The spatial distribution patterns of grassland NPP in south China based on climate productivity models

孙成明,陈瑛瑛,武威,陈雯,韩正笑,李建龙
DOI: https://doi.org/10.16872/j.cnki.1671-4652.2013.04.012
2013-01-01
Abstract:According to the climate data of nearly 31years(1980-2010),the grassland net primary productivity(NPP)in south China was estimated based on 5kinds of climate productivity models.The results showed that the spatial distribution patterns of southern grassland NPP simulated by 5kinds models were consistent.The regions of minimal NPP were distributed in the northwest of Sichuan,then the value of NPP gradually increased from the northwest to the southeast in south China,and the regions of maximal NPP were distributed in the eastern and southern coastal areas.But there were certain differences between the model estimation results.The sequences of maximal NPP estimated by the models from maximum to minimum were successively Miami model,Thornthwaite Memorial model,Zhou Guangsheng model,Chikugo model and Zhu Zhihui model.The sequences of minimal NPP estimated by the models from maximum to minimum were successively Chikugo model,Zhu Zhihui model,Thornthwaite Memorial model,Miami model and Zhou Guangsheng model.The sequences of average NPP estimated by the models from maximum to minimum were successively Miami model,Thornthwaite Memorial model,Chikugo model,Zhu Zhihui model and Zhou Guangsheng model.The maximum,minimum and average value of the estimation results of 5kinds of models was 1 676.26,664.71and 1 201.64g·m-2, respectively,and they were all in the middle level,and close to the Chikugo model estimation results.The model validation showed that the values of relative error,root mean square error(RMSE)and relative root mean square error(RRMSE)of Zhou Guangsheng model were the smallest,and R2 was bigger than other 3 models.The above results showed that it was appropriate to estimate the grassland NPP in south China based on Zhou Guangsheng model.
What problem does this paper attempt to address?