Forest Height Mapping Using Feature Selection and Machine Learning by Integrating Multi-Source Satellite Data in Baoding City, North China

Abstract

Accurate estimation of forest height is crucial for the estimation of forest aboveground biomass and monitoring of forest resources. Remote sensing technology makes it achievable to produce high-resolution forest height maps in large geographical areas. In this study, we produced a 25 m spatial resolution wall-to-wall forest height map in Baoding city, north China. We evaluated the effects of three factors on forest height estimation utilizing four types of remote sensing data (Sentinel-1, Sentinel-2, ALOS PALSAR-2, and SRTM DEM) with the National Forest Resources Continuous Inventory (NFCI) data, three feature selection methods (stepwise regression analysis (SR), recursive feature elimination (RFE), and Boruta), and six machine learning algorithms (k-nearest neighbor (k-NN), support vector machine regression (SVR), random forest (RF), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and categorical boosting (CatBoost)). ANOVA was adopted to quantify the effects of three factors, including data source, feature selection method, and modeling algorithm, on forest height estimation. The results showed that all three factors had a significant influence. The combination of multiple sensor data improved the estimation accuracy. Boruta’s overall performance was better than SR and RFE, and XGBoost outperformed the other five machine learning algorithms. The variables selected based on Boruta, including Sentinel-1, Sentinel-2, and topography metrics, combined with the XGBoost algorithm, provided the optimal model (R2 = 0.67, RMSE = 2.2 m). Then, we applied the best model to create the forest height map. There were several discrepancies between the generated forest height map and the existing map product, and the values with large differences between the two maps were mostly distributed in the steep areas with high slope values. Overall, we proposed a methodological framework for quantifying the importance of data source, feature selection method, and machine learning algorithm in forest height estimation, and it was proved to be effective in estimating forest height by using freely accessible multi-source data, advanced feature selection method, and machine learning algorithm.

Continuous Mapping of Forest Canopy Height using ICESat-2 Data and a Weighted Kernel Integration of Multi-Temporal Multi-Source Remote Sensing Data aided by Google Earth Engine

“…Previous papers have reported the same conclusions (Luo et al, 2023;Xi et al, 2022). However, it is important to acknowledge that the comparison of ML models in estimating FCH has yielded different results in some previous studies as well (Yu et al, 2021;Zhang et al, 2022). This indicates that various factors, such as regional topography, the presence of diverse tree species, and the availability of diverse predictive features, can in uence the performance of ML models (Xi et al, 2022)…”

“…Following the completion of outlier detection procedure, a reliable dataset of 14,897 samples is obtained for the development of ML models. In this study, four ML models were utilized for FCH continuous mapping: KNN (Ahmed et al, 2015), SVR (Mountrakis et al, 2011), RF (Tiwari & Narine, 2022), and XGBoost (Zhang et al, 2022). These algorithms were employed to establish a relationship between the FCH reference values derived from ATL08 and the prepared data cube generated from the multi-source and multi-temporal RS data sources after WK convolution.…”

“…To accomplish this, a grid search methodology was utilized for parameter optimization. Through 10-fold cross-validation on the complete dataset, the parameters of the four ML models were tuned to achieve the lowest Root Mean Squared Error (RMSE) (Zhang et al, 2022). Table 2 summarizes the input parameters of the ML models and their corresponding search space for optimized parameter selection.…”

An Indoor Microwave Radiometer for Measurement of Tropospheric Water

“…An advantage of GBDT is that the relative importance of the features used by the model can be output after model training, which is often used for feature selection to understand which factors have a key impact on prediction [37]. Friedman [38] proposed the computation of GBDT feature selection.…”

Deep Learning Models in Forest Mapping Using Multitemporal SAR and Optical Satellite Data

“…To date, numerous methods and remotely sensed data combinations have been used for forest height estimation in boreal and temperate forests [5], [33], [37]- [39]. Reported accuracies for boreal forest height mapping range typically in the order of 30-40% rRMSE in these studies.…”