Search

• Vegetation communities in the largest karst wetland of China achieved fine classifications. • Fusion of JL101K and UAV multispectral images improved classification accuracy (1.9%–4.0%) • Classification performance of LightGBM algorithm outperformed XGBoost (0.6%–2.5%) and RF algorithms (1.6%–3.5%). • UAV multispectral images produced the highest overall accuracy (87.8%) in three data sources. • Red-edge band and DSM provide great contributions to identify vegetation communities. Fine classification of wetland vegetation communities using machine learning algorithm and high spatial resolution images have attracted increased attention. However, there exist several challenges in image fusion, data dimension reduction and algorithm tuning. To resolve these issues, this paper attempts to fuse Unmanned Aerial Vehicle (UAV) images with spaceborne Jilin-1 (JL101K) multispectral images for classifying vegetation communities of karst wetland using the optimized Random Forest (RF), Extreme gradient boosting (XGBoost) and Light Gradient Boosting (LightGBM) algorithms. This study also quantitatively evaluates image fusion quality from spatial detail and spectral fidelity, and explores the effects of different image feature combinations and classifiers on mapping vegetation communities by variable selection and dimensionality reduction. Finally, this paper further evaluates and quantifies the importance and contribution rate of feature variables for typical vegetation communities using Recursive feature elimination (RFE) algorithm. The results showed that: (1) the Gram-Schmidt (GS)algorithm produced the high-quality fusion image of JL101K and UAV, and the fusion image achieved higher overall accuracy (82.8%) than the original JL101K multispectral image; (2) UAV multispectral image and its derivatives (scheme 3) achieved the highest overall accuracy (87.8%) in all classification schemes; (3) The optimized object-based LightGBM algorithm outperformed XGBoost and RF algorithm, which provided an improvement of 0.6%∼3.5% in overall accuracy (OA). McNemar's test indicated that there existed significant differences in vegetation communities' classification between the three algorithms. (4) The average accuracy (AA) of vegetation communities in karst wetlands was mainly ranged from 60% to 90%. The water hyacinth and herbaceous vegetation were sensitive to the Mean Digital Surface Model (DSM) and Standard RedEdge band. [ABSTRACT FROM AUTHOR]

• IQI-NLS index performed better than other soil quality model in BSZJLC. • The determined MDS for BSZJLC included BD, CEC, SOM, AN, TK, ACu, and AFe. • The soil quality in study area was dominated by moderate grades. • Effects of environmental factors on soil quality has local characteristic. Soil quality assessment is an effective way to improve understanding of the heterogeneity of soil quality and to encourage the adoption of proper agricultural practices. The object of this paper is to select the most effective soil quality indices and the minimum data set (MDS) for the black soil zone of Jilin Province, China, and analyze the association between soil quality and the environmental control. Taking the croplands of the black soil zone of Jilin Province, China, as the study area, different combinations of scoring methods, indices, and data set methods were compared. A minimum data set, based on the integrated quality index derived from non-linear scoring methods (MDS IQI-NLS index), was then selected via principal components analysis to assess the soil quality of the study area. The determined minimum data set for the study area included bulk density, cation exchange capacity, soil organic matter, alkali-dissolvable N, total K, extractable Cu, and extractable Fe. The results showed that the soil quality in the study area was of dominantly moderate grades (grades II, III and IV), accounting for approximately 90% of the total as well as the minimum data set, and the soil quality index increased from the northwest to the southeast. Group analysis was applied because different combinations of environmental factors play different roles in different sub-regions. The analysis results showed that the most important factors affecting soil quality in group 1 (in decreasing order of importance) were: mean annual temperature, DEM, amount of straw return, parent material, and topographic wetness index. For group 2 they were: parent material, annual precipitation, mean annual temperature, chemical fertilizer, DEM, amount of straw return. This analysis proves to be an effective way to determine the factors that affect the spatial distribution of soil quality. [ABSTRACT FROM AUTHOR]