考虑含水量变化信息的土壤有机质光谱预测模型

发布时间：2018-08-18 18:27

【摘要】：地物高光谱技术已被用于土壤有机质(SOM)等理化参数速测,但由于含水量、粗糙度等因素的影响,基于遥感影像的SOM空间反演精度较低。为此引入时相信息,将时像信息与光谱信息结合对研究区SOM进行预测,使预测模型精度显著提高。以黑龙江典型黑土区(北安市南部、海伦市中部、绥化市东部、绥棱县西南部、望奎县中部)为例,获取多期MODIS影像,利用MODIS数据高时间分辨率的优势,研究含水量对土壤反射光谱曲线的影响;基于SOM与含水量对反射率的综合作用分析,建立SOM遥感预测模型。结果表明:(1)利用单期影像建立的SOM光谱预测模型,未加入含水量变化对土壤反射光谱曲线的影响信息,基于年积日(DOY)117,119,130,140,143单期影像建立的SOM预测模型,RMSE分别为0.591,0.522,0.545和0.553,R~2分别为0.505,0.614,0.562,0.568和0.645,模型精度及稳定性较低;(2)利用年积日119和143多时相影像建立的SOM预测模型,考虑了含水量与SOM的综合作用,RMSE为0.442,R~2为0.723,模型精度、稳定性得到显著提高。研究成果对于区域土壤肥力评价、土壤碳库储量估测、精准农业发展有重要意义。
[Abstract]:Ground object hyperspectral technique has been used for rapid measurement of physical and chemical parameters such as soil organic matter (SOM), but the precision of SOM spatial inversion based on remote sensing image is low due to the influence of water content and roughness. For this reason, the time-phase information is introduced and the time-image information is combined with the spectral information to predict the SOM in the study area. The precision of the prediction model is improved significantly. Taking the typical black soil area of Heilongjiang Province (south of Bei'an, central part of Hailun, east of Suihua, southwest of Suilen and central Wangkui) as an example, multi-phase MODIS images are obtained and the advantage of high temporal resolution of MODIS data is utilized. The effect of water content on soil reflectance spectral curve was studied, and the SOM remote sensing prediction model was established based on the comprehensive effect of SOM and water content on reflectivity. The results showed that: (1) the SOM spectral prediction model based on single phase image did not include the information of the influence of water content change on the soil reflectance spectral curve. The SOM prediction model based on annual (DOY) 117119130140143 is 0.591U 0.52220.545 and 0.553U / r ~ (2) is 0.5050.140.562n0.568 and 0.645, respectively, so the accuracy and stability of the model are low. (2) the SOM prediction model is established by using 119d and 143multiphase images. Considering the comprehensive effect of water content and SOM, the model accuracy and stability are improved remarkably. The research results have important significance for regional soil fertility evaluation, soil carbon pool storage estimation and precision agriculture development.
【作者单位】：东北农业大学资源与环境学院;
【基金】：国家自然科学基金项目(41501357) 黑龙江省普通高等学校新世纪优秀人才培养计划项目黑龙江省博士后启动基金项目(LBHQ13026)资助
【分类号】：S153

【相似文献】