Natural and anthropogenic imprints on seasonal river water quality trends across China

• Haoran Zhang, 
• Huihang Sun, 
• Jiarong Li, 
• Yuelei Li, 
• Luyu Zhang, 
• Ruikun Zhao, 
• Xiangang Hu, 
• Nanqi Ren & 
• Yu Tian 

npj Clean Water volume 8, Article number: 49 (2025) Cite this article

• 4972 Accesses
• 6 Citations
• Metricsdetails

Abstract

Climate change and human activities have redefined seasonal river water quality patterns, yet their respective impacts remain unclear. Here, we propose a novel trend-based metric, the T-NM index, to isolate asymmetric human amplification and suppression effects across 195 natural and 1540 managed watersheds in China (2006–2020). Consistent trends in 52–89% of watersheds suggest climatic dominance, while anthropogenic drivers intensified or attenuated trends by 22–158% and 14–56%, especially in summer. Four independent multivariable models simulated seasonal COD and DO concentrations. Attribution analysis showed that seasonal factors explained 47.08% of the variation, while rainfall (25.37%) and slope (17.40%) accounted for COD and DO changes in natural watersheds; in contrast, Shannon Diversity Index (11.58%) and Largest Patch Index (10.66%) dominated in managed watersheds. This study establishes a generalizable framework for distinguishing natural and anthropogenic influences, offering key insights for adaptive water quality management under future climatic and socio-economic transitions.

Similar content being viewed by others

The human factor in seasonal streamflows across natural and managed watersheds of North America

Past and future change in global river flows

Changes in China’s river water quality since 1980: management implications from sustainable development

Introduction

The fluctuations in inland water bodies (quantity) and their complex interactions with physical, chemical, and biological characteristics (quality) are fundamental to sustaining life on Earth. Among them, water quality is sensitive to both natural and anthropogenic disturbances¹. On the one hand, climate change greatly impacts river water quality via both gradual and episodic patterns². Biogeochemical processes in rivers are significantly disrupted by anthropogenic activities such as population growth, land development, and agricultural production³. It is estimated that more than 50% of global river systems are no longer classified as natural⁴. Against this backdrop, comprehending and systematically quantifying the effects of natural and anthropogenic drivers on water quality dynamics is critically important. This understanding is pivotal not only for the supply and safety of water for human consumption but also for food production, the energy supply and the overall health and stability of aquatic ecosystems⁵.

Currently, most studies focus on long-term water quality trends on continental or regional scales^6,7, with fewer studies addressing seasonal trends, primarily in remote, natural watersheds. These studies emphasize the impact of large-scale climate change on river systems, confirming that seasonal climatic variability is closely linked to the migration, transformation, and biochemical reaction rates of water pollutants⁸. However, the influence of anthropogenic interventions on seasonal trends is greater than that on interannual trends. For example, land management measures such as road construction, farmland conservation, and reforestation are typically implemented on a subannual scale, and can influence the seasonal dynamics of river water quality indicators by regulating surface runoff generation, controlling soil erosion, and altering the transport of organic matter⁹. Agricultural activities such as fertilization and irrigation exhibit clear seasonal dynamics, with precipitation exacerbating the uneven effects of nutrient runoff¹⁰. Industrial water use and wastewater discharge vary according to the market demand by season¹¹. Consequently, human activities drive biogeochemical processes involving oxygen, nutrient, and sediment cycling, profoundly altering seasonal river water quality dynamics.

Water quality dynamics reflect both physical processes, such as river–atmosphere gas exchange, and biological processes, including photosynthetic oxygen production and respiratory oxygen consumption. However, systematic attribution analyzes at the national scale are constrained by factors such as the scarcity of meteorological, terrestrial, and landscape gradient sites and inconsistent data collection frequencies¹². Some researchers attribute river changes to climate forcing¹³, whereas others emphasize the role of anthropogenic pressures^14,15. Li et al. proposed an integrative mechanistic perspective that extends beyond singular effects, emphasizing the impact of dynamic changes in geomorphology, structure, and material and energy flows within river–terrestrial systems¹⁶. These processes are shaped and regulated by both natural factors and human activities. Therefore, it is essential to transcend traditional watershed boundaries and systematically assess the relative impacts of natural factors and human activities on river health.

China’s river basins, marked by diverse hydrology, variable climate, and uneven anthropogenic pressures, face concurrent challenges from agricultural intensification, industrial pollution, and urban expansion¹⁷. These characteristics make them ideal for studying water quality dynamics and coupled human–nature responses. We analyzed the interannual and seasonal variations in COD and DO concentrations across China during the period 2006–2020, elucidating the differences in seasonal water quality variations between natural and managed watersheds. Among them, COD and DO are the most nationally representative water quality parameters for identifying pollution levels and assessing the health status of water bodies⁶. Building on this, we compared seasonal water quality trends between 1915 natural watersheds and nearby managed counterparts with similar climates to disentangle climatic and anthropogenic influences. To quantify the direction and strength of human intervention, we introduce the T-NM index, capturing its asymmetric amplification and attenuation effects across seasons. We compiled a 15-year comprehensive dataset for the ten largest watersheds in China, encompassing six major categories and 30 attributes, including seasonal elements, meteorology, watershed attributes, socioeconomics, land use, and landscape. Using a machine learning framework, we conducted a decoupling analysis to distinguish the natural and anthropogenic influences on seasonal water quality variations across natural and managed watersheds nationwide.

Our study addresses three key questions: (1) How have COD and DO evolved over the past 15 years in different watersheds in China? (2) What are the similarities and differences in the seasonal water quality trends between natural and managed watersheds in China? (3) What are the driving factors of the seasonal water quality variations in both types of watersheds?

CLICK HERE FOR MORE INFORMATION

https://www.nature.com/articles/s41545-025-00481-3