Impact of Climate Change on Sugarcane Yield in Maceió

Abstract In this study, a statistical downscaling model (SDSM) based on the analogous methodology was used to project future patterns of precipitation and temperature in the city of Maceió, east coast of northeastern Brazil (9.7° S; 35.7° W; 64.5m), and to evaluate its impacts on sugar cane yield. Historical observed data were used for downscaling precipitation and temperature using different global climate models (CanESM2, CNRM-CM5, IPSL-CM5A-MR, GFDL-ESM2M, MIROC-ESM, MPI-ESM-MR, and NorESM1-M) in relation to the reference period (1961-1990), and then for future scenarios from 2021 to 2080. The data were analysed in a ten days period, the same time step as the agrometeorological model used for culture. Contrary to what was expected, the combination of reduced rainfall and increased evapotranspiration in future scenarios did not decrease the estimate of yield, with an increase forecast. This effect can be attributed to the fact that the reduction in predicted rainfall, which was more pronounced in the rainy season, was not decisive for the decrease in yield, since in most MCGs there is a forecast of increased rainfall in the dry period from September to December, which is crucial for the initial growth of the crop, which has a planting schedule from the first ten-day period of September.

摘要 本研究采用基于类比方法的统计降尺度模型（statistical downscaling model, SDSM），针对巴西东北部东海岸的马塞约市（9.7°S；35.7°W；海拔64.5米），预估未来降水与气温的时空格局，并评估其对甘蔗产量的影响。研究以1961-1990年为参考时段，利用CanESM2、CNRM-CM5、IPSL-CM5A-MR、GFDL-ESM2M、MIROC-ESM、MPI-ESM-MR及NorESM1-M等多套全球气候模式（Global Climate Model, GCM）对历史观测降水与气温数据开展降尺度处理，获取对应序列，进而模拟2021至2080年的未来气候情景。数据以十日为时间步长进行分析，该步长与作物配套的农业气象模型时间步长保持一致。与预期结果相悖的是，未来情景中降水减少与蒸散量（evapotranspiration）升高的协同效应并未导致产量预估下降，反而呈现增产趋势。该现象可归因于：尽管雨季的预估降水减少更为显著，但并未对产量降低产生决定性影响——多数全球气候模式均预估9月至12月的旱季降水有所增加，而该时段对始于9月第一个十日的作物初始生长至关重要。