been used. The fifth phase of the Climate Model Intercomparison Project (CMIP5), a freely available state-of-the-art multi-model dataset (multiple GCMs and RCPs), was designed to advance our knowledge of climate variability and climate change (Taylor et al., 2012). The difficulty encountered in using the scenarios from GCMs has been the mismatch of spatial scales between GCMs and local impact assessments (Anandhi et al., 2011a).

A number of studies have analyzed the climate change and variability in CMIP5 and RCPs for south Asia by studying the changes in precipitation and temperature at seasonal and annual scales (Sabeerali et al., 2015; Sandeep and Ajayamohan, 2015; Seth et al., 2013; Sooraj et al., 2015; Ul Hasson et al., 2016; Wang et al., 2014). Others have analyzed the impacts of climate change and variability on managed ecosystems and water resources (Jayaraman and Murari, 2014). Some of these studies are for smaller regions in south Asia, such as India (Jayaraman and Murari, 2014; Mishra and Lilhare, 2016; Sonali and Kumar, 2016), Pakistan (Zahid and Iqbal, 2015), and China (Chen, 2013; Chong-Hai and Ying, 2012; Sun et al., 2015). Studies have also evaluated the CMIP5 model simulations of precipitation and temperature for smaller regions in south Asia (Menon et al., 2013; Mishra et al., 2014; Raju et al., 2016). Many of these studies have focused on the Indian monsoon (Menon et al., 2013; Mishra et al., 2014; Niu et al., 2015; Raju et al., 2016; Ramesh and Goswami, 2014; Saha et al., 2014). Most of these studies provide information on changes in climate variables (e.g., precipitation and temperature) and the performance and reliability of GCMs in simulating them. However, they cannot be used directly for impact assessments. For climate change impact assessments in managed ecosystems and water resources, scenarios are often built from the time series of climate variables by downscaling them to fit the local scale. These downscaled scenarios are provided as input to ecosystem and/or hydrologic models. These downscaled scenarios can be obtained in a number of ways, and each of them has its advantages and disadvantages (Anandhi et al., 2011a).

The climate scenarios are regarded as “learning machines” and are often used as planning and communication tools to explore an uncertain future climate (Berkhout et al., 2002). Scenario building has also been recognized as a useful tool to examine climate risks and uncertainties and involve decision-makers in the adaptation process (Tschakert and Dietrich, 2010). The objective of our study was to generate synthetic scenarios of precipitation and temperature using innovative scenario funnels. The methodology is explained by empirically exploring climate change in south Asia up to the year 2099 using monthly simulations from 35 GCMs participating in CMIP5 for two future emission scenarios (RCP4.5 and RCP8.5). These GCMs provide a wide range of potential climate change outcomes, which then can be re-sampled from observed data. The results can be used to assess both mitigation and adaptation alternatives to reduce vulnerabilities in managed ecosystems (agricultural and urban) and water resources. South Asia was chosen as the study region because this region supports one quarter of the global population, half of whom live below the poverty line. The scenario funnels generated in this study will be useful to various stakeholder groups with limited resources for developing detailed scenarios for climate change impact assessments, and they can be used in preliminary studies.

Data and Study Region

Monthly simulations of precipitation from 35 GCMs that participated in the CMIP5 were used in this study. Results from historical (1986-2005) and future (2006-2099 and 2006-2099, RCP4.5 and RCP8.5) experiments were investigated. Further information on the experiments and models is available on the CMIP5 website (http://cmip-pcmdi.llnl.gov/cmip5/). The data were extracted and interpolated to a common 1.5° grid square using bilinear interpolation. The future was divided into two periods (2006-2050 and 2051-2099).

The study region extends from the equator (0°) to 40.5° N and from 66° E to 100.5° E. This region completely encloses eight countries (Afghanistan, Bangladesh, Bhutan, India, Myanmar, Nepal, Pakistan, and Sri Lanka) and portions of seven other countries (China, Indonesia, Iran, Tajikistan, Thailand, Turkmenistan, and Uzbekistan). The study region has 729 grid squares. Each grid square has dimensions of 1.5° × 1.5°. These 1.5° grid squares are the smallest unit of spatial resolution on which all calculations are made.

Methods

The methodology involved a three-step process: estimation of mean precipitation and temperature, followed by determination of percent change, and finally the development of synthetic scenarios and scenario funnels. In general, the temporal-scale analysis was performed for three categories and three time periods. The three categories were the monsoon season (June to September), non-monsoon season (January to May and October to December), and the entire year (annual). The three time periods were 1971-2005, i.e., the baseline climate (20C3M) time period, and 2006-2050 and 2051-2099, i.e., two future climate scenario (RCP4.5 and RCP8.5). In general, spatial analysis was performed at grid square (e.g., percent change), country level (e.g., mean, percentile values), or for the entire study region (scenario funnel). In this study, climate variability uses the IPCC third assessment description (McCarthy, 2001). Climate variability refers to variations in the mean state and other statistics (such as percent change, annual mean, etc.) of the climate on all temporal and spatial scales beyond that of individual weather events. Variability may be due to natural internal processes within the climate system (internal variability) or to variations in natural or anthropogenic external forcing (external variability).

Mean Precipitation and Temperatures

Mean precipitation and mean temperatures were first estimated for each GCM, year, and grid square. The grid squares falling in each country were then identified. Depending on the country boundary, some grid squares were in multiple countries. For each country and year, the mean, 10th percentile, and 90th percentile values were calculated from the grid squares and 35 GCMs and plotted (figs. 1 through 6). In this study, the differences between the 10th and 90th percentiles were used to represent the variability in a region. When country boundaries were overlaid on grid square boundaries, certain country boundaries that fell within the study region had fewer than five grid squares (e.g., Sri Lanka). In such cases, only the mean values were calculated and plotted, rather than the 10th and 90th percentile values. In figures 1 through 6, the red lines represent the variability among GCMs in RCP8.5, while the blue band represents the variability in RCP4.5. For RCP8.5, we used red lines rather than bands for precipitation because there was considerable overlap with RCP4.5.

Percent Change

The percent change (Changei,c) between the historical and future GCM ensemble means (GCMh, GCMf) for each grid square was calculated using equation 1 for each category (c) and future time period (i). The representative change values for each country were calculated by aggregating the values within the country. The means were estimated from 35 GCM simulations at each grid square for the three categories and three time scales (figs. 7 and 8). From this, we calculated the ensemble means by aggregating all 35 GCM mean values for each grid square, category, and time period. The variability among the 35 GCMs was represented using the ensemble mean and standard deviation (parametric statistics), while the median and interquartile range represented the non-parametric statistics:

      (1)

For each grid square, Changei,c was percent change for each category (c) and future time period (i), GCMfi,c was the future GCM ensemble mean for each category (c) and future time period (i), and GCMhc was the historical GCM ensemble mean for each category (c).

Developing Scenario Funnels

Scenario funnels were developed from the percent changes calculated from historical conditions. A separate funnel was developed for each climate variable (precipitation and temperature), category (annual, monsoon season, and non-monsoon season), RCP, and time period. The funnel point represented the baseline mean conditions from the historical period (1971-2005), and the funnel length represented the span of the future time period being assessed (2006-2050 or 2051-2099). The width of the funnel mouth for each RCP (4.5 or 8.5) and category represented the percent change among GCMs from historical conditions to future conditions (2006-2050 or 2051-2099). Thus, each funnel diagram for a category included four overlapping funnels: the RCP4.5 scenario for 2006-2050 and 2051-2099, and the RCP8.5 scenario for 2006-2050 and 2051-2099. The point of the funnel (baseline conditions) remained the same for the overlapping funnels. Typically, the RCP4.5 scenario represented a lesser percent change from baseline conditions, which resulted in two narrower funnels, relative to the two wider RCP8.5 funnels, as shown in figure 9. The circles and triangles shown in the funnels were used to differentiate the four overlapping funnels, especially when viewed in gray-scale format. The positions of these symbols within the funnels were random.

Generating Synthetic Scenarios from Scenario Funnels

Synthetic scenarios of future changes in precipitation and temperature were developed from percent change using change factor methodology. Scenarios provide a dynamic view of the future, exploring the effects of various trajectories of changes, leading to a broad range of plausible alternative futures. Synthetic scenarios are not forecasts nor predictions nor projections, but rather possibilities based on data (Anandhi et al., 2008). Studies relating to climate change adaptation and mitigation in managed ecosystems and water resources use scenarios combined with hydrologic and ecosystem models for adaptive management (Matonse et al., 2013; Mukundan et al., 2013; Murugan et al., 2012; Pradhanang et al., 2013).

In some of these studies, scenarios were created using simple downscaling methods such as change factor methodology (Anandhi et al., 2011a; Schoof, 2013). The scenarios thus obtained were also referred to as synthetic scenarios (Carter et al., 1994). They are usually adapted for exploring system sensitivity prior to the application of more credible, model-based scenarios (Matonse et al., 2013; Mukundan et al., 2013; Pradhanang et al., 2013). From

percent change (figs. 7 and 8), we developed a scenario funnel for the entire south Asia region (fig. 9). A similar procedure can be applied at country or local scale.

Depending on the application and the range of change in the scenario funnel, a variety of change factors varying by 3% to 15% (increase or decrease) could be chosen. The change factors, when multiplied with the observed historical precipitation and temperature values, provide the synthetic climate change scenarios. From the scenario funnel, we estimated change factors that can then be applied incrementally by arbitrary amounts (e.g., 1°C, 2°C, and 3°C change in temperature) to the historical climate to obtain future scenarios. These synthetic scenarios need to be used as rough estimates of climate change. Studies have shown that scenario funnels illustrate the variety of long-term consequences of short-term decisions, using widening perspectives and illuminating key issues that may otherwise be missed (Kosow and Gaßner, 2008; Liu et al., 2008; Mahmoud et al., 2009; Timpe and Scheepers, 2003).

Figure 1. Ensemble mean annual precipitation (mm d-1) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.

Results

The ensemble mean annual precipitation (mm d-1) for 15 countries in south Asia was estimated for three categories (annual, monsoon, and non-monsoon), as shown in figures 1 through 3. In general, among the 15 countries, there was no significant change in annual precipitation, especially in RCP4.5. For six countries (Nepal, Bangladesh, Bhutan, China, India, and Sri Lanka), there was an increase in annual and monsoon precipitation, especially in the 90th percentile values. The variability among GCMs was higher for RCP8.5 when compared to RCP4.5, which was observed by the spread of the 10th and 90th percentile values (red lines for RCP8.5 and blue band for RCP4.5). The GCMs showed up to 8 mm d-1 variability (e.g., India). Parts of Iran and Turkmenistan had the least precipitation in the region. Bhutan and Indonesia had the highest precipitation in the region. In south Asia, precipitation occurs mainly during monsoon months in most countries, except portions of Iran, Turkmenistan, and Uzbekistan, where the precipitation is generally low throughout the year.

Figure 2. Ensemble mean monsoon season precipitation (mm d-1) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.

Mean Temperatures

The ensemble temperatures for 15 countries in south Asia were estimated for three categories (annual, monsoon, and non-monsoon), as shown in figures 4 through 6. In general, in all three categories, there was an increase in temperature in south Asia, with RCP8.5 showing a higher increase when compared to RCP4.5. The increases varied with country and category. The variability among GCMs was similar for RCP4.5 and RCP8.5, which can be observed by the spread of the 10th and 90th percentile values (red band for RCP8.5 and blue band for RCP4.5). All GCMs consistently provided the same change in direction. The variability in mean temperature predicted by GCMs range from 1 K to 15 K for the three categories, with most countries having up to 10 K variability.

Percent Change in Mean Precipitation and Temperature

The percent change in the three categories, two future scenarios, and two future time periods are shown in figure 7 for precipitation and in figure 8 for temperature. In the figures, the change is provided at two spatial scales: 1.5° grid square (color shades) and country level (numbers in figure). For precipitation, we observed that the direction of change was not consistent within a country and varied with future scenario, time period, and category. From the mean change in precipitation for each country, there was a general increase of up to 17% in monsoon precipitation in the region (except Turkmenistan). The increase varied with scenario, time period, and country. Most countries in the region also showed an increase in non-monsoon precipitation, except Afghanistan and western Iran, which showed decreases of up to 9%. In general, we observed that among the future scenarios, the change in precipitation was higher and more variable among grid squares for RCP8.5 than for RCP4.5. For the categories, the change in precipitation and the variability among grid squares were highest during monsoon season (except for RCP8.5 during 2051-2099), followed by annual and non-monsoon season. Few GCMs showed a very high change in precipitation due to very low mean precipitation during the historical period. These very high values were not included in the mean percent change estimations.

All countries and grid squares consistently showed an increased temperature change of up to 2.1% at grid square scale and 1.8% at country scale. The increase varied with scenario, time period, and country. In general, the non-monsoon category had a relatively higher increase than monsoon (except for RCP4.5 during 2006-2050). The highest increase was observed in the non-monsoon category for RCP8.5 during 2051-2099. The northern portions of south Asia had a higher increase when compared to the southern portions. The highest percent change was observed in the Himalayan region in the northern part of south Asia.

Figure 3. Ensemble mean non-monsoon season precipitation (mm d-1) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.

Scenario Funnel for Precipitation and Temperature Change

The scenario funnels generated from percent change are shown in figure 9. There were 24 scenario funnels generated in this study: 12 funnels for precipitation (figs. 9a, 9b, and 9c) and 12 for temperature (fig. 9d). The 12 (= 3 × 2 × 2)

Figure 4. Ensemble mean annual temperatures (K) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.

precipitation and 12 temperature scenario funnels were for combinations of the three categories (annual, monsoon, non-monsoon), two time periods (2006-2050 and 2051-2099), and two future scenarios (RCP4.5 and RCP8.5). The 24 funnels were developed from figures 7 and 8 (one from each subplot in the two figures). The length of the funnel is the time period. It was the same for all 24 funnels, although the numbers of years in the two time periods were slightly different. The width of the funnel mouth varied with the ranges shown in the legends in figures 7 and 8. The bottom row in figure 7 (RCP8.5 for 2051-2099) has the widest range in percent change for precipitation for all three categories. Among the three catergories, non-monsoon precipitation has the widest range (-23% to +52%), followed by monsoon (-23% to +43%) and annual (-12% to +37%). The top row in figure 7 (RCP4.5 for 2006-2050) has the narrowest range in percent change for all three categories, with RCP4.5 for 2051-2099 and RCP8.5 for 2006-2050 in between. Thus, the twelve scenario funnels for precipitation form four concentric funnels for each of the three categories (figs. 9a, 9b, and 9c). For temperature (fig. 8), the three columns (three categories) have very negligible difference in their ranges. Therefore, only four scenario funnels were developed, one each for each row (each scenario and time period combination). These four scenario funnels were concentric (fig. 9d). Similarly, scenario funnels can be developed for each country, for portions of each country (e.g., watershed boundary or state boundary), or for an area of interest for precipitation and temperature by using the approximate ranges of percent change from figures 7 and 8.

Synthetic Scenarios of Precipitation and Temperature Change

Synthetic change scenarios can be developed from the scenario funnels. Increments of change can be developed from the funnel width. The varying width of the funnel mouth can be divided evenly or as a ratio 0.5% to 5% (increase and/or decrease) to develop increments. For example, the temperature funnel has a range +0.2% to +2.1% change in temperature. This range can be evenly divided depending on the number of synthetic scenarios required. If three scenarios are required, then the three incremental percent changes in temperature are +0.2%, +1.15% [= 0.2 + (2.1 - 0.2)/2], and +2.1%. If four scenarios are required, then the four incremental percent changes in temperature are +0.2%, 0.83% [= 0.2 + (2.1 - 0.2)/3], 1.46% [= 0.83 + (2.1 - 0.2)/3], and 2.1%. The number of increments can vary with the application, resource availability, and stakeholders making the decision (e.g., managers, scientists, agronomists, and producers). The incremental percent change values, when multiplied with the observed time series of temperature and precipitation, provide synthetic scenarios of climate change for the region. These scenarios are also referred to as “incremental scenarios” because they do not necessarily present a realistic set of changes that are physically plausible (Anandhi et al., 2011a). They are usually adapted for exploring system sensitivity prior to the application of more credible, model-based scenarios. These synthetic scenarios need to be used as rough estimates of climate change, which then can be used for developing adaptation and mitigation strategies in the region. In the absence of resources for developing scenarios of climate change using more sophisticated downscaling techniques, the synthetic scenario methodology and results can be used. The usefulness of synthetic scenarios is further discussed later in this article.

Discussion

The potential impacts of changes in temperature and precipitation, the potential adaptation and mitigation measures in the region’s managed ecosystem and water resources are discussed in this section. Further, the changes observed in earlier studies and the application of the synthetic scenarios that can be generated from this study in general are also discussed.

Figure 5. Ensemble mean monsoon temperatures (K) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.
Figure 6. Ensemble mean non-monsoon temperatures (K) for 15 countries in south Asia. The red lines represent the mean (central), 10th percentile (lower), and 90th percentile (upper) variability among GCMs in RCP8.5. The blue line represents the mean, and the blue bands span the 10th percentile (lower) and 90th percentile (upper) variability among GCMs in RCP4.5. For Sri Lanka, only the mean values are plotted because Sri Lanka had fewer than five grid square values.

GCM Performance and Convergence

The GCMs were evaluated by (1) model performance (testing the GCMs’ ability to simulate “present climate” of the variable of interest), (2) model convergence (identifying groups of models that agree on future climate changes), and (3) combining both performance and convergence. However, it should be noted that neither good performance across an arbitrary suite of measures of observed climate nor agreement in output across a collection of models provides a rigorous basis for assessing the accuracy of a future prediction. The performance of GCMs at the regional scale remains variable (Bollasina and Nigam, 2009; Gleckler et al., 2008; Randall et al., 2007). Model accuracy differs with region and the type of variable simulated (Errasti et al., 2011).

There are a number of approaches for evaluating the GCMs, and there are inherent problems with each of these approaches (Anandhi et al., 2011a). Some studies have

used a single GCM (Ashofteh et al., 2015; Fortier and Mailhot, 2015) or an ensemble mean from multiple climate models (Benestad, 2003; Smith et al., 2009; Tebaldi and Knutti, 2007). Using each approach, only one “likely” outcome was produced, but no information regarding a range of possible outcomes. An alternative approach is to use a subset of climate models (Anandhi et al., 2011b; Elhakeem et al., 2015; Islam and Gan, 2014; Perkins and Pitman, 2009; Perkins et al., 2007; Singh et al., 2015). The question here is how to choose an appropriate subset of GCMs. Studies choose a subset of GCMs based on results from earlier studies on the performance of the GCMs in simulating the local climate, future climate, or both (Ashofteh et al., 2015; Elhakeem et al., 2015; Singh et al., 2015). The number of GCMs selected in these studies ranges from one to nine. Other studies have used results from all the individual climate models, which allows researchers to estimate

Figure 7. Changes in ensemble mean precipitation in the region for two time periods (2006-2050 and 2051-2099) and three categories: annual (entire year), monsoon (June to October), and non-monsoon (November to May). The ensemble mean was estimated by aggregating the changes from 35 GCMs participating in the CMIP5. The numbers in the figures represent the average change in the country obtained by aggregating the changes for each category and time period. The black lines in the figures indicate political boundaries. These boundaries are only indicative and may not be correct.
Figure 8. Changes in ensemble mean temperature in the region for two time periods (2006-2050 and 2051-2099) and three categories: annual (entire year), monsoon (June to October), and non-monsoon (November to May). The ensemble mean was estimated by aggregating the changes from 35 GCMs participating in the CMIP5. The numbers in the figures represent the average change in the country obtained by aggregating the changes for each category and time period. The gray lines in the figures indicate political boundaries. These boundaries are only indicative and may not be correct.

a range of possible outcomes for any particular climate change scenario (Giacomoni and Berglund, 2015; Kollat et al., 2012; Miller et al., 2012). The model performance differs with region and with the type of variable simulated (Anandhi et al., 2011b; Anandhi and Nanjundiah, 2015; Errasti et al., 2011). Eliminating GCMs with a large bias (by choosing a subset) can create the possibility of losing many GCMs if different GCMs have large bias in different regions and across precipitation and temperature. Additionally, a number of evaluation methods (such as skill score, correlation coefficients, mean, median, standard deviation, anomalies, root mean square error, bias, extreme indices, empirical orthogonal functions, and principal component analysis) have been used in the literature (Anandhi and Nanjundiah, 2015; Errasti et al., 2011; Frei et al., 2003; Meehl et al., 2007a, 2007b; Perkins et al., 2007). Evaluating GCMs using multiple methods can result in a different GCMs showing large bias and can also create the possibility of losing multiple GCMs. These approaches have inherent problems that are discussed by Knutti et al. (2010), Knutti (2010), Raisanen (2007), Tebaldi and Knutti (2007), and Weigel et al. (2010). A good review of the available methods and details of earlier studies can be found in Johnson and Sharma (2009), Anandhi and Nanjundiah (2015), and in table 8 of Errasti et al. (2011). In this study, all GCMs were chosen to develop synthetic scenarios because they gave a range of plausible futures for the stakeholders and are useful for exploring system sensitivity prior to the application of more credible, model-based scenarios. Using the whole suite of GCMs also provides preliminary information for selecting wider alternatives to reduce vulnerabilities in managed ecosystems (agricultural and urban) and water resources.

(b)
(a)
(c)
Figure 9. Scenario funnels for (a) annual, (b) monsoon, and (c) non-monsoon precipitation and (d) temperature. The percentage change values in the funnel mouths were extracted from figures 7 and 8 and represent the variability in the percent change observed in south Asia. Symbols differentiate scenarios (circles for RCP4.5; triangles for RCP8.5) and time periods (large symbols for 2006-2050; small symbols for 2050-2099).

Potential Impacts of Temperature and Precipitation Change on Managed Ecosystems and Water Resources

All CMIP5 model simulations of monthly temperature indicated a consistent increase in south Asia. However, the precipitation results indicated that some models showed an increase while others showed a decrease in south Asia. These changes in precipitation and temperature have multiple potential impacts on managed ecosystems (agricultural and urban) and water resources, including:

The Need for Synthetic Scenarios in Developing Management Decisions

Effective adaptation depends on an understanding of projected climatic changes at geographic and temporal scales appropriate for the needed response (Anandhi, 2016; Anandhi et al., 2016). Stakeholders (e.g., decision-makers, producers, and water managers) are becoming more interested in obtaining understandable information about climate change risks in ecological and hydrological systems. They are expressing the need for climate information that can support adaptation-related decision-making, provide straightforward estimations of variability, and be tailored to specific user groups (Mastrandrea et al., 2010), so stakeholders can make informed decisions that lead to economic development. Addressing the consequences of climate change for adaptive and/or mitigative management is now high on the list of priorities for funding agencies (Anandhi, 2015), particularly in developed countries. However, funding is a significant challenge in many Asian cities (Porse, 2013) and developing countries. In many Asian cities, in a time of shrinking public budgets, increasing urban population, and a changing variable climate, water managers must weigh funding among different goals (Porse, 2013).

Our results on changing climate conditions are useful to planners and policy-makers who use rainwater harvesting. This is an increasing practice, with emphasis on harvesting for domestic as well as supplemental irrigation in water-stressed countries such as India and China (Devineni et al., 2013). Managers and policy-makers need to shift from the traditional, linear approach to an adaptive, participatory, and integrated approach. Depending on the degree of shift, the degree of adaptation could vary from incremental to transformational using demand management and supply, and using non-traditional water resources and the concept of fit-for-purpose and decentralization (Brown and Farrelly, 2009). In general, a critical challenge in supporting climate change adaptation is improving the linkage between climate impacts, vulnerability research, public and private planning, and management decisions (Mastrandrea et al., 2010).

The Usefulness of Synthetic Scenarios for Potential Stakeholders

Groundwater use buffers the variability in precipitation for agriculture, and the south Asia region faces groundwater depletion and chronic water stress (Devineni et al., 2013). Rodell et al. (2009) estimated that about 109 km3 of groundwater was extracted in northern India alone between 2002 and 2009, clearly an unsustainable rate of withdrawal (this is about twice India’s surface water reservoir capacity). The two countries with the highest irrigated area in the world are in Asia: India with 57 Mha, followed by China 54 Mha (Scanlon et al., 2012). Asia has 70% of its total area equipped for irrigation (AEI), with south Asia having a subregional-scale AEI as a percentage of cultivated land of 42% (Siebert et al., 2010). Irrigation buffers the changing precipitation and temperature patterns in the region. China is facing a water resources crisis with growing concerns about the reliable supply of water for agricultural, industrial, and domestic needs. In China, water resources per capita are projected to decline from the current 628 m3 to 508 m3 in 2030 due to a changing climate and other stresses (Allan et al., 2013). The simultaneous effects of agricultural growth, industrialization, and urbanization coupled with declining surface and groundwater quantity, inefficiencies in water use practices, and rising concerns about the impacts of climate change on water supply patterns are crucial problems faced by the water sector. This leads to cross-sectoral, intrastate, and interstate water disputes (Devineni et al., 2013).

In these situations, scenario funnels will be useful to water managers for developing synthetic scenarios for preliminary climate change impact assessments, especially under resource constraint conditions. Synthetic scenarios have been used in many developed countries in preliminary analyses for assessing the vulnerability of existing stormwater management systems to a changing climate and for developing adaptation strategies to reduce the negative impacts of the changes (Anandhi et al., 2011a). Water managers can use this information for management of water supply and sewage systems under changing climate conditions by running stormwater models with various synthetic scenarios. Synthetic scenarios can be used in the management of stormwater conveyance systems by changing the stormwater flow pathways and the variability of various components of the water cycle (i.e., infiltration, evapotranspiration, runoff, groundwater, and base flow; Burns et al., 2012).

Synthetic scenarios generated from this study can be used by crop breeders, agronomists, and other researchers in adaptation studies by being input into crop models: (1) to test the robustness of farm adaptations to future climate scenarios; (2) as tools in farm organizations to build farm capacity, minimize risk, and empower farmers; (3) to link with other models to widen the scope of potential impacts, adaptations, and constraints; (4) to probe the interactions of cropping systems with other systems; and (5) to evaluate various indicators of resilience. This quantitative information from the percent change and synthetic scenarios can be used for developing adaptation strategies at different levels, i.e., incremental strategies and/or transformational adaptations. Many incremental strategies (e.g., continued technological advancements) are available to help agriculture adapt to the temperature and precipitation changes projected for the next 25 years (Melillo et al., 2014). The results from this study (changes in temperature and precipitation) can provide quantitative information for crop breeders and researchers to develop new crop varieties (systems adaptation) that are conducive to these changes. For example, irrigation can be expanded through micro-irrigation projects that capture water in small reservoirs or cisterns for very local application, by using other rainfall harvesting techniques, or by using better soil management practices (e.g., reduced tillage, no tillage, cover crops, and rotations) to build soil resilience, increase infiltration, and reduce soil temperatures and evaporation. In an agro-ecosystem, managers, crop advisors, and producers can use this information for selecting hybrids from existing crop varieties (e.g., early or late maturing varieties; incremental strategy), or they can decide to pursue a different land use (transformational adaptation). In addition, climate adaptation will be made largely by farmers, so considering local farmers as a key source of expert information is important to developing the system.

Through development of the scenario funnel, providing information for expanding synthetic scenarios, and discussing the potential impacts of these changes, this study has attempted to improve the linkage between climate impacts and adaptation research. Assessing the performance and convergence of precipitation and temperature simulations from CMIP5 models in south Asia, as well as using the synthetic scenarios for developing adaptation strategies for managed ecosystems and water resources, were deferred for future study.

Conclusions

South Asia supports one quarter of the global population, half of whom live below the poverty line. Given the limited resources in developing countries in this region, the objective of our study was to empirically explore climate change in south Asia up to the year 2099 using monthly simulations from 35 global climate models (GCMs) participating in CMIP5 for two future emission scenarios (RCP4.5 and RCP8.5). The GCMs provide a wide range of potential climate change outcomes. This study was carried out using a three-step procedure by first calculating the mean annual, monsoon, and non-monsoon precipitation and temperatures, followed by estimation of change factors, and finally developing the scenario funnels and synthetic scenarios. This methodology was applied for the entire south Asia region; however, the percent change information generated at 1.5° grid scale can be used to generate scenarios at finer spatial scales.

Results showed a high variability in the future change in precipitation (-23% to 52%, maximum in the non-monsoon category) and a more consistent change in temperature (0.8% to 2.1%) in the region. The temperature in the region was shown to consistently increase, especially in the Himalayan region, which can have serious impacts causing a fast retreat of glaciers, leading to floods in the region. Rivers, which are currently perennial due to snowmelt in the pre-monsoon season, could become seasonal, and this could pose significant challenges for surface water management and even drinking water availability in the summer season. The increasing high rainfall can have impacts on agricultural production and urban water management. Increased propensity of heavy rainfall events, as has been noticed by observational studies, could lead to more frequent occurrence of severe urban flooding, similar to the flooding experienced by Mumbai in July 2005.

The increasing temperatures, extreme climate, and high variability can have important implications in the region by compounding groundwater depletion, changing irrigation patterns (to buffer climate variability), decreasing agricultural production, and causing chronic water stress in the region, and groundwater withdrawal (which is currently unsustainable) could worsen. The increasing high rainfall can also have impacts on agricultural production and urban water management. The maps and scenario funnels generated from a wide range of GCMs would complement existing studies that lack the necessary geographical resolution and wide range of GCMs for assessing regional impacts and prioritizing adaptation policies. In the future, scientists, managers, breeders, policy-makers, and planners can develop synthetic scenarios from our results to explore the potential impacts in a more detailed manner to better cope with a changing climate. Our results can help them shift from the traditional, linear, “old world” approach to an adaptive, participatory, and integrated approach.

Acknowledgements

This material was based on work partially supported by the USDA-NIFA Evans-Allen Project (Grant No. 11979180/ 2016-01711). Dr. Anandhi would like to acknowledge Dr. Magee and Mr. Bentley for support in editing the manuscript.

References

Ainsworth, E. A., & Ort, D. R. (2010). How do we improve crop production in a warming world? Plant Physiol., 154(2), 526-530. http://dx.doi.org/10.1104/pp.110.161349

Alig, R. J., Kline, J. D., & Lichtenstein, M. (2004). Urbanization on the U.S. landscape: Looking ahead in the 21st century. Landscape Urban Plan., 69(2-3), 219-234. http://dx.doi.org/10.1016/j.landurbplan.2003.07.004

Allan, C., Xia, J., & Pahl-Wostl, C. (2013). Climate change and water security: Challenges for adaptive water management. Current Opinion in Environ. Sustain. 5(6), 625-632. http://dx.doi.org/10.1016/j.cosust.2013.09.004

Anandhi, A. (2015). Growing degree days: Ecosystem indicator for changing diurnal temperatures and their impact on corn growth stages in Kansas. Ecol. Indicators, 61(2), 149-158. http://dx.doi.org/10.1016/j.ecolind.2015.08.023

Anandhi, A. (2016). Cista-A: Conceptual model using indicators selected by systems thinking for adaptation strategies in a changing climate: Case study in agro-ecosystems. Ecol. Model., (accepted).

Anandhi, A., & Blocksome, C. (2017). Developing adaptation strategies using an agroecosystem indicator: Variability in crop failure temperatures. Ecol. Indicators, (accepted).

Anandhi, A., & Nanjundiah, R. S. (2015). Performance evaluation of AR4 climate models in simulating daily precipitation over the Indian region using skill scores. Theor. Appl. Climatol., 119(3-4), 551-566. https://doi.org/10.1007/s00704-013-1043-5

Anandhi, A., Frei, A., Pierson, D. C., Schneiderman, E. M., Zion, M. S., Lounsbury, D., & Matonse, A. H. (2011a). Examination of change factor methodologies for climate change impact assessment. Water Resour. Res., 47(3), W03501. https://doi.org/10.1029/2010WR009104

Anandhi, A., Frei, A., Pradhanang, S. M., Zion, M. S., Pierson, D. C., & Schneiderman, E. M. (2011b). AR4 climate model performance in simulating snow water equivalent over Catskill Mountain watersheds, New York, USA. Hydrol. Proc., 25(21), 3302-3311. https://doi.org/10.1002/hyp.8230

Anandhi, A., Perumal, S., Gowda, P. H., Knapp, M., Hutchinson, S., Harrington Jr, J., ... Rice, C. W. (2013a). Long-term spatial and temporal trends in frost indices in Kansas, USA. Clim. Change, 120(1-2), 169-181. https://doi.org/10.1007/s10584-013-0794-4

Anandhi, A., Srinivas, V., Kumar, D. N., & Nanjundiah, R. S. (2009). Role of predictors in downscaling surface temperature to river basin in India for IPCC SRES scenarios using support vector machine. Intl. J. Climatol., 29(4), 583-603. https://doi.org/10.1002/joc.1719

Anandhi, A., Srinivas, V. V., Nanjundiah, R. S., & Nagesh Kumar, D. (2008). Downscaling precipitation to river basin in India for IPCC SRES scenarios using support vector machine. Intl. J. Climatol., 28(3), 401-420. https://doi.org/10.1002/joc.1529

Anandhi, A., Steiner, J. L., & Bailey, N. (2016). A systems approach to assess the exposure of agricultural production to climate change and variability. Clim. Change, 136(3), 647-659. https://doi.org/10.1007/s10584-016-1636-y

Anandhi, A., Zion, M. S., Gowda, P. H., Pierson, D. C., Lounsbury, D., & Frei, A. (2013b). Past and future changes in frost day indices in Catskill Mountain region of New York. Hydrol. Proc., 27(21), 3094-3104. https://doi.org/10.1002/hyp.9937

Ashofteh, P., Haddad, O., & Loáiciga, H. (2015). Evaluation of climatic change impacts on multiobjective reservoir operation with multiobjective genetic programming. J. Water Res. Plan. Mgmt., 141(11), 04015030. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000540

Benestad, R. E. (2003). What can present climate models tell us about climate change? Clim. Change, 59(3), 311-331. https://doi.org/10.1023/A:1024876611259

Berkhout, F., Hertin, J., & Jordan, A. (2002). Socio-economic futures in climate change impact assessment: Using scenarios as “learning machines”. Global Environ. Change, 12(2), 83-95. https://doi.org/10.1016/S0959-3780(02)00006-7

Bollasina, M., & Nigam, S. (2009). Indian Ocean SST, evaporation, and precipitation during the south Asian summer monsoon in IPCC-AR4 coupled simulations. Clim. Dynamics, 33(7), 1017-1032. https://doi.org/10.1007/s00382-008-0477-4

Brown, R., & Farrelly, M. (2009). Delivering sustainable urban water management: A review of the hurdles we face. Water Sci. Tech., 59(5), 839-846. https://doi.org/10.2166/wst.2009.028

Burns, M. J., Fletcher, T. D., Walsh, C. J., Ladson, A. R., & Hatt, B. E. (2012). Hydrologic shortcomings of conventional urban stormwater management and opportunities for reform. Landscape Urban Plan., 105(3), 230-240. https://doi.org/10.1016/j.landurbplan.2011.12.012

Carter, T. R., Parry, M. L., Harasawa, H., & Nishioka, S. (1994). IPCC technical guidelines for assessing climate change impacts and adaptations. London, U.K.: University College London.

Chen, H. (2013). Projected change in extreme rainfall events in China by the end of the 21st century using CMIP5 models. Chinese Sci. Bull., 58(12), 1462-1472. https://doi.org/10.1007/s11434-012-5612-2

Chong-Hai, X., & Ying, X. (2012). The projection of temperature and precipitation over China under RCP scenarios using a CMIP5 multi-model ensemble. Atmos. Oceanic Sci. Letters, 5(6), 527-533. http://dx.doi.org/10.1080/16742834.2012.11447042

Devineni, N., Perveen, S., & Lall, U. (2013). Assessing chronic and climate-induced water risk through spatially distributed cumulative deficit measures: A new picture of water sustainability in India. Water Resources Res., 49(4), 2135-2145. https://doi.org/10.1002/wrcr.20184

Elhakeem, A., Elshorbagy, W., AlNaser, H., & Dominguez, F. (2015). Downscaling global circulation model projections of climate change for the United Arab Emirates. J. Water Resources Plan. Mgmt., 141(9), 04015007. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000507

Errasti, I., Ezcurra, A., Sáenz, J., & Ibarra-Berastegi, G. (2011). Validation of IPCC AR4 models over the Iberian Peninsula. Theor. Appl. Climatol., 103(1), 61-79. https://doi.org/10.1007/s00704-010-0282-y

Fortier, C., & Mailhot, A. (2015). Climate change impact on combined sewer overflows. J. Water Resources Plan. Mgmt., 141(5), 04014073. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000468

Frei, A., Miller, J. A., & Robinson, D. A. (2003). Improved simulations of snow extent in the second phase of the Atmospheric Model Intercomparison Project (AMIP-2). J. Geophys. Res., 108(D12), 4369-4386. https://doi.org/10.1029/2002JD003030

Gerland, P., Raftery, A. E., Ševcíková, H., Li, N., Gu, D., Spoorenberg, T., ...Wilmoth, J. (2014). World population stabilization unlikely this century. Science, 346(6206), 234-237. https://doi.org/10.1126/science.1257469

Giacomoni, M., & Berglund, E. (2015). Complex adaptive modeling framework for evaluating adaptive demand management for urban water resources sustainability. J. Water Resources Plan. Mgmt., 151(11), 04015024. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000543

Gleckler, P. J., Taylor, K. E., & Dutriaux, C. (2008). Performance metrics for climate models. J. Geophys. Res., 113(D6), D06104. https://doi.org/10.1029/2007JD008972

Grimm, N. B., Faeth, S. H., Golubiewski, N. E., Redman, C. L., Wu, J., Bai, X., & Briggs, J. M. (2008). Global change and the ecology of cities. Science, 319(5864), 756-760. http://dx.doi.org/10.1126/science.1150195

Horlings, L., & Marsden, T. K. (2011). Towards the real green revolution? Exploring the conceptual dimensions of a new ecological modernisation of agriculture that could “feed the world”. Global Environ. Change, 21(2), 441-452. https://doi.org/10.1016/j.gloenvcha.2011.01.004

Islam, Z., & Gan, T. (2014). Effects of climate change on the surface-water management of the south Saskatchewan River basin. J. Water Resources Plan. Mgmt., 140(3), 332-342. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000326

Jayaraman, T., & Murari, K. (2014). Climate change and agriculture: Current and future trends, and implications for India. Rev. Agrarian Studies, 4(1), 1-49.

Johnson, F., & Sharma, A. (2009). Measurement of GCM skill in predicting variables relevant for hydroclimatological assessments. J. Clim., 22, 4373-4382. https://doi.org/10.1175/2009JCLI2681.1

Knutti, R. (2010). The end of model democracy? Clim. Change, 102(3), 395-404. https://doi.org/10.1007/s10584-010-9800-2

Knutti, R., Furrer, R., Tebaldi, C., Cermak, J., & Meehl, G. A. (2010). Challenges in combining projections from multiple climate models. J. Clim., 23(10), 2739-2758. https://doi.org/10.1175/2009JCLI3361.1

Kollat, J., Kasprzyk, J., Thomas, W., Jr.,, Miller, A., & Divoky, D. (2012). Estimating the impacts of climate change and population growth on flood discharges in the United States. J. Water Resources Plan. Mgmt., 138(5), 442-452. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000233

Kosow, H., & Gaßner, R. (2008). Methods of future and scenario analysis. Bonn, Germany: German Development Institute (DIE). https://doi.org/10.1007/s10584-010-9800-2

Lansigan, F., De Los Santos, W., & Coladilla, J. (2000). Agronomic impacts of climate variability on rice production in the Philippines. Agric. Ecosys. Environ., 82(1), 129-137. https://doi.org/10.1016/S0167-8809(00)00222-X

Lavalle, C., Micale, F., Houston, T. D., Camia, A., Hiederer, R., Lazar, C., ... Genovese, G. (2009). Climate change in Europe: 3. Impact on agriculture and forestry: A review. Agron. Sustain. Devel., 29(3), 433-446. https://doi.org/10.1051/agro/2008068

Leggett, J., Pepper, W. J., Swart, R. J., Edmonds, J., Meira Filho, L., Mintzer, I., & Wang, M. (1992). Emissions scenarios for the IPCC: An update. In J. T. Houghton, B. A. Callander, and S. K. Varney (Eds.), Climate change 1992: Supplementary report to the IPCC scientific assessment (pp. 71-95). Cambridge, U.K.: Cambridge University Press.

Liu, Y., Mahmoud, M., Hartmann, H., Stewart, S., Wagener, T., Semmens, D., ... White, D. (2008). Formal scenario development for environmental impact assessment studies. Paper 77. Washington, DC: USEPA.

Mahmoud, M., Liu, Y., Hartmann, H., Stewart, S., Wagener, T., Semmens, D., ... Dominguez, F. (2009). A formal framework for scenario development in support of environmental decision-making. Environ. Model. Software, 24(7), 798-808. https://doi.org/10.1016/j.envsoft.2008.11.010

Mastrandrea, M., Heller, N., Root, T., & Schneider, S. (2010). Bridging the gap: Linking climate impacts research with adaptation planning and management. Clim. Change, 100(1), 87-101. https://doi.org/10.1007/s10584-010-9827-4

Matonse, A. H., Pierson, D. C., Frei, A., Zion, M. S., Anandhi, A., Schneiderman, E., & Wright, B. (2013). Investigating the impact of climate change on New York City’s primary water supply. Clim. Change, 116(3-4), 437-456. https://doi.org/10.1007/s10584-012-0515-4

McCarthy, J. J. (2001). Climate change 2001: Impacts, adaptation, and vulnerability. Contribution of Working Group II to the Third Assessment Report of the IPCC. Cambridge, U.K.: Cambridge University Press.

Meehl, G., Arblaster, J., & Tebaldi, C. (2007a). Contributions of natural and anthropogenic forcing to changes in temperature extremes over the United States. Geophys. Res. Letters, 34(19), L19709. https://doi.org/10.1029/2007gl030948

Meehl, G., Covey, C., Delworth, T., Latif, M., McAvaney, B., Mitchell, J., Stouffer, R., & Taylor, K. (2007b). The WCRP CMIP3 multimodel dataset: A new era in climate change research. Bull. American Meteorol. Soc., 88(9), 1383-1394. https://doi.org/10.1175/BAMS-88-9-1383

Melillo, J. M., Richmond, T., & Yohe, G. W. (2014). Climate change impacts in the United States: The third national climate assessment. Washington, DC: U.S. Global Change Research Program. https://doi.org/10.7930/J0Z31WJ2

Menon, A., Levermann, A., Schewe, J., Lehmann, J., & Frieler, K. (2013). Consistent increase in Indian monsoon rainfall and its variability across CMIP5 models. Earth System Dynamics, 4(2), 287-300. https://doi.org/10.5194/esd-4-287-2013

Miller, W., Butler, R., Piechota, T., Prairie, J., Grantz, K., & DeRosa, G. (2012). Water management decisions using multiple hydrologic models within the San Juan River basin under changing climate conditions. J. Water Resources Plan. Mgmt., 138(5), 412-420. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000237

Mishra, V., Kumar, D., Ganguly, A. R., Sanjay, J., Mujumdar, M., Krishnan, R., & Shah, R. D. (2014). Reliability of regional and global climate models to simulate precipitation extremes over India. J. Geophys. Res. Atmos., 119(15), 9301-9323. https://doi.org/10.1002/2014jd021636

Mishra, V., & Lilhare, R. (2016). Hydrologic sensitivity of Indian sub-continental river basins to climate change. Global Planet. Change, 139, 78-96. https://doi.org/10.1016/j.gloplacha.2016.01.003

Mukundan, R., Pradhanang, S. M., Schneiderman, E. M., Pierson, D. C., Anandhi, A., Zion, M. S., ... Steenhuis, T. S. (2013). Suspended sediment source areas and future climate impact on soil erosion and sediment yield in a New York City water supply watershed, USA. Geomorphol., 183, 110-119. https://doi.org/10.1016/j.geomorph.2012.06.021

Murugan, M., Shetty, P., Ravi, R., Anandhi, A., & Rajkumar, A. (2012). Climate change and crop yields in the Indian Cardamom hills, 1978-2007 CE. Clim. Change, 110(3-4), 737-753. https://doi.org/10.1007/s10584-011-0115-8

Nakicenovic, N., Alcamo, J., Davis, G., de Vries, B., Fenhann, J., Gaffin, S., ... Kram, T. (2000). Special report on emissions scenarios. A Special Report of Working Group III of the IPCC. Cambridge, U.K.: Cambridge University Press.

Niu, X., Wang, S., Tang, J., Lee, D. K., Gutowski, W., Dairaku, K., ... Sasaki, H. (2015). Projection of Indian summer monsoon climate in 2041-2060 by multiregional and global climate models. J. Geophys. Res. Atmos., 120(5), 1776-1793. https://doi.org/10.1002/2014jd022620

Perkins, S., & Pitman, A. (2009). Do weak AR4 models bias projections of future climate changes over Australia? Clim. Change, 93(3), 527-558. https://doi.org/10.1007/s10584-008-9502-1

Perkins, S. E., Pitman, A. J., Holbrook, N. J., & McAneney, J. (2007). Evaluation of the AR4 climate models’ simulated daily maximum temperature, minimum temperature, and precipitation over Australia using probability density functions. J. Clim., 20(17). https://doi.org/10.1175/JCLI4253.1

Porse, E. C. (2013). Stormwater governance and future cities. Water, 5(1), 29-52. https://doi.org/10.3390/w5010029

Pradhanang, S. M., Mukundan, R., Schneiderman, E. M., Zion, M. S., Anandhi, A., Pierson, D. C., ... Steenhuis, T. S. (2013). Streamflow responses to climate change: Analysis of hydrologic indicators in a New York City water supply watershed. JAWRA, 49(6), 308-1326. https://doi.org/10.1111/jawr.12086

Raisanen, J. (2007). How reliable are climate models? Tellus A, 59(1), 2-29. https://doi.org/10.1111/j.1600-0870.2006.00211.x

Raju, K. S., Sonali, P., & Kumar, D. N. (2016). Ranking of CMIP5-based global climate models for India using compromise programming. Theor. Appl. Climatol., 2016, 1-12. https://doi.org/10.1007/s00704-015-1721-6

Ramesh, K., & Goswami, P. (2014). Assessing reliability of regional climate projections: The case of Indian monsoon. Sci. Reports, 4, 4071. https://doi.org/10.1038/srep04071

Randall, D. A., Wood, R. A., Bony, S., Colman, R., Fichefet, T., Fyfe, J., ... Taylor, K. E. (2007). Climate models and their evaluation. Cambridge, U.K.: Cambridge University Press.

Rodell, M., Velicogna, I., & Famiglietti, J. S. (2009). Satellite-based estimates of groundwater depletion in India. Nature, 460(7258), 999-1002. https://doi.org/10.1038/nature08238

Sabeerali, C., Rao, S. A., Dhakate, A., Salunke, K., & Goswami, B. (2015). Why ensemble mean projection of south Asian monsoon rainfall by CMIP5 models is not reliable? Clim. Dyn., 45(1-2), 161-174. https://doi.org/10.1007/s00382-014-2269-3

Saha, A., Ghosh, S., Sahana, A., & Rao, E. (2014). Failure of CMIP5 climate models in simulating post-1950 decreasing trend of Indian monsoon. Geophys. Res. Letters, 41(20), 7323-7330. https://doi.org/10.1002/2014GL061573

Sandeep, S., & Ajayamohan, R. (2015). Poleward shift in Indian summer monsoon low-level jetstream under global warming. Clim. Dyn., 45(1-2), 337-351. https://doi.org/10.1007/s00382-014-2261-y

Scanlon, B. R., Faunt, C. C., Longuevergne, L., Reedy, R. C., Alley, W. M., McGuire, V. L., & McMahon, P. B. (2012). Groundwater depletion and sustainability of irrigation in the U.S. High Plains and Central Valley. Proc. Natl. Acad. Sci., 109(24), 9320-9325. https://doi.org/10.1073/pnas.1200311109

Schoof, J. T. (2013). Statistical downscaling in climatology. Geography Compass, 7(4), 249-265. https://doi.org/10.1111/gec3.12036

Seth, A., Rauscher, S. A., Biasutti, M., Giannini, A., Camargo, S. J., & Rojas, M. (2013). CMIP5 projected changes in the annual cycle of precipitation in monsoon regions. J. Clim., 26(19), 7328-7351. https://doi.org/10.1175/JCLI-D-12-00726.1

Shrestha, A. B., Wake, C. P., Mayewski, P. A., & Dibb, J. E. (1999). Maximum temperature trends in the Himalaya and its vicinity: An analysis based on temperature records from Nepal for the period 1971-1994. J. Clim., 12(9), 2775-2786. https://doi.org/10.1175/1520-0442(1999)012<2775:MTTITH>2.0.CO;2

Siebert, S., Burke, J., Faures, J.-M., Frenken, K., Hoogeveen, J., Döll, P., & Portmann, F. T. (2010). Groundwater use for irrigation: A global inventory. Hydrol. Earth Syst. Sci., 14(10), 1863-1880. https://doi.org/10.5194/hess-14-1863-2010

Singh, H., Sinha, T., & Sankarasubramanian, A. (2015). Impacts of near-term climate change and population growth on within-year reservoir systems. J. Water Resources Plan. Mgmt., 141(6), 04014078. https://doi.org/10.1061/(ASCE)WR.1943-5452.0000474

Smith, R. L., Tebaldi C., Nychka D., & Mearns L.O. (2009). Bayesian modeling of uncertainty in ensembles of climate models. J. Am. Stat. Assoc., 104(485), 97-116. https://doi.org/10.1198/jasa.2009.0007

Sonali, P., & Kumar, D. N. (2016). Detection and attribution of seasonal temperature changes in India with climate models in the CMIP5 archive. J. Water Clim. Change, 7(1), 83-102. http://dx.doi.org/10.2166/wcc.2015.072

Sooraj, K., Terray, P., & Mujumdar, M. (2015). Global warming and the weakening of the Asian summer monsoon circulation: Assessments from the CMIP5 models. Clim. Dyn., 45(1-2), 233-252. https://doi.org/10.1007/s00382-014-2257-7

Sun, Q., Miao, C., & Duan, Q. (2015). Projected changes in temperature and precipitation in ten river basins over China in 21st century. Intl. J. Climatol., 35(6), 1125-1141. https://doi.org/10.1002/joc.4043

Taylor, K. E., Stouffer, R. J., & Meehl, G. A. (2012). An overview of CMIP5 and the experiment design. Bull. American Meteor. Soc., 93(4), 485. https://doi.org/10.1175/BAMS-D-11-00094.1

Tebaldi, C., & Knutti, R. (2007). The use of the multi-model ensemble in probabilistic climate projections. Phil. Trans. Royal Soc. A, 365(1857), 2053-2075. https://doi.org/10.1098/rsta.2007.2076

Timpe, C., & Scheepers, M. (2003). A look into the future: Scenarios for distributed generation in Europe. Petten, The Netherlands: Energy Research Centre of the Netherlands (ECN). Retrieved from http://www.ecn.nl/docs/library/report/2004/c04012.pdf

Tschakert, P., & Dietrich, K. A. (2010). Anticipatory learning for climate change adaptation and resilience. Ecol. Soc., 15(2), article 11. https://doi.org/10.5751/ES-03335-150211

Tscharntke, T., Clough, Y., Wanger, T. C., Jackson, L., Motzke, I., Perfecto, I., ... Whitbread, A. (2012). Global food security, biodiversity conservation, and the future of agricultural intensification. Biol. Conserv., 151(1), 53-59. https://doi.org/10.1016/j.biocon.2012.01.068

Ul Hasson, S., Pascale, S., Lucarini, V., & Böhner, J. (2016). Seasonal cycle of precipitation over major river basins in south and southeast Asia: A review of the CMIP5 climate models data for present climate and future climate projections. Atmos. Res., 180, 42-63. https://doi.org/10.1016/j.atmosres.2016.05.008

van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., ... Rose, S. K. (2011). The representative concentration pathways: An overview. Clim. Change, 109, 5-31. https://doi.org/10.1007/s10584-011-0148-z

Wang, B., Yim, S.-Y., Lee, J.-Y., Liu, J., & Ha, K.-J. (2014). Future change of Asian-Australian monsoon under RCP4.5 anthropogenic warming scenario. Clim. Dyn., 42(1-2), 83-100. https://doi.org/10.1007/s00382-013-1769-x

Weigel, A. P., Knutti, R., Liniger, M. A., & Appenzeller, C. (2010). Risks of model weighting in multimodel climate projections. J. Clim., 23(15), 4175-4191. https://doi.org/10.1175/2010JCLI3594.1

Zahid, M., & Iqbal, W. (2015). Multi-model cropping seasons projections over Pakistan under representative concentration pathways. Model. Earth Syst. Environ., 1(3), 1-12. https://doi.org/10.1007/s40808-015-0008-3

Zhang, L., Dawes, W., & Walker, G. (2001). Response of mean annual evapotranspiration to vegetation changes at catchment scale. Water Resources Res., 37(3), 701-708. https://doi.org/10.1029/2000WR900325