首页|基于遥感影像的阿拉尔垦区不同土地利用类型土壤有机质含量研究

基于遥感影像的阿拉尔垦区不同土地利用类型土壤有机质含量研究

扫码查看
土壤有机质是土壤肥力的重要指标,对阿拉尔垦区土壤有机质空间分布制图可以更好了解其分布和变化规律.本研究以阿拉尔垦区为研究区,基于257个表层土壤有机质含量数据,采用监督分类法将研究区划分为5种典型土地利用类型,提取9种不同的植被指数作为建模因子,采用随机森林算法进行建模,对研究区内不同土地利用类型土壤有机质含量和空间分布情况进行对比分析.研究结果表明,随机森林(RF)模型能够准确评估土壤有机质含量,其建模集R2值为0.73,RPD值为1.87,RMSE值为1.59 g·kg-1;验证集R2值为0.71,RPD值为1.91,RMSE值为1.59 g·kg-1.基于RF模型对阿拉尔垦区不同土地利用类型土壤有机质含量进行制图,结果表明,土壤有机质由耕地土壤向外围区域递减,耕地土壤有机质含量最多,荒漠的有机质含量最少,林草地和未利用地有机质含量适中呈现西高东低的趋势.研究结果可为阿拉尔垦区土壤肥力提升以及土壤有机质监测提供科学依据.
Research on Soil Organic Matter Content of Different Land use Types in Alar Reclamation Area Based on Remote Sensing Images
Soil organic matter is an important indicator of soil fertility,and mapping the spatial distribution of soil organic matter in the Alar reclamation area can better understand its distribution and variation patterns.This research focuses on the Alar Reclama-tion Area.Based on 257 surface soil organic matter contents,the study area is divided into 5 typical land use types using supervised classification method,and 9 different vegetation indices are extracted as modeling factors.The random forest is applied for modeling to compare and analyze soil organic matter content and spatial distribution of different land use types in the study area.The research results indicate that the Random Forest(RF)model has strong predictive ability,with an R2 value of 0.73,an RPD value of 1.87,and an RMSE value of 1.59 g·kg-1 in the modeling set;The R2 value of the validation set is 0.71,the RPD value is 1.91,and the RMSE value is 1.59 g·kg-1 Based on the RF model,the mapping of soil organic matter in the Alar reclamation area is carried out.The spa-tial distribution map of soil organic matter content shows that soil organic matter decreases from farmland soil to peripheral areas,with farmland soil having the highest organic matter content and desert soil having the lowest organic matter content.The organic matter content in forests,grasslands,and construction land shows a trend of high in the west and low in the east.The research results provide scientific basis for improving soil fertility and monitoring soil organic matter in the Alar Reclamation Area.

remote sensingsupervised classificationsoil organic matterandom forest

王江竹、王玮瑱、李渊博、罗德芳

展开 >

塔里木大学农学院 新疆,阿拉尔 843300

遥感 监督分类 土壤有机质 随机森林

2025

现代农业研究
黑龙江省科学技术情报研究所

现代农业研究

影响因子:0.166
ISSN:2096-1073
年,卷(期):2025.31(1)