Abstract
Ecosystem services (ES) assessment is crucial in ecology, with numerous studies seeking to evaluate them. Despite the abundance of methods and indicators, standardization is lacking, hindering comparability and progress in ES understanding and monitoring. This paper reviews indicators used in scientific literature to evaluate mountain ES and then examined whether the most used indicators are also the best. Using ISI Web of Knowledge we searched papers published between 2015 and 2020 containing “ecosystem services” AND “mountains” and we selected 400 papers effectively applying at least one indicator to evaluate ES. For each article, we extracted the following information: type of ES evaluated; indicator(s) used; input data used; type of value-domain (ecological, economic or social); scale of analysis; country; mapping; and management suggestions. From the literature, we extracted a list of 130 most frequently used indicators. The results indicated that regulating services were the most frequently assessed ES, followed by provisioning, cultural, and lastly supporting. The scale of analysis was mainly regional (51%) and local (40%), while studies at national scale were less frequent (9%) and only 2 studies were at global scale. Mountain areas most studied were in Europe (50%) and Asia (31%). Ecological value-domain was the most frequent used (55%), followed by social (26%) and economic (18%). 84% of studies considered only one dimension of the value and few studies combined multiple value-domains (15%). Almost half of studies mapped ES and around one third provided management suggestions. We examined the quality of indicators based on six criteria evaluated by experts in the field: significance, simplicity, cost, replicability, ease of interpretation and policy relevance. Indicators used exhibit significant diversity, and there is no clarity in nomenclature. There are cases where the indicators used represent inadequately the ecological parameters to be measured. Although many indicators score high in some of the properties evaluated, only 13 indicators perform well in all properties. These indicators have a universal value both in terms of semantics, but also in terms of properties and should be promoted for standardization among ES assessments. We envisage the necessity to condense the array of indicators commonly used down to a small set of standardized high-quality metrics.