基于Sentinel-2遥感影像的黑土区土壤有效磷反演

Inversion of available phosphorus in black soil based on Sentinel-2 remote sensing image

关键词:  Sentinel-2影像  土壤有效磷  偏最小二乘回归  玉米农田  黑土区

Key words:  Sentinel-2 image  soil available phosphorus  partial least-squares regression  maize farmland  Black-soil area

摘要:

【目的】 土壤养分含量的整体状况对供给作物的生长有着重要意义，对农田土壤速效养分含量的估算是作物生长发育状况评价过程中的关键环节，高光谱遥感技术的发展为区域有机质和土壤养分的动态监测提供了有效的手段。相比传统方法耗时、耗力及测量复杂等缺点，该文提出了基于遥感影像的大面积土壤速效养分含量的精准估算方法，对促进玉米作物的精准施肥和黑土地肥力保护具有重大理论意义。【方法】 文章以东北黑土区的吉林省农安县玉米农田为研究区，基于Sentinel-2遥感影像和土壤有效磷含量实测数据，利用偏最小二乘回归方法构建土壤有效磷含量的反演模型，最终获取土壤有效磷含量的空间分布图，实现了黑土区土壤有效磷含量的大面积、快速、精准估算。【结果】 Coastal aerosol波段（海岸/气溶胶波段）、可见光波段及红边波段与土壤有效磷含量的相关性较强，模型的决定系数R²为0.649、均方根误差为2.44 mg/kg。【结论】 利用偏最小二乘回归模型估算所得土壤有效磷含量空间分布情况与其空间插值结果相一致，从空间分布情况上看，农安县西北部的土壤有效磷含量最高，东北部的土壤有效磷含量高于中部与南部地区；模型估算的土壤有效磷含量均值为18.08 mg/kg，空间插值的土壤有效磷含量均值为17.63 mg/kg。由此可见，该研究所建立的土壤有效磷估算模型具有一定的准确性与可靠性。

Abstract:

[Purpose] The overall condition of soil nutrient content is of great significance to the growth of crops，and the estimation of available nutrient content in farmland soil is the key link in the evaluation process of crop growth and development. The development of hyperspectral remote sensing technology provides an effective means for dynamic monitoring of regional organic matter and soil nutrients. Compared with the shortcomings of traditional methods，such as time-consuming，labor-consuming，and complicated measurement，this paper puts forward an accurate estimation method of soil available nutrient content based on remote sensing images，which is of great theoretical significance for promoting accurate fertilization of corn crops and fertility protection of black soil.[Method] Based on Sentinel-2 remote sensing image and measured data of soil available phosphorus content，the inversion model of soil available phosphorus content was established by Partial Least-Squares Regression（PLSR）method，and finally the spatial distribution map of soil available phosphorus content was obtained，which realized the large-area，rapid and accurate estimation of soil available phosphorus content in black soil area.[Result] Coastal aerosol wave band（coast/aerosol wave band），visible light wave band and red edge wave band have strong correlation with soil available phosphorus content. The determination coefficient R² of the model is 0.649 and the Root-Mean-Square Error，RMSE）is 2.44 mg/kg.[Conclusion] The spatial distribution of soil available phosphorus estimated by partial least squares regression model is consistent with its spatial interpolation results. From the spatial distribution，the soil available phosphorus content in the northwest of Nong’an County is the highest，and that in the northeast is higher than that in the central and southern areas. The estimated average soil available phosphorus content of the model is 18.08 mg/kg，and the spatial interpolation average soil available phosphorus content is 17.63 mg/kg. Therefore，the estimation model of soil available phosphorus established in this research has certain accuracy and reliability.