Recently, convolutional neural networks (CNNs) showed excellent performance in many tasks, such as computer vision and remote sensing semantic segmentation. Especially, the ability to learn high-representation features of CNN draws much attention. And random forest (RF) algorithm, on the other hand, is widely applied for variables selection, classification, and regression. Based on the previous fusion models that fused CNN with the other models, such as conditional random fields (CRFs), support vector machine (SVM), and RF, this article tested a method based on the fusion of an RF classifier and the CNN for a very high resolution remote sensing (VHRRS) based forests mapping. The study area is located in the south of China and the main purpose was to precisely distinguish Lei bamboo forests from the other subtropical forests. The main novelties of this article are as follows. First, a test was conducted to confirm if a fusion of CNN and RF make an improvement in the VHRRS information extraction. Second, based on RF, variables with high importance were selected. Then, a test was again conducted to confirm if the learning from the selected variables will further give better results.

Above-ground biomass (AGB) directly relates to the productivity of forests. Precisely, AGB mapping for regional forests based on very high resolution (VHR) imagery is widely needed for evaluation of productivity. However, the diversity of variables and algorithms and the difficulties inherent in high resolution optical imagery make it complex. In this paper, we explored the potentials of the state-of-art algorithm convolutional neural networks (CNNs), which are widely used for its high-level representation, but rarely applied for AGB estimation. Four experiments were carried out to compare the performance of CNNs and other state-of-art Machine Learning (ML) algorithms: (1) performance of CNN using bands, (2) performance of Random Forest (RF), support vector regression (SVR), artificial neural network (ANN) on bands, and vegetation indices (VIs). (3) Performance of RF, SVR, and ANN on gray-level co-occurrence matrices (GLCM), and exploratory spatial data analysis (ESDA), and (4) performance of RF, SVR, and ANN based on all combined data and ESDA+VIs. CNNs reached satisfactory results (with R² = 0.943) even with limited input variables (i.e., only bands). In comparison, RF and SVR with elaborately designed data obtained slightly better accuracy than CNN. For examples, RF based on GLCM textures reached an R² of 0.979 and RF based on all combined data reached a close R² of 0.974. However, the results of ANN were much worse (with the best R² of 0.885).

Subtropical forests have great potential as carbon sinks; however, the relationship between net ecosystem productivity (NEP) and climate change is still unclear. This study took Zhejiang Province, a subtropical region, as an example. Based on remote sensing classification data of forest resources, the integrated terrestrial ecosystem carbon cycle (InTEC) model was used to simulate the spatiotemporal dynamics of the forest NEP in Zhejiang Province during 1985–2015 and analyze its response to meteorological factors such as temperature, precipitation, relative humidity, and radiation. Three patterns emerged: (1) The optimized InTEC model can better simulate the forest NEP in Zhejiang Province, and the correlation coefficient between the simulated NEP and observed NEP was up to 0.75. (2) From 1985 to 2015, the increase in the total NEP was rapid, with an average annual growth rate of 1.52 Tg·C·yr⁻¹. During 1985–1988, the forests in Zhejiang Province were carbon sources. After 1988, the forests turned into carbon sinks and this continued to increase. During 2000–2015, more than 97% of the forests in Zhejiang Province were carbon sinks. The total NEP reached 32.02 Tg·C·yr⁻¹, and the annual mean NEP increased to 441.91 gC·m⁻²·yr⁻¹. The carbon sequestration capacity of forests in the east and southwest of Zhejiang Province is higher than that in the northeast of Zhejiang Province. (3) From 2000 to 2015, there was an extremely significant correlation between forest NEP and precipitation, with a correlation coefficient of 0.85. Simultaneously, the forest NEP showed a negative correlation with temperature and radiation, with a correlation coefficient of −0.56 for both, and the forest NEP was slightly negatively correlated with relative humidity. The relative contribution rates of temperature, precipitation, relative humidity, and radiation data to NEP showed that the contribution of precipitation to NEP is the largest, reaching 61%, followed by temperature and radiation at 18% and 17%, respectively. The relative contribution rate of relative humidity is the smallest at only 4%. During the period of 1985–1999, due to significant man-made disturbances, the NEP had a weak correlation with temperature, precipitation, relative humidity, and radiation. The results of this study are important for addressing climate change and illustrating the response mechanism between subtropical forest NEP and climate change.

Dynamic monitoring of carbon storage in forests resources is important for tracking ecosystem functionalities and climate change impacts. In this study, we used multi-year Landsat data combined with a Random Forest (RF) algorithm to estimate the forest aboveground carbon (AGC) in a forest area in China (Hang-Jia-Hu) and analyzed its spatiotemporal changes during the past two decades. Maximum likelihood classification was applied to make land-use maps. Remote sensing variables, such as the spectral band, vegetation indices, and derived texture features, were extracted from 20 Landsat TM and OLI images over five different years (2000, 2004, 2010, 2015, and 2018). These variables were subsequently selected according to their importance and subsequently used in the RF algorithm to build an estimation model of forest AGC. The results showed the following: (1) Verification of classification results showed maximum likelihood can extract land information effectively. Our land cover classification yielded overall accuracies between 86.86% and 89.47%. (2) Additionally, our RF models showed good performance in predicting forest AGC, with R² from 0.65 to 0.73 in the training and testing phase and a RMSE range between 3.18 and 6.66 Mg/ha. RMSEr in the testing phase ranged from 20.27 to 22.27 with a low model error. (3) The estimation results indicated that forest AGC in the past two decades increased with density at 10.14 Mg/ha, 21.63 Mg/ha, 26.39 Mg/ha, 29.25 Mg/ha, and 44.59 Mg/ha in 2000, 2004, 2010, 2015, and 2018. The total forest AGC storage had a growth rate of 285%. (4) Our study showed that, although forest area decreased in the study area during the time period under study, the total forest AGC increased due to an increment in forest AGC density. However, such an effect is overridden in the vicinity of cities by intense urbanization and the loss of forest covers. Our study demonstrated that the combined use of remote sensing data and machine learning techniques can improve our ability to track the forest changes in support of regional natural resource management practices.

Analysis of urban land use dynamics is essential for assessing ecosystem functionalities and climate change impacts. The focus of this study is on monitoring the characteristics of urban expansion in Hang-Jia-Hu and evaluating its influences on forests by applying 30-m multispectral Landsat data and a machine learning algorithm. Firstly, remote sensed images were preprocessed with radiation calibration, atmospheric correction and topographic correction. Then, the C5.0 decision tree was used to establish classification trees and then applied to make land use maps. Finally, spatiotemporal changes were analyzed through dynamic degree and land use transfer matrix. In addition, average land use transfer probability matrix (ATPM) was utilized for the prediction of land use area in the next 20 years. The results show that: (1) C5.0 decision tree performed with precise accuracy in land use classification, with an average total accuracy and kappa coefficient of more than 90.04% and 0.87. (2) During the last 20 years, land use in Hang-Jia-Hu has changed extensively. Urban area expanded from 5.84% in 1995 to 21.32% in 2015, which has brought about enormous impacts on cultivated land, with 198,854 hectares becoming urban, followed by forests with 19,823 hectares. (3) Land use area prediction based on the ATPM revealed that urbanization will continue to expand at the expense of cultivated land, but the impact on the forests will be greater than the past two decades. Rationality of urban land structure distribution is important for economic and social development. Therefore, remotely sensed technology combined with machine learning algorithms is of great significance to the dynamic detection of resources in the process of urbanization.

The highly accurate multiresolution leaf area index (LAI) is an important parameter for carbon cycle simulation for bamboo forests at different scales. However, current LAI products have discontinuous resolution with 1 km mostly, that makes it difficult to accurately quantify the spatiotemporal evolution of carbon cycle at different resolutions. Thus, this study used MODIS LAI product (MOD15A2) and MODIS reflectance data (MOD09Q1) of Moso bamboo forest (MBF) from 2015, and it adopted a hierarchical Bayesian network (HBN) algorithm coupled with a dynamic LAI model and the PROSAIL model to obtain high-precision LAI data at multiresolution (i.e., 1000, 500, and 250 m). The results showed the LAIs assimilated using the HBN at the three resolutions corresponded with the actual growth trend of the MBF and correlated significantly with the observed LAI with a determination coefficient (R²) value of >0.80. The highest-precision assimilated LAI was obtained at 1000-m resolution with R² values of 0.91. The LAI assimilated using the HBN algorithm achieved better accuracy than the MODIS LAI with increases in the R² value of 2.7 times and decreases in the root mean square error of 87.8%. Therefore, the HBN algorithm applied in this study can effectively obtain highly accurate multiresolution LAI time series data for bamboo forest.