Abstract-The classification of tree species through remote sensing data is of great significance to monitoring forest disturbances, biodiversity assessment, and carbon estimation. The dense time series and a wide swath of Sentinel-2 data provided the opportunity to map tree species accurately and in a timely manner over a large area. Many current studies have applied machine learning (ML) algorithms combined with Sentinel-2 images to classify tree species, but it is still unclear which algorithm is more effective in the automotive extraction of tree species. In this study, five machine learning algorithms were compared to identify the composition of tree species with multi-temporal Sentinel-2 images in the JianShe forest farm, Northeast China. Three major types of deep neural networks (Conv1D, AlexNet and LSTM) were tested to classify Sentinel-2 time series, which represent three disparate but effective strategies to apply sequential data. The other two models are Support Vector Machine (SVM) and Random Forest (RF), which are renowned for extensive adoption and high performance for various remote sensing applications. The results show that the overall accuracy of neural network models is better than that of SVM and RF. The Conv1D model had the highest classification accuracy (84.19%), followed by the LSTM model (81.52%), and the AlexNet model (76.02%). For non-neural network models, RF's classification accuracy (79.04%) is higher than that of SVM (72.79%), but lower than that of Conv1D and LSTM. Therefore, the deep neural networks combined with multitemporal Sentinel-2 images can efficiently improve the accuracy of tree species classification.

No abstract

Abstract:The increasing global demands for land resource with increasing population have resulted in occurrence of soil degradation in many regions of the world. Assessment of soil quality has become the basic work for agricultural sustainable development and selecting regional indicators effectively has become very important since there are no standard evaluation methods and universal indicators. In this study, taking the Corn Belt of Northeast China as the study area, seven indicators-obstacle horizon thickness, cation exchange capacity, pH, soil organic matter, total nitrogen, total potassium, and available Fe-were selected to constitute the minimum data set from sixteen indictors of the total data set to assess the soil quality. The soil quality of the study area was dominated by moderate grade, increasing from west to east. The soil quality of Yushu, Changchun and Shuangyang had higher values, and that of Nongan was the lowest. We found that the distribution of cation exchange capacity has a good consistency with the assessment result of the soil quality. Black soils were distributed in the middle part of the study region from north to south and accounted for a higher quality, exactly where the areas of rapid urbanization are located. An ANOVA analysis showed that soil quality in the Corn Belt of Northeast China was greatly affected by topographic factors and agricultural management and climate was not the principal factor affecting soil quality. Though the minimum data set slightly reduced the evaluation accuracy, a large sampling density in our study was able to improve the precision loss that resulted from reducing the number of indicators to a certain extent.