Extreme Temeperature Datasets Utilized in the Multivariate Logarithmic Modeling of Grain Production in the Yangtze River Basin

Data: gridded to a spatial resolution of 0.5deg *0.5deg, <strong>coldwave_frequencies.nc</strong> and <strong>hwcase19802022.nc</strong> contain historical annual terrestral coldwave/heatwave respectively in the timeframe 1980~2022 derived from CPC global unified temperature dataset provided by NOAA PSL. <strong>ssp_ete.zip</strong> is a compiled set of future coldwave and heatwave frequencies derived from CMIP6 datasets MIROC6 and BCC_CSM2_MR. Said zip archive is utilized for near-future per-hectare grain yield prediction in the original paper. Methods: the 90th percentile of the daily Tmax 1981-2010 climatology distribution defines the minimum heatwave-qualifying temperature of a specific location, and a day will be labeled as 'heatwave day' when daily Tmax exceeds the threshold over at least three consecutive days. Coldwave is determined following similar principles: an event of coldwave contains more than 3 consecutive cold days with Tmin lower than the 10th percentile of the daily Tmin 1981-2010 climatology distribution. <strong>cwave.py</strong> is an implementation of this method in the determination of heatwaves and coldwaves from SSP experiment datasets.