Abstract: By comprehensively utilizing CASA and the VPM models,the estimation method of ε_max value was studied on the pixel scale,to improve the spatial precision of ε_max value,and provide technical support for guiding agricultural production and developing precision agriculture.Taking Henan Province as research area,NPP value was calculated by combining with CASA model and relational model using MODIS data.Further combined VPM model,ε_max value was obtained.Finally,the land use data was used to carry out the temporal and spatial analysis of the ε_max values for the four vegetation cover types,i.e.forest,grass,paddy field and dry land. And the influence of related environmental factors on ε_max value was also explored.The results showed that,the value of εmax was between 0.000 and 4.796 g/MJ during 2001 and 2015.And it was spatially higher in the northwest and southwest,and in other regions was low.On the monthly scale change,all types of vegetation cover reached the peak value between June and August.The ε_max value of dry land showed obvious bimodal distribution,which was consistent with the cropping system of yield two crops a year in Henan Province.Among the four vegetation cover types,the leaves of forest for photosynthesis were denser and larger,so ε_max value was significantly higher.Among environmental impact factors,fertilizer application amount(convert into purification),carbon dioxide content,vegetation water content index and effective irrigation area showed a significant positive correlation with εmax value,and the correlation coefficients were 0.66,0.61,0.56 and 0.53,respectively.It can be seen that the change of ε_max value was the result of the combined effects of natural and human factors.The results indicate that an appropriate increase in human influence can increase the utilization of light energy at the critical stage of crop growth,thereby increasing crop production potential and yield.

Key words: Carnegie-ames-stanford approach model, Vegetation photosynthesis model, Maximum light use efficiency, Net primary productivity, Gross primary productivity, Photosynthesis, Temporal and spatial distribution, Vegetation cover type

摘要： 通过综合利用CASA模型和VPM模型，在像元尺度上开展ε_max值估算研究，从而提高ε_max值的空间精细度，为指导农业生产，开展精准农业提供技术支持。以河南省为研究区，首先基于MODIS数据，结合CASA模型和关系模型计算NPP值，进一步结合VPM模型得到像元上的ε_max值，最后根据土地利用数据对林地、草地、水田和旱田4种植被覆盖类型的ε_max值进行时空分析，并探索相关环境因素对ε_max值的影响。结果表明，2001—2015年，ε_max值介于0.000~4.796 g/MJ，空间上呈西北、西南高，其他区域低的格局；月尺度变化上，不同植被覆盖类型ε_max值均在6—8月份达到峰值，其中，旱田ε_max值呈现明显的双峰分布，和河南省大部分地区一年两熟的作物种植制度相符。4种植被覆盖类型中，因为用于光合作用的叶面积较大，林地ε_max值较高；环境影响因素中，化肥使用折纯量、二氧化碳含量、植被含水指数和有效灌溉面积与ε_max值呈现明显的正相关，相关系数分别为0.66、0.61、0.56、0.53，说明ε_max值的变化是自然、人为因素共同影响的结果。可见，在农作物生长的关键阶段，适当增加人为的影响可以提高光能利用率，从而增加作物生产潜力和产量。

关键词: CASA模型, VPM模型, 最大光能利用率, 净初级生产力, 总初级生产力, 光合作用, 时空分布, 植被覆盖类型