Enhanced Growth Rate and Reduced Water Demand of Crop Due to Climate Change in the Eastern Mediterranean Region

Abstract

The specific objectives of this study were to: (a) test the reliability of a regional climate model (RCM) as a tool for climate change projection in the Eastern Mediterranean, (b) compare the observed yield variables of maize and wheat in the region with results of two crop models, (c) compare the models DSSAT and SWAP and (d) use DSSAT and SWAP to generate future productivity of wheat and maize under the A2 global warming scenario. Reference evapotranspiration was highly correlated with the models with average r² = 0.98 and a unit slope. The two models accurately predicted observed dry mass production (DMP) and leaf area index (LAI) of wheat and maize. The correlations strengthen the legitimacy of DSSAT, SWAP and RCM to serve as predicting models for future climate change on a regional scale.

A simulation was carried out to describe the effects of climate change on crop growth and irrigation water requirements for a wheat-maize-wheat cropping sequence. Climate change scenarios were projected using data of three general circulation models (CGCM2, ECHAM4 and MRI) for the period of 1990–2100 and one RCM for the period of 2070–2079. Daily RCM data were consistent with actual meteorological data in the region and therefore were used for computations of present and future water balance and crop development. Predictions derived from the models about changes in irrigation and crop growth covered the period of 2070–2079 relative to a baseline period of 1994–2003. The effects of climate change on wheat and maize water requirements and yields were predicted using the detailed crop growth subroutine of the DSSAT (Decision Support System for Agrotechnology Transfer) and SWAP (Soil-Water-Atmosphere-Plant) models. Precipitation was projected to decrease by about 163, 163 and 105 mm during the period of 1990–2100 under the A2 scenario of the CGCM2, ECHAM4 and MRI models respectively (an average of about 1.3 mm/year). The models projected a temperature rise of 4.3, 5.3 and 3.1 °C, by the year 2100. An increase in temperature may result in a higher evaporative demand of the atmosphere under combined doubling CO₂ concentration and temperature rise by about 2 °C for the period of 2070–2079. The temperature rise accelerated crop development and shortened the growing period by a maximum of thirteen days for wheat and nine days for maize during the period 2070–2079. When yield and available water (rain + applied irrigation) were normalised by extension of the growing period with respect to the baselines years, DMP of maize increased by 1–3 ton ha⁻¹ and that of wheat by 3–4 ton ha⁻¹. Consequently, water use efficiency (WUE) increased for both crops. It was concluded, therefore, that the effect of increased temperature and doubling CO₂ on agro-productivity may be positive. This positive effect can be explained if elevated temperature meets the optimal level of a crop response to temperature. Effects of elevated CO₂ on crop tolerance to water stress may counteract the expected negative effects of rising temperature. Increased atmospheric CO₂ levels have important physiological effects on crops such as the increase in photosynthetic rate, which is associated with higher yield and WUE, at least for some cereal crops in the Eastern Mediterranean.

J. Ben-Asher, Professor Emeritus, Ben Gurion University Agroecology Group, The Katif R&D Center, Ministry of Science and Technology, Sedot Negev Regional Council 85200, Israel; e-mail: benasher@bgu.ac.il.

T. Yano, Professor Emeritus, Tottori University, Arid Land Research Center, Tottori, Japan; e-mail: yano@ant.bbiq.jp.

M. Aydın, Retired Professor, Mustafa Kemal University, Department of Soil Science, Antakya, Turkey; e-mail: maydin08@yahoo.com.

A. Garcia y Garcia, Assistant Professor, Department of Agronomy and Plant Genetics, University of Minnesota, Southwest Research and Outreach Center, Lamberton, MN, United States; e-mail: axel@umn.edu.

Appendix

1.1 Climate Change Scenarios

Based on a range of several current climate models, the mean annual global surface temperature is projected to increase by 1.4 to 5.8 °C over the period of 1990–2100 (IPCC 2001), with changes in the spatial and temporal patterns of precipitation (Southworth et al. 2001; Raisanen 2001).

The objective of this appendix was to support the above study by generating future climate data that can be used in the models of the Çukurova plain using GCMs and a known RCM of this region.

The climate change data were obtained from the outputs of the three GCMs: the second version of the Canadian Global Coupled Model – CGCM2 – (Flato/Boer 2001), the model developed from the atmospheric model of the European Centre for Medium Range Weather Forecasting, and parameterised at HAMburg—ECHAM4 (Roeckner et al. 1996), and the general circulation model developed at the Meteorological Research Institute of Japan – MRI (Yukimoto et al. 2001; Kitoh et al. 2005). The impacts of climate change based on GCM data were estimated for the A2 scenario in the Special Report on Emission Scenarios (SRES). The A2 scenario describes a very heterogeneous world of high population growth, slow economic development and strong regional cultural identities. Scenario A2 is one of the emission scenarios with the highest projected CO₂ increase (up to 750–800 ppm) by the end of the 21st century (Nakicenovic/Swart 2000). The GCM-based climate change data are available with seven climate models on the IPCC website (http://ipcc-ddc.cru.uea.ac.uk/). Monthly temperature and precipitation values projected by CGCM2 and ECHAM4 were obtained from the IPCC database. The MRI model can be used to explore climate change associated with anthropogenic forcing (Yukimoto et al. 2001); however, its data are not available on the IPCC website. The MRI control run simulates the current climate condition, while the MRI global warming run is performed based on the A2 scenario of the SRES. As was mentioned previously in the materials and methods section, the MRI data were computed for Adana from the projected values at the four nearest grid points. It is unreasonable to expect that a large GCM grid box completely represents climate for any particular point. In order to move from the coarse grid scale of the GCM outputs to the specific location, the following procedure was used: GCM data from the four nearest grid points were used to compute climatic data for the specific site. The actual values were calculated using the inverse distance weighted method between the specific site and the GCM grid points (Schulze 2000).

Monthly mean precipitation predictions for Adana for the future period (2070–2079) based on the CGCM2, ECHAM4 and RCM models are depicted in Fig. 13.A1.

Fig. 13.A1

Comparison of monthly precipitation created from the CGCM2, ECHAM4 and RCM models for a period of ten years from 2070. Source Yano et al. (2007): p 2,305

Projected mean annual precipitation would decrease by 133, 56, and 306 mm (equivalent to a 25, 12 and 46% decrease), according to the sums of CGCM2, ECHAM4 and MRI models respectively. The discrepancies between the RCM and GCM results can be attributed to the spatial resolutions of the models (Fig. 13.A1).

Variations in mean annual precipitation for 111 years from 1990 to 2100 in Adana are shown in Fig. 13.A2.

Fig. 13.A2

Variations in annual precipitation from 1990 to 2100 downscaled for Adana with the CGCM2, ECHAM4 and MRI models. Source Yano et al. (2007): p 2,303

Precipitation is projected to decrease by about 163 mm, 163 mm and 105 mm over the period of 1990–2100 under the A2 scenario of the CGCM2, ECHAM4 and MRI models respectively (Fig. 13.A2).

Decreasing rainfall trends in Turkey have already been observed during the 20th century (Türkeş 1996).

1.2 Global Climate Change and Future Agro-productivity

The impact of climate change on two crops – summer maize, which is associated with low rainfall amounts and large irrigation water requirements (Table 13.B1), and winter wheat, with larger rainfall amounts and minimal irrigation water requirement (Table 13.B2) – were simulated using the DSSAT and SWAP models. The water balance components, such as precipitation (Rain), actual crop evapotranspiration (ETa) and irrigation amount (irrig), and crop growth components, such as dry matter production (DMP), grain yield (Grain) and growing duration, are shown in Tables 13.B1 and 13.B2.

Table 13.B2 Tabulated results of final values for major production variables of present and future wheat. Source The authors

Both tables were obtained from daily runs of the models and used for the average values that are presented in Tables 13.2 and 13.3 in the main part of this article.

Precipitation amount predicted and observed are comparable, especially in the summer, although some considerable deviations are evident.