气候变化背景下“气候产量”计算方法的探讨

气候变化对于粮食安全的影响已成为全球变化研究的热点问题之一。在研究过去几十年尺度上的气候变化对粮食产量影响的程度时,常规的研究方法是利用实际产量的数据拟合出以时间为变量的趋势产量,实际产量与趋势产量的差值即为气候产量。但这种方法在评估气候变暖对粮食产量的影响时存在一定缺陷。在常规算法的基础上,论文提出一种改进算法,即通过选取基准时段,建立反映气候要素与作物单产关系的函数表达式,来估计非基准时段内的气候产量,这种方法主要适用于气候要素存在趋势性变化的时段。以黑龙江省为例,将改进算法和常规算法进行比较,结果显示应用改进算法得到的气候产量更能真实地反映气候变暖对水稻产量的影响。

A method for calculating the climatic yield of grain under climate change

The impact of climate change on grain production has become one of the hotspots in the research on global change.The impact of past climate change on crop yield is an effective approach for this research,however,it is difficult to separate the contribution of climate changes and human activities in such studies.Traditionally,the real yield is separated into trend yield that fits the real crop yield by a trend line depended on time only,and the fluctuant yield which is the residual yield between the real yield and the trend.The fluctuant yield is regarded as the climatic yield.But the main disadvantage of the traditional method is that it could not show the actual contribution of trend of climate change to yield of grain.In this paper a new method is put forward to calculate climate yield,which may overcome the disadvantage of traditional method. Climate influence coefficient that is the ratio of climate yield to trend yield is given to express the influence on the yield by meteorological factors.In the new method,a referenced period that satisfies the hypothesis of traditional method is selected to construct a function on the main meteorological factor and climate influence coefficient by regression.This function can be used to calculate climate yields beyond the referenced period.As a case,this method is used to calculate the climate yield of rice in Heilongjiang Province.Traditional methods,such as exponential function,polynomial function and moving average,are used to fit the per-unit-area rice yield for comparison.The comparative result shows that the contribution by climate warming can be revealed more accurately by using the new method.