用于监测柑橘叶片冻害的叶绿素含量光谱反射模型研究

果树叶片的叶绿素含量（SPAD）是植株营养状况的表现,也决定其光合作用能力的强弱,并且越冬前的树体营养状态,对果树抵御极端低温和顺利越冬是一个重要的影响因素,而果树受冻时叶绿素渗出和降解也是冻害发生程度的指标。通过对不同生长期的柑橘叶片进行光谱扫描,采用逐步回归法、红边参数法和光谱指数法分析叶片光谱反射率和叶绿素含量之间的关系,构建了柑橘叶片叶绿素光谱反射模型。结果表明,柑橘叶片叶绿素含量与反射光谱之间有较强的相关性,且两种方法所得模型预测值与实测值的相对误差都小于10%,说明模型具有良好的预测结果。两种方法中,选择波段的逐步回归法比光谱指数法的精度更高,但从建模参数的物理意义和逻辑性方面考虑,推荐光谱指数法建模,该模型可为远距离遥感监测果园营养状况和冻害情况提供参考。

Model for Citrus Leaves Chlorophyll Content Based on Spectral Signature

The chlorophyll content of leaves not only reflects the capability of photosynthesis,but also symptomizes tree＇s condition of the storage nutrition.The tree＇s storage nutrition prior winter is an important factor to resist extreme low temperature in winter.Accompanied by frozen injury in winter,the damage of protoplast resulted in exudation of cell solutes and decrease of chlorophyll content.In this research,we investigated the relationship between the chlorophyll content and spectrum reflectance of leaves,aiming to establish a mathematic model inspecting tree＇s nutrient condition by spectrum reflectance.Two methods,including stepwise regression and spectral indexes method were performed respectively.The results showed that there was a good correlation between SPAD and spectrum reflectance of citrus＇s leaves.The relative errors of predicted models and observed data were all less than 10%,indicating two models were available in remote evaluation for leaves chlorophyll content.The model established with stepwise regression method had better estimation than the model with spectral indexes method,but considering the physical meaning and logic aspect,the latter method were recommended.These models provided a possible reference for remote monitoring growth status and cold injury situation in orchards.