Landscape Characteristics Affecting Spatial Patterns of Water Quality Variation in a Highly Disturbed Region.

Gang Xie,Yi Zhu,Huijian Shi,Hongqi Wang,Xinqi Hu

doi:10.3390/ijerph16122149

Abstract

Spatial patterns of water quality trends for 45 stations in control units of the Shandong Province, China during 2009–2017 were examined by a non-parametric seasonal Mann-Kendall’s test (SMK) for dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD), permanganate index (CODMn), total phosphorus (TP) and ammonia nitrogen (NH3-N). The DO concentration showed significant upward trends at approximately half of the stations, while other parameters showed significant downward trends at more than 40% of stations. The stations with downward trends presented significant spatial autocorrelation, and were mainly concentrated in the northwest and southwest regions. The relationship between the landscape characteristics and water quality was explored using stepwise multiple regression models, which indicated the water quality was better explained using landscape pattern metrics compared to the percentage of land use types. Decreased mean patch area and connectedness of farmland will promote the control of BOD, COD and CODMn, whereas the increased landscape percentage of urban areas were not conducive to the water quality improvement, which suggested the sprawling of farmland and urban land was not beneficial to pollution control. Increasing the grassland area was conducive to the reduction of pollutants, while the effect of grassland fragmentation was reversed.

Highlights

The increasing economic growth, rapid urbanization and higher population concentrated in cities of developing countries had led to deterioration of its water bodies [1]
The stations showing downward trend of biochemical oxygen demand (BOD), chemical oxygen demand (COD), CODMn and NH3 -N were mainly concentrated in the northwest and southeast, especially the area surrounding the Nansi Lake, which may be attributed to the continuous and effective water pollution control actions implemented by the national and local government since 2000, including the 10th, 11th and 12th Five-Year Major Science and Technology Program for Water Pollution Control and Treatment Plan from 2001 to 2015, which were aimed at improving the water environment of several major basins in China, including the Hai River
Different from previous similar studies on sub-basin and buffer scales [68], as shown in Table S1, we found out landscape pattern metrics dominate higher relevance to water quality parameters than percentage of land use types as indicated by the generally higher R2 values, which was the unique conclusion we got about the relationship between landscape features and water quality at the control unit scale in Shandong province

Summary

Introduction

The increasing economic growth, rapid urbanization and higher population concentrated in cities of developing countries had led to deterioration of its water bodies [1]. This water quality deterioration resulted due to point sources and non-point sources pollutants entering the water bodies, i.e., lakes, rivers and streams. Due to the dynamic changes in pollution source emissions and the spatial heterogeneity of the underlying surface, surface water quality presents dynamic and regional differences in time and space [2,3,4]. The evolution trend of water quality data in long time series can be detected by using appropriate trend analysis method [5,6]. Non-parametric tests have been widely used to detect temporal trends in environmental and hydrological data, including surface water quality concentrations [7,8,9,10]

Objectives

Methods

Results

Discussion

Conclusion