Regional crop yield, water consumption and water use efficiency and their responses to climate change in the North China Plain

Abstract

The North China Plain (NCP) is one of the most important regions for food production in China, with its agricultural system being significantly affected by the undergoing climate change and vulnerable with water stress. In this study, the Vegetation Interface Processes (VIP) model is used to evaluate crop yield, water consumption (ET), and water use efficiency (WUE) of a winter wheat (Triticum aestivum L.)–summer maize (Zea mays L.) double cropping system in the NCP from 1951 to 2006. Their responses to future climate scenarios of 21st century projected by the GCM (HadCM3) with Intergovernmental Panel on Climate Change Special Report on Emission Scenario (IPCC SRES) A2 and B1 emissions are investigated. The results show a rapid enhancement of crop yield in the past 56 years, accompanying with slight increment of ET and noticeable improvement of WUE. There exist spatial patterns of crop yield stemmed mainly from soil quality and irrigation facilities. For climate change impacts, it is found that winter wheat yield will significantly increase with the maximum increment in A2 occurring in 2070s with a value of 19%, whereas the maximum in B1 being 13% in 2060s. Its ET is slightly intensified, which is less than 6%, under both A2 and B1 scenarios, giving rise to the improvement of WUE by 10% and 7% under A2 and B1 scenarios, respectively. Comparatively, summer maize yield will gently decline by 15% for A2 and 12% for B1 scenario, respectively. Its ET is obviously increasing since 2050s with over 10% relative change, leading to a lower WUE with more than 25% relative change under both scenarios in 2090s. Therefore, possible adaptation countermeasures should be developed to mitigate the negative effects of climate change for the sustainable development of agro-ecosystems in the NCP.

Introduction

Greenhouse gases emission from fossil combustion, cement production and land use/cover changes have propelled the global climate changing, which appears as a widespread rising of surface air temperatures, alteration of precipitation patterns and global hydrologic cycle, and increased frequency of severe weather events, such as drought spells and flooding. In many regions, agricultural crops are sensitive to climate change (Izaurralde et al., 2003, Lobell and Field, 2008). Usually, air warming will accelerate the crop development, alter the phenological period, and enhance the maintenance respiration; on the other hand, atmospheric CO₂ enrichment will increase the leaf photosynthetic rate and reduce transpiration simultaneously, adding additional carbon to the ecosystems, and hence leading to changes in the cycling of water, nutrients and energy balance (Polley, 2002, Fuhrer, 2003). Due to complex interactions between the climatic and other environmental factors, the impacts of climate change on agricultural ecosystems may be interacted under diversified agronomical practices. For example, the responses of wheat yield to global warming are different between rain-fed and irrigated conditions, and between well and less fertilized conditions (Tubiello et al., 2000).

The physically process-based models, designed for crop ecosystem simulation, in which the environmental and management factors and their interactions are integrated, are broadly applied to project the responses of crops to future climate change scenarios (Brown and Rosenberg, 1997, Mearns et al., 2001, van Ittersum et al., 2003, Trnka et al., 2004, Zhang and Liu, 2005, Thomson et al., 2006, Walker and Schulze, 2006). With the crop models, a lot of researches on the responses of wheat and maize to global change were conducted. For example, Southworth et al. (2002) predicted the wheat responses in the Midwestern United States for 2050–2059 with atmospheric CO₂ concentration of 555 ppm, elucidating that wheat yields would increase 60–100% above current yields across the central and northern areas, but both small increases and decreases were found in the southern areas. Trnka et al. (2004) reported that winter wheat would enhance its productivity under both Intergovernmental Panel on Climate Change (IPCC) Special Report on Emission Scenarios (SRES) A2 and B1 scenarios. Under the three emission scenarios (A2a, B2a, and GGa1), Zhang and Liu (2005) predicted with EPIC model that the productivities and evapotranspiration of wheat and maize would noticeably increase over the Loess Plateau. Generally, it is projected that winter wheat as a C3 crop will increase its productivity and water use efficiency in most cases due to the atmospheric CO₂ fertilization, but the annual variability and vulnerability of crop yield are also exaggerated. Maize as a C4 crop has low growth response to elevated CO₂ concentration and then benefits less from its enrichment, due to their CO₂ concentrating mechanism in the photosynthetic path (Kim et al., 2007). The increased air temperature and changed precipitation pattern will significantly affect crop phenological process and stomatal conductance, which lead to alteration of yield and water use efficiency (Kattge and Knorr, 2007). However, the results from Free Air Carbon Enrichment (FACE) experiments show that the stimulation of grain yield by CO₂ enrichment is lower than expected (Kimball et al., 2002, Long et al., 2006). This discrepancy is possibly related to a fact that the crop models usually predict with non-limited supply of water and nutrition and near optimum temperature for crop growth. Usually, assessment of climate change impact is intended to seek the adaptation measures that may be the choices to mitigate the negative feedbacks to agro-ecosystems, to maintain and even increase the crop yields under future scenarios. These measures include selection of the most favorable crops, guidance for new cultivars breeding, and application of dynamic cropping (Hanson et al., 2007).

The North China Plain (NCP) is the country's most productive region of agriculture, accounting for about 69% of wheat (Triticum aestivum L.) and 35% of maize (Zea mays L.) grain yields of the whole country (Liu et al., in press). In the recent two decades, the intensive agricultural systems are mainly composed of wheat–maize double cropping, which is strongly dependent on the available irrigation from aquifer pumping, reservoir and river withdrawing. However, as stressed by the climatic variability and economic development, the water resources in this region are vulnerable. In order to meet the irrigation requirement of the intensified agriculture, groundwater has been pumped in excess of recharge, giving rise to continuously dropping of groundwater table during the last several decades and forming the so-called “groundwater funnel” in some areas. Consequently, the overuse of water resources has deteriorated the agricultural sustainability and caused serious environmental hazards. As an aspect of water deficit mitigation, the responses of agricultural system in the NCP to climate change have been highly concerned.

So far, there are several reports on the response of crop yield of the NCP to climate change. For example, Thomson et al. (2006) reported that under A2 and B2 scenarios wheat yield and soil carbon sequestration would significantly increase in the NCP. Liu et al. (in press) chose two typical counties each in the south and north of the plain respectively to explore the response of crop yield to climate change with the Vegetation Interface Processes (VIP) model under different scenarios.

However, there have not been many researches on the regional responses of crop yield, water consumption (ET) and water use efficiency (WUE) to climate change in wheat–maize double cropping system over the plain. Because water is the critical limited factor in the NCP, it is better to study the three variables of crop yield, ET and WUE together than just to study each of them alone. Except some statistic yield data at county and provincial scales, the observed data of ET and WUE over a region are usually not available. The yield data sometimes contain uncertainty because of some unavoidable factors (Mo et al., 2005). Using a physically process-based crop model to simulate crop yield, ET and WUE is an effective way for the analysis. As in situ measurements are always at point scale, models can be used to upscale information from point to large area. The successful simulation of crop productions in the past several decades will improve the reliability of projection on the future climate change responses.

The purpose of this study is to explore the regional crop response to climate change. For this, firstly the spatial variability and evolution of crop yield, ET and WUE with a process-based crop model in the NCP is explored. The contribution of climate change to their enhancement in the past 56 years is then identified. Further, the impacts of future climate changes under A2 and B1 scenarios on the wheat–maize double cropping system are assessed. Finally discussions and conclusions are given.