基于蒸腾速率与光谱特征的松材线虫病害预测

Prediction of Bursaphelenchus xylophilus based on transpiration rate and spectral characteristics

【目的】明确马尾松感染松材线虫病初期蒸腾速率与光谱特征的变化规律及相互关系,以解决病害潜伏期表征不易鉴别的问题,为在遥感监测中特征参数的选取提供理论参考。【方法】以福建省泰宁县马尾松林为例,分别对感染松材线虫病后4个病害阶段(健康期及病害初期、中期、末期)马尾松的蒸腾速率进行测定,分析其在不同病害阶段的变化规律;再对健康期及病害初期马尾松的光谱数据进行采集,以这2个阶段的蒸腾速率为基准,分析其蒸腾速率与原始光谱及光谱特征参数(红边位置、红边斜率、红谷位置、红谷反射率、红边比值植被指数和红边归一化植被指数)的关系。【结果】马尾松蒸腾速率随着病害的加重而逐渐减弱;健康期及病害初期马尾松蒸腾速率与原始光谱有一定的相关性,相关关系因波长的不同而不同,最大值在753nm处,相关系数达到0.86;蒸腾速率与红边参数有较高相关性,拟合结果R2均大于0.7,其中与红边比值植被指数相关性最高,R2达到0.851 5。【结论】在外界因素一致的情况下,马尾松蒸腾速率对预测病害的发生有一定的指示作用,近红外波段及红边参数与蒸腾速率有较高相关性,可以利用特征光谱对松材线虫病进行早期预测。

【Objective】This study investigated the changes in transpiration rate and spectral characteristics of Pinus massoniana at the initial stage of Bursaphelenchus xylophilus infection and their relationship to improve the identification of Bursaphelenchus xylophilus during incubation and the selection of feature parameters in remote sensing monitoring.【Method】Transpiration rates at four different disease periods(HP,EPD,MPD,and TPD)of P.massonianainfected by B.xylophilus were measured to investigate the change laws in Taining,Fujian.Then transpiration rates at HP and EPD were analyzed with collected spectral data and spectral characteristics(REP,RES,RBP,RRB,RERVI,and RENDVI).【Result】The transpiration rate of P.massonianagradually weakened with the deterioration of disease.There was certain relationship between the transpiration rates and raw spectral data at health and early disease stages.The correlation coefficients were dependent on wave length(the maximal value was 0.86 at 753nm).Transpiration rate and red edge parameters had high relationship with R2 of larger than 0.7,especially for RERVIwith R2 of 0.851 5.【Conclusion】The transpiration rate of P.massonianacould be an indicator of disease.Transpiration rate had high correlation with infrared band and red edge parameters.Thus spectral characteristics can be used for early forecast of B.xylophilus.