United Nations University Institute for the Advanced Study of Sustainability (UNU-IAS), 5 Chome-53-70 Jingumae, Shibuya-Ku, Tokyo, 150-8925, Japan.
Institute for Future Initiatives (IFI), University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku, Tokyo, 113-8654, Japan.
Environ Sci Pollut Res Int. 2021 May;28(20):25920-25938. doi: 10.1007/s11356-020-12285-8. Epub 2021 Jan 21.
This paper aims to assess the influence of land use and land cover (LULC) indicators and population density on water quality parameters during dry and rainy seasons in a tourism area in Indonesia. This study applies least squares regression (OLS) and Pearson correlation analysis to see the relationship among factors, and all LULC and population density were significantly correlated with most of water quality parameter with P values of 0.01 and 0.05. For example, DO shows high correlation with population density, farm, and built-up in dry season; however, each observation point has different percentages of LULC and population density. The concentration value should be different over space since watershed characteristics and pollutions sources are not the same in the diverse locations. The geographically weighted regression (GWR) analyze the spatially varying relationships among population density, LULC categories (i.e., built-up areas, rice fields, farms, and forests), and 11 water quality indicators across three selected rivers (Ayung, Badung, and Mati) with different levels of tourism urbanization in Bali Province, Indonesia. The results explore that compared with OLS estimates, GWR performed well in terms of their R values and the Akaike information criterion (AIC) in all the parameters and seasons. Further, the findings exhibit population density as a critical indicator having a highly significant association with BOD and E. Coli parameters. Moreover, the built-up area has correlated positively to the water quality parameters (Ni, Pb, KMnO and TSS). The parameter DO is associated negatively with the built-up area, which indicates increasing built-up area tends to deteriorate the water quality. Hence, our findings can be used as input to provide a reference to the local governments and stakeholders for issuing policy on water and LULC for achieving a sustainable water environment in this region.
本文旨在评估土地利用和土地覆盖（LULC）指标以及人口密度对印度尼西亚旅游区干湿季水质参数的影响。本研究采用最小二乘法回归（OLS）和皮尔逊相关分析来观察因素之间的关系，所有 LULC 和人口密度与大多数水质参数均呈显著相关，P 值均为 0.01 和 0.05。例如，在旱季，DO 与人口密度、农田和建成区高度相关；然而，每个观测点的 LULC 和人口密度的百分比不同。由于流域特征和污染源在不同地点不同，因此空间上的浓度值应该不同。地理加权回归（GWR）分析了人口密度、LULC 类别（即建成区、稻田、农场和森林）与 11 个水质指标在印度尼西亚巴厘省三个选定河流（阿勇河、巴东河和马蒂河）之间的空间变化关系，这些河流具有不同程度的旅游城市化。结果表明，与 OLS 估计相比，GWR 在所有参数和季节的 R 值和赤池信息量准则（AIC）方面表现良好。此外，研究结果表明，人口密度是一个关键指标，与 BOD 和大肠杆菌参数高度相关。此外，建成区与水质参数（镍、铅、高锰酸钾和 TSS）呈正相关。参数 DO 与建成区呈负相关，这表明建成区的增加往往会恶化水质。因此，我们的研究结果可以作为提供给地方政府和利益相关者的参考，以制定有关水和 LULC 的政策，实现该地区可持续的水环境。
