This post was adapted from a post published on the blog of the Tropical Ecology and Entomology Lab led by Associate Professor Eleanor Slade.

Insects are responding rapidly to global environmental change, yet understanding why insect populations change remains one of the greatest challenges in ecology and conservation. While monitoring efforts increasingly reveal shifts in insect abundance, diversity, biomass, and distributions, observed patterns are often shaped by multiple interacting environmental drivers, biological mechanisms, and even the way insects are detected.
 
In our new paper published in Current Opinion in Insect Science, we propose an integrative framework that links environmental drivers, mechanistic pathways, detection processes, causal inference, and ecological forecasting within a single conceptual structure.

Environmental pressures such as climate change, habitat modification, pollution, pesticides, and urbanisation influence insect populations through interacting physiological, behavioural, ecological, trait-mediated, and evolutionary processes. However, the signals we observe are also filtered through detection, because monitoring methods capture activity-dependent and context-dependent observations rather than direct measures of true population change. As a result, apparent increases, declines, or stability in insect populations may partly reflect variation in detectability, sampling design, or environmental conditions rather than underlying ecological responses alone.
 
Predictive entomology therefore emphasises explicit causal inference to separate true driver effects from confounding processes and detection artefacts. Achieving this requires understanding not only the environmental drivers of insect populations but also the biological and observational processes that generate the patterns we observe. By recognising that ecological observations reflect both true population change and the processes that influence detectability, predictive entomology provides a stronger foundation for attributing observed population change to its underlying causes and improving forecasts of future responses under environmental change.

The predictive entomology framework linking (A) environmental drivers, (B) mechanistic pathways, (C) detection, (D) causal inference, and (E) forecasting.

Realising predictive entomology will require coordinated advances in long-term monitoring, mechanistic and experimental research, causal inference, and process-explicit forecasting models. Together, these advances can help move insect ecology toward a more predictive science capable of identifying emerging risks, improving biodiversity forecasting, and guiding proactive conservation under rapid global environmental change.

This research is part of the Climate Transformation Programme, and is supported by the Ministry of Education, Singapore, under its TIER 3 Programme (Award MOET32022-0006).