Géant Chuma B, Gustave Mushagalusa N, Schmitz Serge
Faculty of Agriculture and Environmental Sciences, Université Evangélique en Afrique (UEA), P.O Box: 3323, Bukavu, Democratic Republic of the Congo.
Department of Geography, University of Liège, UR SPHERES-Laplec, Bât. B11, Quartier Village 4, Clos Mercator 3, Liège, Belgium.
Sci Rep. 2023 Oct 17;13(1):17626. doi: 10.1038/s41598-023-43292-7.
There are several techniques for mapping wetlands. In this study, we examined four statistical models to assess the potential distribution of wetlands in the South-Kivu province by combining optical and SAR images. The approach involved integrating topographic, hydrological, and vegetation indices into the four most used classifiers, namely Artificial Neural Network (ANN), Random Forest (RF), Boosted Regression Tree (BRT), and Maximum Entropy (MaxEnt). A wetland distribution map was generated and classified into 'wetland' and 'non-wetland.' The results showed variations in predictions among the different models. RF exhibited the most accurate predictions, achieving an overall classification accuracy of 95.67% and AUC and TSS values of 82.4%. Integrating SAR data improved accuracy and precision, particularly for mapping small inland wetlands. Our estimations indicate that wetlands cover approximately 13.5% (898,690 ha) of the entire province. BRT estimated wetland areas to be ~ 16% (1,106,080 ha), while ANN estimated ~ 14% (967,820 ha), MaxEnt ~ 15% (1,036,950 ha), and RF approximately ~ 10% (691,300 ha). The distribution of these areas varied across different territories, with higher values observed in Mwenga, Shabunda, and Fizi. Many of these areas are permanently flooded, while others experience seasonal inundation. Through digitization, the delineation process revealed variations in wetland areas, ranging from tens to thousands of hectares. The geographical distribution of wetlands generated in this study will serve as an essential reference for future investigations and pave the way for further research on characterizing and categorizing these areas.
有几种绘制湿地地图的技术。在本研究中，我们通过结合光学和合成孔径雷达（SAR）图像，研究了四种统计模型，以评估南基伍省湿地的潜在分布。该方法包括将地形、水文和植被指数整合到四种最常用的分类器中，即人工神经网络（ANN）、随机森林（RF）、增强回归树（BRT）和最大熵（MaxEnt）。生成了一幅湿地分布图，并将其分为“湿地”和“非湿地”。结果显示不同模型的预测存在差异。随机森林表现出最准确的预测，总体分类准确率达到95.67%，曲线下面积（AUC）和真技能统计（TSS）值为82.4%。整合SAR数据提高了准确性和精度，特别是在绘制小型内陆湿地方面。我们的估计表明，湿地覆盖了全省约13.5%（898,690公顷）的面积。增强回归树估计湿地面积约为16%（1,106,080公顷），而人工神经网络估计约为14%（967,820公顷），最大熵估计约为15%（1,036,950公顷），随机森林估计约为10%（691,300公顷）。这些区域的分布在不同地区有所不同，在温加、沙本达和菲齐的湿地面积值较高。其中许多地区常年被洪水淹没，但其他地区则经历季节性淹没。通过数字化，划定过程揭示了湿地面积的变化，范围从几十公顷到数千公顷不等。本研究中生成的湿地地理分布图将为未来的调查提供重要参考，并为进一步研究这些区域的特征和分类铺平道路。
