Establishing the relationship between the integrated multidimensional landscape pattern and stream water quality in subtropical agricultural catchments. (August 2021)

• Record Type: Journal Article
• Authors: Liu, Huanyao; Meng, Cen; Wang, Yi; Li, Yong; Li, Yuyuan; Liu, Xinliang; Wu, Jinshui
• Is Part Of: Ecological indicators. Volume 127(2021)
• Publication Date: 2021-08
• Subjects: Catchment environment -- Landscape -- Water quality -- Topography -- Soil type -- Land use; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2021.107781
• Languages: English
• ISSNs: 1470-160X

Abstract

Highlights: Landscape analyze needs to adopt a multidimensional perspective. Landscape composition and configuration affects stream water N and P levels. The most influential landscape composition and configuration for the N and P levels were identified. The composition and configuration of landscape predicted stream water NP levels well. Multidimensional landscape optimization benefits catchment environmental management. Abstract: Although a landscape is a multidimensional integrated synthesis, the relationship between land use and water quality was merely studied in relation to protecting the aquatic ecological service function from the 'landscape perspective'. This study analyzed the integrated landscape pattern by coupling land use, soil property, and topography and determined the stream water total nitrogen (TN) and total phosphorus (TP) in Chinese subtropical catchments from 2010 to 2017 to reveal and quantify the relationship between multidimensional landscape patterns and stream water N and P levels using the GIS and partial least squares regression (PLSR) techniques. The results suggested that the areal proportions of tea field and residential area in the Ultisols and on the first three slope categories (0–28.02°) and paddy field in the Ultisols and on slope category I (0–5.35°) were positively correlated with the TN and TP, while forest in the Ultisols and on slope category IV (28.02–80.30°) was negatively correlated with the TN (p ≤ 0.05). For the multidimensional landscape metrics, the Euclidean nearest-neighbor distance (ENN_MN) and interspersion and juxtaposition index (IJI) were positively correlated with the TN, while the ENN_MN was positively correlated with the TP ( p ≤ 0.05). The PLSR analysis explained the variations of TN and TP levels well (R 2 ≥ 0.66), in which the areal proportions of the tea field and residential area in the Ultisols and on the first three slope categories (0–28.02°) and the paddy field in the Ultisols and on slope category I (0–5.35°) were the important variables for both TN and TP; however, the areal proportions of forest in the Ultisols and on slope category IV (28.02–80.30°) and the IJL were only important to the TN. Therefore, the multidimensional landscape pattern providing specific land use, soil property, and topography can be more capable in restoring and protecting stream N and P pollution at an applicable level.