On the spatial aggregation of condition metrics for ecosystem accounting

Anders Lorentzen Kolstad , Hanno Sandvik, Bálint Czúcz, Chloé R. Nater 

In face of the ongoing nature crisis, the international community is setting targets and deciding on actions
to combat the current biodiversity loss. For this to be effective they need tools to accurately describe
the current situation and to monitor trends in ecosystems over time. Ecosystem condition accounting
(ECA) is one such tool that use variables and indicators to describe key ecosystem characteristics, reflecting
their condition and deviations from a reference condition. Because the purpose is to inform decisions at
relatively high political levels, these ECA metrics are often spatially aggregated to represent larger areas,
such as countries. However, spatial aggregation of information has the potential to alter the descriptive
and normative interpretations one can make from these metrics. For example, aggregation displacement
causes the information held in variables and indicators to diverge when these are aggregated spatially. This
process is influenced also by the order of steps involved in normalising and aggregating variables, i.e. the
aggregation pathway. Although aggregation displacement and the type of aggregation pathway chosen for
the indicator clearly impact both the indicator values and their interpretation, there are no clear guidelines
or deliberation on these topics in the SEEA EA standard for ecosystem accounting. This paper outlines
the consequences of different aggregation pathways, emphasising their impact on the credibility of ECAs,
and how these are interpreted by users. We introduce a standardised terminology for aggregation pathways
specific to ecosystem condition indicators following the SEEA EA standard and provide recommendations
for selecting appropriate pathways in various contexts. Our discussion of this topic is aimed at raising
general awareness of spatial aggregation issues and to guide indicator developers in choosing and reporting
spatial aggregation methods.