Modelling The Effects Of Low Temperature Stress At Elongation And Booting Stage On Growth And Yield Formation In Wheat

Extreme temperature events are become more frequent due to temperature fluctuations in the context of climate change,which brings extremely adverse effects on crop production.Wheat becomes more sensitive to low temperature with the the decrease of cold resistance after elongation stage.Extreme low temperature events in spring,such as spring cold and late spring frost,have become a major threat to the safety of wheat production after elongation stage.Process-based crop models have proven to be key tools for understanding the processes of the complex interconnections in cropping systems,and for assessing climate impacts and adaptation options for crop systems.However,the existing crop growth models lack a relatively rational low-temperature stress effect algorithm,and there is great uncertainty in the evaluation of extreme low-temperature effect under climate change.Therefore,it is urgent to improve the ability of crop growth models to predict low-temperature stress response in elongation-booting stage.In this study,we first analyzed the spatio-temporal characteristics of low temperature stress after elongation stage and its impact on grain yield based on the historical climate data at different meteorological stations and phenological development and yield data at different agro-meteorological experimental stations during last 29 years across the main winter wheat growing area in China.After systematic analysis of low temperature effects on wheat growth and yield,algorithms and routines for phenology,leaf photosynthesis,leaf area index,biomass partitioning,and grain yield under low temperature stress were proposed and integrated into WheatGrow model to improve the response accuracy of low temperature effects.The results will provide a quantitative tool for accurately assessing the impact of extreme low temperature on wheat productivity in the future.Low temperature stress indices(including accumulated frost days,accumulated frost degree days and temperature-drop-rate)were calculated with historical climate and phenology data from 161 meteorological stations and 141 agro-meteorological experimental stations to characterize the spatial and temporal variability of low temperature stress for the growth period from stem elongation to flowering across the main winter wheat growing area in China from 1981-2009.To understand low temperature impact on winter wheat yield,we set 2℃ as low temperature stress threshold and calculate the accumulated frost degree-days(AFDD)as an index for frost risk affecting yield.Surprisingly,low temperature stress risk is smaller in the cooler northern and warmer southern regions than in the central winter wheat growing region of China.Huang-Huai Subregion(HHS)has the greatest low temperature duration,intensity and suffers the largest yield losses due to spring frost.Low temperature stress risk during stem elongation to flowering has not been significantly decreased in the winter wheat-growing regions of China under climate warming.While rising temperature reduces frost events in general,it also accelerates wheat phenology which increases the risk of low temperature stress.The yield of HHS was the most sensitive to low temperature stress among the four sub-regions,and the yield decreased by 1.5-2.1%for every 1℃·d increase of AFDD.Quantifying future trends and impacts of low temperature damage on wheat production will be critical for developing appropriate adaptation and mitigation strategies for food availability in China and other regions of the world affected by low temperature stress.Four low temperature stress response routine(CERES-Wheat routine,CropSyst routine,WOFOST routine and STICS routine)imbedded in the WheatGrow model were evaluated with phytotron experimental datasets under low temperature stress at elongation and booting stage.The result indicates that the leaf area index,biomass of aboveground parts,and grain yield decreased significantly under low temperature stress during elongation and booting stages.The four integrated low-temperature stress algorithms improved the original model in simulating the dynamic changing of leaf area index.However,most algorithms overestimate the damage of low temperature stress to leaves in the following short time after treatment.CropSyst and STICS routine are better than the other two algorithms in predicting the dynamic change of yield and aboveground biomass accumulation.None of the four low-temperature stress algorithms considered the negative effects caused by low temperature stress on stem growth and dry matter partitioning.Therefore,the deficiencies are existing in simulating the changing dynamic of stem accumulation and aboveground biomass.The results of this study clearly demonstrate the shortcomings of the existing models under low temperature stress,and provide instructions to improve the algorithms for low temperature stress.Low temperature stress significantly declines the growth rate of wheat and slows down their growth process.Five widely used temperature response routines(HERMES、DSSAT-CERES、Expert-N-CERES、DSSAT-Nwheat and Apsim-E)of wheat phenology imbedded in the WheatGrow model were tested to simulate low temperature stress effects on elongation-flowering durations with datasets from four-years environment-controlled phytotron experiments.In order to improve the prediction ability of growth model under low temperature stress,a new algorithm for limiting daily thermal sensitivity(DTS)under low temperature stress was proposed and applied in six temperature response routines.The result indicates that the flowering period of wheat was significantly delayed under a low temperature stress at elongation and booting stage.All the temperature response routines underestimated the delaying effects on wheat growth process caused by low temperature stress.The DTS algorithm can well simulate the process of wheat growth,including the growth rate declines under low temperature stress and growth rate recovery after treatment.The model test results show that the six models with the improved DTS algorithm all significantly improved the simulation accuracy of elongation-flowering duration under low temperature stress.The normalized root mean squared error(NRMSE)and the root-mean-square error(RMSE)of elongation-flowering duration reduces 3%and 1.37 days respectively,and the consistency coefficient,as D-index shows,has been improved to more than 0.9.The results provide theoretical support for the simulation of wheat growth period under extreme climate conditions,and an effective tool for quantitative evaluation of the impact of low temperature stress on wheat growth period under climate change.Dynamic of leaf photosynthesis,biomass accumulation,biomass partitioning and grain yield were analyzed for different low temperature stress treatment.Based on environment-controlled phytotron data,LT50(lethal temperature at which 50%are killed)and LDI(low tempearture damage index)were proposed to build the low temperature stress algorithm on photosynthesis,dry matter production and partitioning of wheat.Compared to the original model,the improved WheatGrow model with a new algorithm decreased simulation errors of photosynthetic rate of single leaf,leaf area index,total aboveground biomass,biomass of each organ and yield under low temperature stress.Among them,RMSE of photosynthetic rate of single leaf,total aboveground biomass and yield are decreased by 64.8%,77.4%and 74.7%respectively on average.The consistency coefficient(D-value)and R2 are both above 0.9.The results provide an important digital tool for quantitative assessment of the impact of low temperature stress on wheat productivity in the context of climate change.