Robust estimates of biodiversity change require high-resolution time series

Abstract

Abstract Robust estimates of biodiversity change are essential to inform management and conservation, and to address global biodiversity loss. However, common temporal data limitations may compromise trend accuracy. Here, we test how temporal resolution influences biodiversity trends using 1,353 river invertebrate time series collected annually for ≥ 10 years across 18 European countries. We simulate reduced sampling frequencies and durations and compare these trends to the complete time series. Reducing frequency from annual to every 2–6 years resulted in 87–73% of sites matching in trend directions, but only 78–39% matching in magnitude. Reducing duration from 10 to 9–2 years resulted in 88–52% direction matches and 86–8% magnitude matches. Similar results were observed for longer time series ( ≥ 20 years). Additionally, a comparison of two real-world monitoring datasets with different temporal resolutions shows that 53% of sites matched in direction, but only 12% in magnitude. Our findings indicate that biodiversity change magnitude is more sensitive to temporal resolution than direction. Consequently, accurate estimates of both direction and magnitude require high-resolution time series, whereas lower-resolution data may only reliably capture direction. These results highlight the value and limitations of temporal biodiversity data, and help plan future monitoring.