LAUSR

Decision-makers are often described as increasingly interested in learning how investing in nature influences and steers the provision of ecosystem goods and services. Researchers, in response, have developed predominantly quantitative decision-support tools to assess ecosystem-services provision based on a wide range of different, often spatial input data for multiple demand and supply variables. Echoing this, Rieb and colleagues (2017) stated in their recent article that many decision-support tools, although providing important advantages of accessibility or generality, often fail to include sufficient complexity to comprehensively assess when, where, and how much nature is needed to provide ecosystem services (ES) and to sustain and improve human well-being. Rieb and colleagues come up with three research frontiers to improve the existing tools: (1) understanding the complex dynamics of ES in space and time, (2) linking ES provision to human well-being, and (3) determining the potential for technology to substitute for or enhance ES. We agree that these frontiers are important workspaces to make significant progress in ES assessments with decision-support tools. We miss, however, a deeper consideration of these workspaces for the important and globally spanning context of urban systems, where an increasing number of people live and along with them a growing number of practitioners and policy analysts making increasingly more frequent decisions about land and environmental resources at local and regional but also global levels (Kabisch et al. 2017). Current urban expansion will significantly affect natural resources worldwide with severe effects on ecosystems and the services they provide. But there are also enormous opportunities to improve the human–nature relationship in cities as, for example, (environmental) education can reach out to more and more people of different ages worldwide (Russ and Krasny 2017). Thus, the improvement of human well-being through the sustainable and resilient provision of ES in cities is of utmost importance and should be considered in the development of decisionsupport tools integratively, as has already been noted by Wachsmuth and colleagues (2016) and McPhearson and colleagues (2016a). The call for an integrated view to sustainable development particularly in cities and urban areas is emphasized in the 2030 Agenda for Sustainable Development, too, and rather opens a window of opportunity to widen the focus of the ES assessment debate to the urban space. In this vein, this viewpoint is complementary to the messages by Rieb and colleagues (2017) in a critical but constructive sense. In their first frontier, Rieb and colleagues (2017) argue that the complex dynamics of ES in space and time fall short in current decision-support tools. They plead for a closer collaboration between the ES scientific community and the remote sensing (RS) community to integrate advances in RS products into decision-support tools and, respectively, into decisionmaking. We agree that recent advances in RS—including the increasing access to the data archives of Earthobservation satellites with high-resolution and hyperspectral data—are relevant to biodiversity research and are particularly valuable for detecting spectral plant traits (Lausch et al. 2016), allowing for the assessment of ecosystem processes and the performance of ecosystem functions in space and time. Nevertheless, we have doubts that integrating these data into decision-support tools is easily applicable for decision-makers or policy analysts, particularly in the urban context. RS data are valuable for detecting changes in vegetation composition and structure and may also allow for assessing vegetation responses to global stressors related to heat or intense soil sealing. Still, such RS imagery data demand huge server and download capacities, specific expertise in processing and calibration, and expert knowledge and time to get the most and right information out of it. Given the current reality in many urbanand regional-planning departments facing limited financial, staff, and time resources, it is unlikely that such complex dependence on heavy RS data in decision-support tools will be generally feasible and regarded as a common working instrument. There remains little substitute for researchers working closely with decision-makers to improve the efficacy of ES-based solutions for urban challenges. In their second frontier, Rieb and colleagues (2017) argue that the interrelations between ES provisioning and human well-being need to be more reflected in decision-support tools