Development and assessment of a landscape-scale ecological threat index for the Lower Colorado River Basin

Abstract

Anthropogenic disturbances impact freshwater biota but are rarely incorporated into conservation planning due to the difficulties in quantifying threats. There is currently no widely accepted method to quantify disturbances, and determining how to measure threats to upstream catchments using disturbance metrics can be time consuming and subjective. We compared four watershed-scale ecological threat indices for the Lower Colorado River Basin (LCRB) using landscape-level threats of land use (e.g., agricultural and urban lands), waterway development and diversions (e.g., number of canals, dams), and human development (e.g., road and railroads density, pollution sites). The LCRB is an ideal region to assess ecological threat indices because of the increasing need for conservation to ensure the persistence of native fishes in highly altered habitat. Each threat was measured for severity (i.e., level of influence on the upstream watershed) and frequency throughout each watershed; both severity and frequency were measured using two different methods. Severity values were based either on peer-reviewed literature and weighted in accordance to their published ecological impact, or assumed equal severity across stressors. Threat frequency was calculated according to either the presence/absence of each stressor, or on the relative density of each stressor in the watershed. Each measure of severity was combined with a measure of frequency, creating four ecological threat indices, and transformed to a 0&ndash;100 scale. Threat indices were highly correlated (slopes of 0.94&ndash;1.63; R 2 of 0.82&ndash;0.98), and were highest for watersheds close to urban centers, including Phoenix, Tucson, and Flagstaff, Arizona, and Las Vegas, Nevada. Road crossings and density appeared to be the most influential stressors in the index, but the removal of any individual stressor only changed the index by <5.1 units. Our results indicate that a simpler index with less subjectivity (i.e., presence/absence of a stressor in a watershed) provides similar results to the more subjective measure of threats (i.e., peer-reviewed threat severity). Because these threats have been linked to ecological health, the development of the index should be a useful tool to identify regions of greatest potential threat to aquatic biota and can aid in conservation planning for the Lower Colorado River Basin.