2013~2021廉江营仔镇红树林的分布变化
The Distribution Change of Mangrove in Yingzai Town, Lianjiang from 2013 to 2021

本研究基于Landsat-8卫星影像数据，对湛江廉江营仔镇的红树林分布变化进行了研究。选取了2013年、2017年和2021年的三期影像进行地物分类，提取红树林面积。采用面向对象分类方法进行地物类别分类。通过Google Earth高分辨率卫星影像数据及混淆矩阵进行精度评价，结果显示总体精度和Kappa系数分别为94.84%和0.91。研究结果表明，2013年红树林面积为715.50 ha，2017年为729.72 ha，2021年为742.50 ha。2013年至2021年红树林增长总面积为27.00 ha，年增长率为0.47%。红树林保持稳定的比例为90.05%，转变为水体的比例为4.15%，转变为陆地的比例为5.70%，转变为人工养殖区的比例为0.10%。水体转化为红树林占3.80%，陆地转化为红树林占0.49%，人工养殖区转化为红树林占0.10%。总体而言，营仔镇红树林的面积呈增长趋势，但增长率较低。明显增长区域共有5处。红树林分布较为稳定，且向水体扩张。本研究采用的面向对象分类方法和精度评价技术对遥感影像的分类和准确性评估具有参考价值，对相关领域的研究和应用具有推动作用。
This study investigated the distribution changes of mangroves in Yingzi Town, Lianjiang, Zhanjiang, based on Landsat-8 satellite imagery data. Three sets of images from 2013, 2017, and 2021 were selected for land cover classification and extraction of mangrove areas. The object-oriented classification method was employed for land cover classification. Accuracy assessment was performed using high-resolution satellite imagery data from Google Earth and a confusion matrix, with an overall accuracy of 94.84% and a Kappa coefficient of 0.91. The results of the study revealed that the man-grove area was 715.50 ha in 2013, 729.72 ha in 2017, and 742.50 ha in 2021. The total increase in mangrove area from 2013 to 2021 was 27.00 ha, with an annual growth rate of 0.47%. Among the mangrove areas, 90.05% remained stable, 4.15% converted to water bodies, 5.70% converted to land, and 0.10% converted to aquaculture areas. Additionally, 3.80% of water bodies transformed into mangroves, 0.49% of land transformed into mangroves, and 0.10% of aquaculture areas transformed into mangroves. Overall, the mangrove area in Yingzi Town showed an increasing trend, but with a low growth rate. Five distinct areas exhibited significant growth, and the distribution of mangroves remained relatively stable while expanding towards water bodies. The adopted object-oriented classification method and accuracy assessment techniques in this study hold reference value for the classification and accuracy evaluation of remote sensing imagery. They can contribute to research and applications in related fields by driving advancements and improvements.