Low flow gave no clear change in warmth slopes of EPT family richness in two English regions in 1991-2017

We tested whether low flow made the link between warmth and recorded river invertebrate richness more negative. We used 8,542 site-date samples from northeast and northwest England in 1991-2003 and 2004-2017. The outcome was the number of recorded mayfly, stonefly and caddisfly families. Exposure measures covered the six months before each sample. Four models controlled site and season, year and month, analysis method, sample day and curved exposure terms. Interaction estimates were +0.012, +0.031, +0.022 and +0.052 families per degree C per ten percentage points of low-flow days. All four corrected intervals included zero. Autumn-only estimates changed sign between periods. These data and models do not establish a common added loss under combined warmth and low flow. Selected sites, missing flow matches, uncertain sampling effort, air temperature as a proxy for water temperature, and few year clusters limit inference. This is a weather association test. It tests no causal climate-change or air-pollution effect. The package includes a model panel, source hashes, code, checks and a figure.

macroinvertebrates · low flow · temperature · EPT richness · observation bias · regional replication · null result