Setting reference levels and limits for good ecological condition in terrestrial ecosystems – Insights from a case study based on the IBECA approach

Abstract Effective evidence-based nature conservation and habitat management relies on developing and refining our methodological toolbox for detecting critical ecological changes at an early stage. This requires not only optimizing the use and integration of evidence from available data, but also optimizing methods for dealing with imperfect knowledge and data deficiencies. For policy and management relevance, ecological data are often synthesized into indicators, which are assessed against reference levels and limit values. Here we explore challenges and opportunities in defining ecological condition in relation to a reference condition reflecting intact ecosystems, as well as setting limit values for good ecological condition, linked to critical ecological thresholds in dose–response relationships between pressures and condition variables. These two concepts have been widely studied and implemented in aquatic sciences, but rarely in terrestrial systems. In this paper, we address practical considerations, theoretical challenges and possible solutions using different approaches to determine reference and limit values for good ecological condition in terrestrial ecosystems, based on empirical experiences from a case study in central Norway. We present five approaches for setting indicator reference values for intact ecosystems: absolute biophysical boundaries, reference areas, reference communities, ecosystem dynamics based models, and habitat availability based models. We further present four approaches for identifying indicator limit values for good ecological condition: empirically estimated values, statistical distributions, assumed linear relationships, and expert judgement-based limits. This exercise highlights the versatile and robust nature of ecological condition assessments based on reference and limit values for different management purposes, for situations where knowledge of the underlying relationships is lacking, and for situations limited by data availability.