Abstract

This study analyzes the role of large river flow events in annual loads, for three constituents and for up to 32 years of daily data at multiple watersheds with different land-uses. Prior studies were mainly based on simple descriptive statistics, such as the percentage of nutrient loadings transported during several of the largest river flows, while this study uses log-regression and analysis of covariance (ANCOVA) to describe and quantify the relationships between large flow events and nutrient loadings. Regression relationships were developed to predict total annual loads based on loads exported by the largest events in a year for nitrate plus nitrite nitrogen (NO3-N + NO2-N, indicated as total oxidized nitrogen; TON), total phosphorus (TP), and suspended solids (SS) for eight watersheds in the Lake Erie and Ohio River basins. The median prediction errors for annual TON, TP, and SS loads from the top five load events for spatially aggregated watersheds were 13.2%, 18.6%, and 13.4%, respectively, which improve further on refining the spatial scales. ANCOVA suggests that the relationships between annual loads and large load events are regionally consistent. The findings outline the dominant role of large hydroclimatic events, and can help to improve the design of pollutant monitoring and agricultural conservation programs.

Highlights

  • Increased nutrient and sediment loading from agricultural and urban activities is a primary factor in declining water quality in receiving streams, lakes, and estuaries [1,2]

  • The top five load events were chosen for predicting annual loads, as in most cases, they carried a large percentage of annual loads

  • Data from all Ohio watersheds, excluding Cuyahoga, indicated that, on average, 56% of total annual total oxidized nitrogen (TON) loads were exported during the top five load events, which lasted for approximately 17% of the days in a year

Read more

Summary

Introduction

Increased nutrient and sediment loading (flux) from agricultural and urban activities is a primary factor in declining water quality in receiving streams, lakes, and estuaries [1,2]. Nitrogen (N) and phosphorus (P) from intensively cropped and artificially drained watersheds and other nonpoint sources, has caused the severe impairment of surface water quality worldwide Increased enrichment of these pollutants leads to eutrophication followed by hypoxia, thereby severely disturbing ecological balances and reducing biodiversity. Over the past several decades, water quality sampling strategies have changed, moving from traditional fixed-frequency water sampling (bi-weekly, monthly, quarterly) to mixed strategies, such as fixed-frequency water sampling supplemented with some storm-based sampling, to capture high flows [9,10] These approaches are time- and resource-intensive, as it is essential to characterize all flow regimes and capture most of the high load periods throughout the year in order to avoid the underestimation of annual pollutant loading values. The Cuyahoga and Grand watersheds contain higher precipitation and runoff than the other watersheds because of the lake-effect [26]

Separation of Large Events
Predictability of Annual Loads
Similarity of Predictions
Applicability of Regression Lines in Other Regions
Large Event Characteristics
AAnnnnuuaall LLooaadd Predictions
ANCOVA
Error Analysis
Spatial Transferability
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call