基于便携式高光谱近感快速监测内陆水体叶绿素a浓度OA

Phytoplankton chlorophyll-a(Chl.a)concentration is a key indicator for assessing water eutrophication status.Conven-tional monitoring approaches face significant limitations:laboratory analyses are time-consuming and labor-intensive,while in situ sensors are susceptible to biofouling,low accuracy,and high maintenance costs.Traditional satellite remote sensing is also unsuita-ble for high-precision,real-time monitoring due to challenges such as atmospheric correction errors,technical complexity,and low temporal resolution.The emergence of hyperspectral proximal sensing technology offers a promising alternative for improving Chl.a monitoring efficiency.In this study,we utilized a novel portable hyperspectral proximal sensing device to collect 533 synchronized in situ Chl.a measurements across eight lakes,reservoirs,and rivers between 2021 and 2024.We developed and compared high-accuracy Chl.a inversion models using both linear regression and machine learning methods.Among the evaluated algorithms—lin-ear regression,random forest,extreme gradient boosting(XGBoost),and support vector machine—the XGBoost-based model ex-hibited the best performance(R2=0.87,RMSE=6.02 μg/L,MAE=3.98 μg/L).This approach enables simultaneous spectral acquisition and Chl.a estimation,streamlining field monitoring workflows,lowering technical barriers,and significantly improving operational efficiency.