| 引用本文: | 赵泽艺,高阳,李朝阳,等.南疆盐碱土水氮盐光谱特征及其反演模型[J].灌溉排水学报,2023,():-. |
| Zhao Zeyi,Gao Yang,Li Zhaoyang,et al.南疆盐碱土水氮盐光谱特征及其反演模型[J].灌溉排水学报,2023,():-. |
|
| 摘要: |
| 【研究目的】探讨南疆盐碱土在不同水、氮、盐含量下的光谱特征，构建适合南疆沙土的水氮盐含量反演模型，为快速检测南疆沙土中水氮盐含量提供科学依据。【研究方法】选取南疆代表性盐碱沙土为研究对象，设置不同的土壤水分、盐分和氮含量，获取并分析不同处理的土壤光谱特征，采用偏最小二乘回归（PLSR）、支持向量回归（SVR）和BP神经网络（BPNN）建立土壤水、氮、盐反演模型。【研究结果】土壤水的特征波段在1400、1900nm附近，土壤氮的特征波段在1448~1515，1841~2500nm波段，土壤盐分的特征波段在1878~1881，1908~1940nm波段；PLSR模型对水、氮、盐的反演效果最好，BPNN次之，SVR最差。【研究结论】1900nm波段附近是水、氮、盐共同的特征波段，对南疆盐碱土水、氮、盐的最优反演方法是采用Savitzky-Golay方法平滑，运用主成分分析降维，通过偏最小二乘回归建立反演模型。 |
| 关键词: 土壤光谱特征；盐碱土；反演模型；土壤盐分；土壤氮含量；土壤水分 |
| DOI： |
| 分类号:TP79 |
| 基金项目:兵团财政科技计划项目（S2021BC1158），国家自然科学基金项目 (51879267，51669032)，兵团节水灌溉试验计划项目（BTJSSY-202210） |
|
| Spectral Characteristics of Water Nitrogen and Salt in Saline Soils of South Xinjiang and its Inversion Model |
|
Zhao Zeyi1, Gao Yang2, Li Zhaoyang3, Wang Hongbo3, Zhang Nan3, Li Guohui3, Tang Maosong3, Wang Xingpeng3
|
|
1.College of water Resource and Architecture Engineering,Tarim University;2.Farmland lrrigation Research lnstitute,Chinese Academy of Agricultural Sciences;3.Traim University
|
| Abstract: |
| 【Objective】 The main discussion is the spectral characteristics of saline soils in South Xinjiang under different water, nitrogen and salt contents, and the construction of a water, nitrogen and salt content inversion model suitable for South Xinjiang sandy soils, which provides a scientific basis for rapid detection of water, nitrogen and salt contents in its soils. 【Method】 Selecting saline sandy soil in South Xinjiang as the research object, and setting up different soil moisture, salt and nitrogen contents in order to obtain and analyze the soil spectral characteristics of different treatments. Meanwhile, soil water, nitrogen and salt inversion models were established by using partial least squares regression (PLSR), support vector regression (SVR) and BP neural network (BPNN) 【Results】 The characteristic bands of soil water are around 1400 and 1900 nm, the characteristic bands of soil nitrogen are between 1448-1515 and 1841-2500 nm, and the characteristic bands of soil salts are between 1878-1881 and 1908-1940 nm. The PLSR model has the best inversion for water, nitrogen and salt, BPNN the second and SVR the worst. 【Conclusion】 The characteristic band common to water, nitrogen and salt is around 1900 nm. The optimal inversion method for water, nitrogen, and salt of saline soils in South Xinjiang was smoothed by Savitzky-Golay method, using principal component analysis for dimensionality reduction and partial least squares regression to develop the inverse model. |
| Key words: soil spectral characteristics; saline soils; inversion model; soil salinity; soil nitrogen content; soil moisture |
