本文以叶绿素a（Chl-a）的低成本、高精度估算为目标，利用近十年南海西部调查航次数据，基于随机森林（Random Forest, RF）算法，以水文气象对海洋生化过程的影响及贡献为基础，以水文气象参数（Hydro-Meteorological Parameters，HMPs）作为输入数据，构建低成本Chl-a的估算模型对南海西部表层Chl-a进行了估算。为验证基于水文气象参数估算Chl-a的可靠性，利用QAA算法以实测固有光学特性参数为基础，推导得到原位遥感反射率（remote sensing reflectance, R_rs）。在此基础上，结合OC4、Aiken和Tassan等经典水色产品经验算法对 Chl-a进行了估算及评价，评价结果表明OC4算法的估算精度最高，R²可达0.438。与RF模型0.568的R²比较不难看出，得益于HMPs的大数据量，基于HMPs的RF模型其Chl-a估算结果表现出较为优秀的稳定性和泛化性，与实测结果的空间分布一致性更好。通过对特征参数重要性进行研究发现，盐度是基于HMPs估算Chl-a的机器学习模型中最重要的特征变量，其次依次是温度、风与气压，贡献率最低的是相对湿度。

With the goal of low-cost and high-accuracy estimation of Chlorophyll-a (Chl-a), a model for estimating Chl-a in the surface layer of the Western South China Sea (WSCS) was constructed in this study. Using the data from the WSCS cruises in the past ten years, based on the influence and contribution of hydro-meteorology conditions to the oceanic biochemical processes, the Hydro-Meteorological Parameters (HMPs) were used as the input data of Random Forest (RF) algorithm. To evaluate the reliability of estimating Chl-a based on HMPs, the QAA algorithm was used to derive the in-situ remote sensing reflectance (R_rs) based on the measured inherent optical property parameters. Then Chl-a was estimated and evaluated by the combination of classical empirical algorithms for water color products such as OC4, Aiken and Tassan, the evaluation results showed that the OC4 algorithm had the highest estimation accuracy, with an R² of up to 0.438. The comparison with the R² of 0.568 of the RF-based model shows that, owing to the large data volume of HMPs, the Chl-a estimation results of the RF model based on HMPs show more excellent stability and generalization, and better spatial distribution consistency with the measured results. The importance of feature parameters was found that in the machine learning model for estimating Chl-a based on HMPs, salinity is the most important feature variable, followed by temperature, wind and air pressure in that order, and the lowest contributor is relative humidity.