Increasing the cost-effectiveness of neotropical stream assessments based on physical habitat, fish, and macroinvertebrates

Site-sampling effort, including site extent and levels of sample processing, plays a key role in ecological research. Insufficient effort increases the likelihood of both Type I and Type II statistical errors, whereas excessive effort can lead to unnecessary costs. This indicates the existence of an optimal effort level that minimizes both the risk of insufficient data and the burden of excessive expenditure. Starting from the standard sampling approach defined by the U.S. Environmental Protection Agency (USEPA), we sampled physical habitat structure and fish and benthic macroinvertebrate assemblages in 160 wadeable stream reaches, each with 10 sections divided by 11 transects, across four watersheds in the Brazilian Cerrado biome. Here, we assessed the effect of reduced sampling effort within transects (from 11 down to 6) on distinguishing among least-disturbed, moderately disturbed, and most-disturbed sites. For physical habitat, metrics such as canopy cover, human disturbance and presence of fine sediments did not change with sampling effort, neither did their variances. In contrast, reducing sampling effort resulted in an underestimation of taxa richness for both macroinvertebrates and fish down to around 70% of the full sample, although rarefying sampling effort per transect did not alter the ability to distinguish whether each site belonged to least-disturbed, moderately disturbed, or most-disturbed categories. Biotic indices, namely Multi Metric Index (MMI) for both and Average Score Per Taxon (ASPT) for benthos remained seldom changed regardless the number of sampling transects. As such, reducing sampling effort had limited effects on physical habitat variables and biotic indices, but a clear effect on taxa richness, which proved more sensitive to effort reduction. At the temporal level, the average field time per site was reduced from around 6 h to 3 h, and the average benthos processing time per site decreased from 15 to 10 days. Taken together, these results point to the potential of reducing sampling effort to enable faster, more cost-effective, and more feasible research in Cerrado streams, with valuable implications for underfunded stream biomonitoring efforts elsewhere. However, it is important to note that these results are applicable only to the specific outcome of rapid assessments of habitat quality, and may not be suitable for detailed assessments of biotic communities, such as full species inventories, particularly those involving specialist or rare species. • Reduced sampling effort retained key bioassessment capacity for macroinvertebrates. • Fish metrics were the most sensitive to reduced sampling effort for some indices. • % fines and anthropogenic disturbance distinguished site classes reliably. • Cost-effective sampling enables faster feedback for biomonitoring in the Cerrado. • Findings support practical stream assessments in data-poor tropical regions.