Machine learning algorithms, in particular, kernel-based machine learning methods such as Gaussian processes regression (GPR) have shown to be promising alternatives to traditional empirical methods for retrieving vegetation parameters from remotely sensed data. However, the performance of GPR in predicting forest biophysical parameters has hardly been examined using full-spectrum airborne hyperspectral data. The main objective of this study was to evaluate the potential of GPR to estimate forest leaf area index (LAI) using airborne hyperspectral data. To achieve this, field measurements of LAI were collected in the Bavarian Forest National Park (BFNP), Germany, concurrent with the acquisition of the Fenix airborne hyperspectral images (400−2500 nm) in July 2017. The performance of GPR was further compared with three commonly used empirical methods (i.e., narrowband vegetation indices (VIs), partial least square regression (PLSR), and artificial neural network (ANN)). The cross-validated coefficient of determination (R_cv²) and root mean square error (RMSE_cv) between the retrieved and field-measured LAI were used to examine the accuracy of the respective methods. Our results showed that using the entire spectral data (400−2500 nm), GPR yielded the most accurate LAI estimation (R_cv² = 0.67, RMSE_cv = 0.53 m² m⁻²) compared to the best performing narrowband VIs SAVI2 (R_cv² = 0.54, RMSE_cv = 0.63 m² m⁻²), PLSR (R_cv² = 0.74, RMSE_cv = 0.73 m² m⁻²) and ANN (R_cv² = 0.68, RMSE_cv = 0.54 m² m⁻²). Consequently, when a spectral subset obtained from the analysis of VIs was used as model input, the predictive accuracies were generally improved (GPR RMSE_cv = 0.52 m² m⁻²; ANN RMSE_cv = 0.55 m² m⁻²; PLSR RMSE_cv = 0.69 m² m⁻²), indicating that extracting the most useful information from vast hyperspectral bands is crucial for improving model performance. In general, there was an agreement between measured and estimated LAI using different approaches (p > 0.05). The generated LAI map for BFNP using GPR and the spectral subset endorsed the LAI spatial distribution across the dominant forest classes (e.g., deciduous stands were generally associated with higher LAI values). The accompanying LAI uncertainty map generated by GPR shows that higher uncertainties were observed mainly in the regions with low LAI values (low vegetation cover) and forest areas which were not well represented in the collected sample plots. This study demonstrated the potential of GPR for estimating LAI in forest stands using airborne hyperspectral data. Owing to its capability to generate accurate predictions and associated uncertainty estimates, GPR is evaluated as a promising candidate for operational retrieval applications of vegetation traits.

机器学习算法，尤其是基于内核的机器学习方法，例如高斯过程回归（GPR），已显示出是从遥感数据中检索植被参数的传统经验方法的有希望的替代方法。但是，几乎没有使用全光谱机载高光谱数据检查GPR在预测森林生物物理参数方面的性能。这项研究的主要目的是评估GPR使用机载高光谱数据估算森林叶面积指数（LAI）的潜力。为了实现这一目标，2017年7月在德国巴伐利亚森林国家公园（BFNP）收集了LAI的现场测量数据，同时还获得了Fenix机载高光谱图像（400-2500 nm）。GPR的性能与三种常用的经验方法（即，窄带植被指数（VIs），偏最小二乘回归（PLSR）和人工神经网络（ANN））进行了比较。交叉验证的确定系数（R _cv ²）和取回的LAI和实地测量的LAI之间的均方根误差（RMSE _cv）用于检验相应方法的准确性。我们的结果表明，使用整个频谱数据（400-2500纳米），GPR得到最准确的估计LAI（ř _CV ² = 0.67，RMSE _CV =0.53米²米^-2相比表现最佳的窄带的VI SAVI2（）- [R _cv ² = 0.54，RMSE _cv = 0.63 m ² m ^-2），PLSR（R _cv ² = 0.74，RMSE _cv = 0.73 m ²m ^-2）和ANN（R _cv ² = 0.68，RMSE _cv = 0.54 m ² m ^-2）。因此，将通过VI分析获得的光谱子集用作模型输入时，预测精度通常得到改善（GPR RMSE _cv = 0.52 m ² m ^-2； ANN RMSE _cv = 0.55 m ² m ^-2； PLSR RMSE _cv = 0.69 m ² m ^-2），表明从宽高光谱带中提取最有用的信息对于改善模型性能至关重要。一般而言，使用不同方法测得的LAI与估计的LAI之间存在一致性（p> 0.05）。使用GPR和光谱子集为BFNP生成的LAI图谱认可了主要森林类别的LAI空间分布（例如，落叶林通常与较高的LAI值相关联）。由GPR生成的伴随的LAI不确定性图表明，主要在LAI值低（植被覆盖率低）的区域和未在所收集的样地中很好地表示森林的区域中观察到较高的不确定性。这项研究证明了GPR在使用机载高光谱数据估算林分中的LAI方面的潜力。