Modeling linkages between erosion and connectivity in an urbanizing landscape

Alexander Michalek; Amirreza Zarnaghsh; Admin Husic

doi:10.1016/j.scitotenv.2020.144255

Modeling linkages between erosion and connectivity in an urbanizing landscape

Sci Total Environ. 2021 Apr 10:764:144255. doi: 10.1016/j.scitotenv.2020.144255. Epub 2020 Dec 24.

Authors

Alexander Michalek¹, Amirreza Zarnaghsh¹, Admin Husic²

Affiliations

¹ Dept. of Civil, Environmental and Architectural Engineering, University of Kansas, United States of America.
² Dept. of Civil, Environmental and Architectural Engineering, University of Kansas, United States of America. Electronic address: ahusic@ku.edu.

PMID: 33385647
DOI: 10.1016/j.scitotenv.2020.144255

Abstract

Erosion and connectivity are spatially varied processes key to determining sediment transport and delivery to downstream waterbodies. However, we find few studies that explicitly model the linkages of where erosion and connectivity coincide and where they contradict, particularly in urbanizing settings. In this study, we couple in-stream aquatic sensing, the Revised Universal Soil Loss Equation (RUSLE), the Index of Connectivity (IC), and the Sediment Delivery Ratio (SDR), together with Monte Carlo uncertainty analysis, to generate a new Erosion-Connectivity Mapping (ECM) framework. We evaluate ECM accuracy with field assessment of thirty-five sites spread across five lowland watersheds (mean slope <5°) in Johnson County, Kansas, USA, which differ primarily in their land use, ranging from 21% to 89% urban. RUSLE modeling results indicate erosion is controlled by topography with high risk areas near streambanks roadway systems. SDR and IC were positively related at the five sites (R2 = 0.78, p < 0.05) with the highest values in the most urbanized watershed, indicating that anthropogenic change augments connectivity. The ECM results indicate that while only 5±1% of the study area is both highly erodible and highly connected, these areas represent 37±4% of total watershed-scale erosion. Our modeling results indicate that erosion is more likely to be the limiting factor in sediment transport, as opposed to connectivity, as there are generally more locations that are well-connected to hydrologic transport but resistant to erosion. Our field assessment provided broad support for the ECMs; however, field assessment indicated that geospatial modeling underpredicts how closely related erosion and connectivity are in the field and we suggest that future models consider this coupling more explicitly. This study provides a method for combining RUSLE and IC in a new tool (ECM) to identify spatial patterns in sediment erosion-connectivity to aid in the understanding and management of watershed sedimentation.

Keywords: Disconnectivity; Index of Connectivity; RUSLE; Sediment Delivery Ratio.