With the Lutou Forest Farm as the research area, the Lasso algorithm was used for characteristic selection, and the optimal combination of variables was input into the support vector regression (SVR) model. The most suitable SVR model was selected to estimate the aboveground biomass of the forest through the comparison of the kernel function and optimal parameters, and the spatial distribution map of the aboveground biomass in the study area was drawn. The significance analysis of special variables showed good correlations between forest aboveground biomass and each vegetation index. There was a more significant correlation with some remote sensing bands, a less significant correlation with some texture features, and a strong correlation with DEM in the terrain features. When the parameters C is 2 and g is 0.01, the SVR model has the highest precision, which can illustrate 73% of the forest aboveground biomass, with the validation set $R^2$ being 0.62. The statistical analysis of the results shows that the total aboveground biomass of the Lutou Forest Farm is $4.82\times {10}^5$ t. The combination of Lasso with the SVR model can improve the estimation accuracy of forest aboveground biomass, and the model has a strong generalization ability.

Affected by global warming and increased extreme weather, Hunan Province saw a phased and concentrated outbreak of forest fires in 2022, causing significant damage and impact. Predicting the occurrence of forest fires can enhance the ability to make early predictions and strengthen early warning and responses. Currently, fire prevention and extinguishing in China’s forests and grasslands face severe challenges due to the overlapping of natural and social factors. Existing forest fire occurrence prediction models mostly take into account vegetation, topographic, meteorological and human activity factors; however, the occurrence of forest fires is closely related to the forest fuel moisture content. In this study, the traditional driving factors of forest fire such as satellite hotspots, vegetation, meteorology, topography and human activities from 2004 to 2021 were introduced along with forest fuel factors (vegetation canopy water content and evapotranspiration from the top of the vegetation canopy), and a database of factors for predicting forest fire occurrence was constructed. And a forest fire occurrence prediction model was built using machine learning methods such as the Random Forest model (RF), the Gradient Boosting Decision Tree model (GBDT) and the Adaptive Augmentation Model (AdaBoost). The accuracy of the models was verified using Area Under Curve (AUC) and four other metrics. The RF model with an AUC value of 0.981 was more accurate than all other models in predicting the probability of forest fire occurrence, followed by the GBDT (AUC = 0.978) and AdaBoost (AUC = 0.891) models. The RF model, which has the best accuracy, was selected to predict the monthly forest fire probability in Changsha in 2022 and combined with the Inverse Distance Weight Interpolation method to plot the monthly forest fire probability in Changsha. We found that the monthly spatial and temporal distribution of forest fire probability in Changsha varied significantly, with March, April, May, September, October, November and December being the months with higher forest fire probability. The highest probability of forest fires occurred in the central and northern regions. In this study, the core drivers affecting the occurrence of forest fires in Changsha City were found to be vegetation canopy evapotranspiration and vegetation canopy water content. The RF model was identified as a more suitable forest fire occurrence probability prediction model for Changsha City. Meanwhile, this study found that vegetation characteristics and combustible factors have more influence on forest fire occurrence in Changsha City than meteorological factors, and surface temperature has less influence on forest fire occurrence in Changsha City.

For the Sentinel-2 multispectral satellite image remote sensing data, due to the rich spatial information, the traditional water body extraction methods cannot meet the needs of practical applications. In this study, a random forest-based RF_16 optimal combination model algorithm is proposed to extract water bodies. The research process uses Sentinel-2 multispectral satellite images and DEM data as the basic data, collected 24 characteristic variable indicators (B2, B3, B4, B8, B11, B12, NDVI, MSAVI, B5, B6, B7, B8A, NDI45, MCARI, REIP, S2REP, IRECI, PSSRa, NDWI, MNDWI, LSWI, DEM, SLOPE, SLOPE ASPECT), and constructed four combined models with different input variables. After analysis, it was determined that RF_16 was the optimal combination for extracting water body information in the study area. Model. The results show that: (1) The characteristic variables that have an important impact on the accuracy of the model are the improved normalized difference water index (MNDWI), band B2 (Blue), normalized water index (NDWI), B4 (Red), B3 (Green), and band B5 (Vegetation Red-Edge 1); (2) The water extraction accuracy of the optimal combined model RF_16 can reach 93.16%, and the Kappa coefficient is 0.8214. The overall accuracy is 0.12% better than the traditional Relief F algorithm. The RF_16 method based on the optimal combination model of random forest is an effective means to obtain high-precision water body information in the study area. It can effectively reduce the “salt and pepper effect” and the influence of mixed pixels such as water and shadows on the water extraction accuracy.

In order to locate forest fire smoke more precisely and expand existing forest fire monitoring methods, this research employed Himawari-8 data with a sub-pixel positioning concept in smoke detection. In this study, Himawari-8 data of forest fire smoke in Xichang and Linzhi were selected. An improved sub-pixel mapping method based on random forest results was proposed to realize the identification and sub-pixel positioning of smoke. More spatial details of forest fire smoke were restored in the final results. The continuous monitoring of smoke indicated the dynamic changes therein. The accuracy evaluation of smoke detection was realized using a confusion matrix. Based on the improved sub-pixel mapping method, the overall accuracies were 87.95% and 86.32%. Compared with the raw images, the smoke contours of the improved sub-pixel mapping results were clearer and smoother. The improved sub-pixel mapping method outperforms traditional classification methods in locating smoke range. Moreover, it especially made a breakthrough in the limitations of the pixel scale and in realizing sub-pixel positioning. Compared with the results of the classic PSA method, there were fewer “spots” and “holes” after correction. The final results of this study show higher accuracies of smoke discrimination, with it becoming the basis for another method of forest fire monitoring.

The formation of forest fire burned area, influenced by a variety of factors such as meteorology, topography, vegetation, and human intervention, is a dynamic process of fire line burning that develops from the point of ignition to the boundary of the burned area. Accurately simulating and predicting this dynamic process can provide a scientific basis for forest fire control and suppression decisions. In this study, five typical forest fires located in different regions of China were used as the study object. The straight path distances from the ignition point grid to each grid on fire line in Sentinel-2 imageries for each forest fire were used as the target variables. We obtained the values of 11 independent variables for each pathway, including wind speed component, Temperature, Relative Humidity, Elevation, Slope, Aspect, Degree of Relief, Normalized Difference Vegetation Index, Vegetation Type, Fire Duration, and Gross Domestic Product reflecting human intervention capacity for fires. The value of each target variable and that of its corresponding independent variable constituted a sample. Four machine learning models, such as Random Forest (RF), Gradient Boosting Decision Trees (GBDT), Support Vector Machine (SVM), and Multilayer Perceptron (MLP), were trained using 80% effective samples from four forest fires, and 20% used to verify the above models. The hyper-parameters of each model were optimized using grid search method. After analyzing the validation results of models which showed temperature as a non-significant variable, the training and validation process of models above was repeated after excluding temperature. The results show that RF is the optimal model with 49.55 m for root mean square error (RMSE), 29.19 m for mean absolute error (MAE) and 0.9823 for coefficient of determination (R²). This study used the RF model to construct the shape of burned areas by predicting lengths of all straight path distances from the ignition point to the fire line. The study can dynamically capture the development of forest fire scenes.