A fish-based multimetric index for Brazilian savanna streams

摘要: Abstract Multimetric indices (MMI) have been widely used to assess ecosystem conditions because they are low-cost, employ a rapid field method, and can incorporate various biological metrics at different levels of organization. Our objective was create fish-based multimetric index applicable all streams the Brazilian savanna biome (Cerrado), second largest in Brazil deemed global biodiversity hotspot. We evaluated 156 sites two river basins (Parana Sao Francisco) selected capable distinguishing stream-sites across gradient anthropogenic disturbances. employed MMI approaches determine if an based on natural variation-adjusted performed better than one unadjusted metrics. In addition, we assessed performance final MMIs their sensitivity pressures local (LDI), catchment (CDI) both scales integrated (IDI). Finally, power probability sample survey design infer headwater stream hydrologic region approximately 47,000 km2. for included six metrics: % common species; characiform individuals; loricariid trichomycterid invertivore species, Poecilia reticulata individuals. MMI1 (unadjusted metrics) MMI2 (natural discriminating least- most-disturbed sites, but distinguished intermediate from MMI1. Both were negatively correlated with CDI scores; however, only IDI scores. inferred that 709 km (9.35%) studied good condition, 8115 km (82.73%) intermediate, 641 km (7.91%) poor condition. conclude proposed this study great potential widespread application integrate data most important represents more 20% country. Furthermore, retained indexes easy access low-cost method. However, feasibility areas influenced by mining, as well biomes, should be tested.