基于植物净初级生产力模型的区域冬小麦估产研究

Regional yield estimation for winter wheat based on net primary production model

该文以中国冬小麦主要种植区黄淮海平原典型县市的冬小麦为研究对象,以植物净初级生产力模型对冬小麦估产进行研究。其中光合有效辐射数据(PAR)主要通过TOM S传感器紫外反射率月数据来计算获得。并且通过投影转换和内插方法,将分辨率由经度1.25度、纬度1度转为250 m。光合有效辐射分量(f PAR)主要通过250 m分辨率M OD IS的最大值合成法生成的N DV I月数据和f PAR之间的统计直线关系(f PAR=a N DV I+b)来反演。在研究中光能转化有机质效率(ε)被视为常数,其值通过前人研究结果确定。然后计算冬小麦净初级生产力(N P P=ε×f PAR×PAR)。文中主要考虑冬小麦产量形成关键期内N P P的形成,然后将累积的N P P转化为作物干物质的量,最后通过冬小麦收获指数修正,得到估计的冬小麦产量。而且利用地面实际调查产量数据对所预测的植物净初级生产力N P P和所预测的产量进行了验证,通过N P P计算的冬小麦生物量与实际生物量间相对误差为-4.30%;预测冬小麦产量与实际小麦产量间相对误差平均为-4.41%,结果令人满意。

The timely and accurate crop yield data have always been useful to government. Winter wheat is one of the most important crops in China. So, the authors select a widely-used net primary production （NPP） model to estimate the yield of winter wheat in the Huanghuaihai Plain in China. The authors selected the data which come from the critical growth stage which is significant to the final yield of winter wheat. The parameter of photosynthetically active radiation （PAR, 400～700 nm） was calculated using the ultraviolet （UV） reflectance data at 370 nm which come from the Total Ozone Mapping Spectrometer（TOMS）. And the fraction of the photosynthetically active radiation （fPAR） can be retrieved from MODIS Normalized Difference Vegetation Index （NDVI） image using the linear regression between the fPAR and NDVI （fPAR=aNDVI＋b）. And biological efficiency of PAR conversion to dry matter （ε） is supposed to be a constant in this paper through referring to local experiments in field and published literature. Then the authors used the equation NPP=ε×fPAR×PAR to calculate the winter wheat net primary production. Finally, estimation yield data through converting the calculated NPP to dry matter and multiplying the dry matter by the Harvest Indices （HI） for specific winter wheat varieties were obtained. The authors also analyzed the accuracy of the prediction of NPP. And the accuracy of final yield estimation versus the field yield data was also analyzed. The relative error between the winter wheat biomass calculated from NPP and the actual biomass was -4.30% and relative error between the predicted yield of winter wheat and the actual yield was -4.41%. So the results are satisfactory and using the NPP parameter to estimate winter wheat yield is practical and reliable.