A new diatom-based multimetric index (MMI-D) for ecological health monitoring in the Tropical Rift Valley Lake (Lake Hawassa, Ethiopia)

Gelaw, TW and Mengistou, S and Barker, PA (2024) A new diatom-based multimetric index (MMI-D) for ecological health monitoring in the Tropical Rift Valley Lake (Lake Hawassa, Ethiopia). African Journal of Aquatic Science, 49 (4). pp. 330-343. ISSN 1608-5914

Text (final_submitted_after_peer-reviewed_Manuscript_TW_Gelaw)
final_submitted_after_peer-reviewed_Manuscript_TW_Gelaw.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (383kB)

Abstract

Multimetric assessment is one of the important tools for diagnosing, detecting and measuring impairment in ecosystem function in lentic ecosystems. It enhances detection capabilities across a broader variety of stressors and offers a more complete picture of ecological conditions than single metrics or biological indicators. In this context, a diatom-based multimetric index (MMI-D) has been developed to evaluate the ecological health of Lake Hawassa. Physicochemical and benthic diatom sampling was conducted at nine sites along the lakeshore, representing varying levels of human disturbance, from February to November in2015 and 2016. The sampling sites were classified a priori into three categories: minimally disturbed (three sites), moderately disturbed (three sites) and highly disturbed (three sites). This classification, was based on a clustering analysis using the percent disturbance score (PDS). Of the 24-diatom candidate metrics, only ten were chosen as core metrics for the development of MMI-D, based on redundancy analysis, reaction to environmental conditions, percent discriminatory efficiency (%DE) and box plots. The newly established MMI-D clearly distinguished between reference and non-reference sites, and between the lake’s three clusters. The MMI-D’s performance was validated using independent data sets from lakes Hawassa and Ziway and it demonstrated the best capability for discrimination between different disturbance levels. MMI-D 2-stage Least Squares (2SLS) regression analysis revealed an inverse but robust connection with the PDS, indicating its responsiveness to Lake Hawassa habitat quality degradation (n = 9, R2 = 0.921, p < 0.001). The MMI-D revealed a high %DE (95.1%) and a negative but significant connection with nutrients, total suspended solids (TSS), and turbidity (R2 > 0.6; p < 0.05). Generally, it can be concluded that this index is a powerful tool that could assist endusers by providing a practical method for measuring the ecological quality of Lake Hawassa.