Crop models are essential tools for assessing the threat of climate change to local and global food production 1 . Present models used to predict wheat grain yield are highly uncertain when simulating how crops respond to temperature 2 . Here we systematically tested 30 di erent wheat crop models of the Agricultural Model Intercomparison and Improvement Project against field experiments in which growing season mean temperatures ranged from 15 • C to 32 • C, including experiments with artificial heating. Many models simulated yields well, but were less accurate at higher temperatures. The model ensemble median was consistently more accurate in simulating the crop temperature response than any single model, regardless of the input information used. Extrapolating the model ensemble temperature response indicates that warming is already slowing yield gains at a majority of wheat-growing locations. Global wheat production is estimated to fall by 6% for each • C of further temperature increase and become more variable over space and time.Understanding how different climate factors interact and impact food production 3 is essential when reaching decisions on how to adapt to the effects of climate change. To implement such strategies the contribution of various climate variables on crop yields need to be separated and quantified. For instance, a change in temperature will require a different adaptation strategy than a change in rainfall 4 . Temperature changes alone are reported to have potentially large negative impacts on crop production 5 , and hotspots-locations where plants suffer from high temperature stress-have been identified across the globe 6,7 . Crop simulation models are useful tools in climate impact studies as they deal with multiple climate factors and how they interact with various crop growth and yield formation processes that are sensitive to climate. These models have been applied in many studies, including the assessment of temperature impacts on crop production 1,8 . However, none of the crop models have been tested systematically against experiments at different temperatures in field conditions. Although many glasshouse and controlled-environment temperature experiments have been described, they are often not suitable for model testing as the heating of root systems in pots 9 and effects on micro-climate differ greatly from field conditions 10 . Detailed information on field experiments with a wide range of sowing dates and infrared heating recently became available for wheat 11,12 . Such experiments are well suited for testing the ability of crop models to quantify temperature responses under field conditions. Testing the temperature responses of crop models is particularly important for assessing the impact of climate change on wheat production, because the largest uncertainty in simulated impacts on yield arises from increasing temperatures 2 .In a 'Hot Serial Cereal' (HSC) well-irrigated and fertilized experiment with a single cultivar, the observed days after sowing (DAS) to maturity declined...

Projections of climate change impacts on crop yields are inherently uncertain(1). Uncertainty is often quantified when projecting future greenhouse gas emissions and their influence on climate(2). However, multi-model uncertainty analysis of crop responses to climate change is rare because systematic and objective comparisons among process-based crop simulation models(1,3) are difficult(4). Here we present the largest standardized model intercomparison for climate change impacts so far. We found that individual crop models are able to simulate measured wheat grain yields accurately under a range of environments, particularly if the input information is sufficient. However, simulated climate change impacts vary across models owing to differences in model structures and parameter values. A greater proportion of the uncertainty in climate change impact projections was due to variations among crop models than to variations among downscaled general circulation models. Uncertainties in simulated impacts increased with CO2 concentrations and associated warming. These impact uncertainties can be reduced by improving temperature and CO2 relationships in models and better quantified through use of multi-model ensembles. Less uncertainty in describing how climate change may affect agricultural productivity will aid adaptation strategy development and policymaking

The yield and quality of food crops is central to the well being of humans and is directly affected by climate and weather. Initial studies of climate change on crops focussed on effects of increased carbon dioxide (CO 2 ) level and/or global mean temperature and/or rainfall and nutrition on crop production. However, crops can respond nonlinearly to changes in their growing conditions, exhibit threshold responses and are subject to combinations of stress factors that affect their growth, development and yield. Thus, climate variability and changes in the frequency of extreme events are important for yield, its stability and quality. In this context, threshold temperatures for crop processes are found not to differ greatly for different crops and are important to define for the major food crops, to assist climate modellers predict the occurrence of crop critical temperatures and their temporal resolution. This paper demonstrates the impacts of climate variability for crop production in a number of crops. Increasing temperature and precipitation variability increases the risks to yield, as shown via computer simulation and experimental studies. The issue of food quality has not been given sufficient importance when assessing the impact of climate change for food and this is addressed. Using simulation models of wheat, the concentration of grain protein is shown to respond to changes in the mean and variability of temperature and precipitation events. The paper concludes with discussion of adaptation possibilities for crops in response to drought and argues that characters that enable better exploration of the soil and slower leaf canopy expansion could lead to crop higher transpiration efficiency.

Stochastic weather generators are used in a wide range of studies, such as hydrological applications, environmental management and agricultural risk assessments. Such studies often require long series of daily weather data for risk assessment and weather generators can produce time series of synthetic daily weather data of any length. Weather generators are also used to interpolate observed data to produce synthetic weather data at new sites, and they have recently been employed in the construction of climate change scenarios. Any generator should be tested to ensure that the data that it produces is satisfactory for the purposes for which it is to be used. The accuracy required will depend on the application of the data, and the performance of the generator may vary considerably for different climates. The aim of this paper is to test and compare 2 commonly-used weather generators, namely WGEN and LARS-WG, at 18 sites in the USA, Europe and Asia, chosen to represent a range of climates. Statistical tests were selected to compare a variety of different weather characteristics of the observed and synthetic weather data such as, for example, the lengths of wet and dry series, the distribution of precipitation and the lengths of frost spells. The LARS-WG generator used more complex distributions for weather variables and tended to match the observed data more closely than WGEN, although there are certain characteristics of the data that neither generator reproduced accurately. The implications for the development and use of stochastic weather generators are discussed.

Multi-model ensembles of climate predictions constructed by running several global climate models for a common set of experiments are available for impact assessment of climate change. Multi-model ensembles emphasize the uncertainty in climate predictions resulting from structural differences in the global climate models as well as uncertainty due to variations in initial conditions or model parameterisations. This paper describes a methodology of using multi-model ensembles from global climate models for impact assessments which require local-scale climate scenarios. The approach is based on the use of a weather generator capable of generating the localscale daily climate scenarios used as an input by many process-based impact models. A new version of the LARS-WG weather generator, described in the paper, incorporates climate predictions from 15 climate models from the multi-model ensemble used in the IPCC Fourth Assessment Report (AR4). The use of the AR4 multi-model ensemble allows assessment of the range of uncertainty in the impacts of climate change resulting from the uncertainty in predications of climate. As an example, the impact of climate change on the probability of heat stress during flowering of wheat, which can result in significant yield losses, was assessed using local-scale climate scenarios in conjunction with a wheat simulation model at 4 European locations. The exploitation of much larger perturbed physics ensembles is also discussed.

No abstract

Crop models of crop growth are increasingly used to quantify the impact of global changes due to climate or crop management. Therefore, accuracy of simulation results is a major concern. Studies with ensembles of crop models can give valuable information about model accuracy and uncertainty, but such studies are difficult to organize and have only recently begun. We report on the largest ensemble study to date, of 27 wheat models tested in four contrasting locations for their accuracy in simulating multiple crop growth and yield variables. The relative error averaged over models was 24-38% for the different end-ofseason variables including grain yield (GY) and grain protein concentration (GPC). There was little relation between error of a model for GY or GPC and error for in-season variables.Thus, most models did not arrive at accurate simulations of GY and GPC by accurately simulating preceding growth dynamics. Ensemble simulations, taking either the mean (emean) or median (e-median) of simulated values, gave better estimates than any individual model when all variables were considered. Compared to individual models, e-median ranked first in simulating measured GY and third in GPC. The error of e-mean and e-median declined with an increasing number of ensemble members, with little decrease beyond 10 Accepted ArticleThis article is protected by copyright. All rights reserved. models. We conclude that multimodel ensembles can be used to create new estimators with improved accuracy and consistency in simulating growth dynamics. We argue that these results are applicable to other crop species, and hypothesize that they apply more generally to ecological system models.

This is a repository copy of Similar estimates of temperature impacts on global wheat yield by three independent methods.