Mapping Cultural Ecosystem Services using Earth Observationsg.email@example.com
Managed and semi-natural ecosystems provide important benefits to people, in particular crop and livestock food production, flood protection, and carbon sequestration. Methods to measure, map and monitor these services, including approaches that rely on the wealth of freely available Earth Observation (EO) data, are developing steadily. Furthermore, evidence on the linkages between biodiversity (in the broad sense including genetic, taxonomic, and ecosystem diversity) and these ecosystem services is emerging. Unlike these services, however, the capacity of ecosystems to provide cultural and recreational services, and the dependence of that on biodiversity, is a relatively unexplored research field and does not makes use of the complexity of EO based mapping [Barbosa et al. 2015]. the challenge of quantifying cultural ecosystem services (CES) lies in their apparent disconnect from specific ecosystem functions, their inherent complex interconnection to people’s cognitive perception of aesthetics and outdoors enjoyment, and the difficulty of localizing them.
For many ecosystem services, there is strong causal relationships between specific ecosystem functions, such as phenology and net primary production (for carbon sequestration) or maintaining soil porosity and enhancing infiltration (for flood control, say by trees). In such cases, it is the ecosystem functions that can be directly, or more often indirectly, monitored using EO. This is not the case for CES, which to date is treated as a complex and subjective value, studied primarily using social-science tools (e.g. surveys [van Berkel et al. 2014]) and related mostly to landscape features and land cover forms, see e.g. [Plieninger et al. 2013].
Cultural Ecosystem Services are also notoriously difficult to quantify independently of people. Whereas for most ecosystem services, it is possible to conceive and quantify a ‘potential’ supply of the service (often the end-point in most ecosystem services studies!), which is the maximal amount of utilization of that service from a particular locality, this is hard to justify for CES. While it is possible to quantify aesthetic qualities of photos, such studies tend to be very local and limited (but see [Seresinhe et al 2015]). Furthermore, factors such as accessibility and environmental conditions (e.g. temperature, humidity) that are controlled (to an extent) by ecosystems and contribute to human comfort are rarely considered.
Finally, people may not experience CES at specific locations, but rather gain their enjoyment from the larger landscape around them or by moving through a landscape (e.g. walking). It may be impractical to ask people to pinpoint exactly where they ‘use’ CES, as this is not actually well defined nor the spatial scale from which they benefit changes across the studied landscape. How people value CES may also depend on difference between daily interactions (commuting, daily use of space around us) and during particular recreational activities (when we seek out the outdoors).
The above-mentioned difficulties in developing more fundamental understanding and better mapping of CES are challenging. To be able to tackle such challenges, it would require considerable volume of social-science data on CES (e.g. questionnaires, surveys etc.) and a spatial dataset that extends across multiple spatial and temporal scales, and is capable of quantifying different facets of CES supply and demand.
The aim of this PhD project is to study the spatial and temporal scale dependence of CES on biodiversity and the landscape, and to do this using novel Earth Observation-based approaches. It will attempt to address questions such as:
At what spatial scale do people aesthetically value landscapes, and how is that linked to biodiversity – is it very local (e.g. a rare flower), at their immediate area (unique plant communities), at the landscape scale (a diverse forest), or at the viewshed scale (landscape heterogeneity) ?
How does the temporal dimension affect CES value – for example, do people prefer landscapes wherein they pass through larger variety of ecosystems (forest next to a field, next to a river etc.) while hiking? Do they prefer locations already familiar to them (less time spent reaching the area)? Do they value places where seasonality is more evident – e.g. are deciduous trees more attractive than evergreen trees?
Knowing the spatial and temporal scaling relationship of CES, can we improve the existing approaches for mapping and monitoring of CES at large scales - combining one-off Earth Observation with 1-3 meter resolution (LiDAR and hyperspectral), frequent (every few days) multispectral imagery at 10-30 meter resolution (e.g. Landsat, Sentinel-2) and daily 250 to 500 meter resolution multispectral data (e.g. MODIS) ?
The project will take advantage of a large volume of questionnaires, interviews and participatory GIS data on CES, as well as Earth Observation data (LiDAR and hyperspectral at 1-3m resolution) - both collected as part of the Wessex-BESS project (Biodiversity and Ecosystem Services in Multifunctional Landscapes - a six-year (2011-2017) programme funded by NERC) over an area of nearly 1,400 km2 (Figure 1).
Potential for high impact outcome
There is a growing academic, industrial and political movement towards accounting for nature in every aspect of life. The EU has adopted a Biodiversity Strategy 2020 which calls upon all Member States to include Ecosystem Services (including CES) as part of their National Accounts. Projects such as the Natural Capital Coalition and big companies including Dow, Nestle and Unilever are developing innovative approaches for Natural Capital Accounting – most recent example being the Natural Capital Protocol. On the other hand, enormous budgets and effort is spent on space programmes, in particular the Copernicus programme of the European Union, and demand for better use of these large volumes of Earth Observation data in mapping Natural Capital is high. This PhD project would contribute to this developing area of academic and non-academic interest, which so-far made little progress on monitoring of CES from space.
Figure 1: The Wessex Chalk landscape in southern England comprises 1400 km2 of open, rolling chalk land with small hilltop woodlands and rivers within narrow floodplains. This landscape contains a mix of productive arable land, extensively managed grassland undergoing biodiversity restoration and high nature conservation value, ancient grasslands. It is an exceptional prehistoric ritual landscape with widespread earthworks and monuments
The student taking this PhD project will gain deep understanding and knowledge in the area of natural resources management, in particular accounting for Ecosystem Services and Natural Capital. He/she will develop transferable skills in statistical analysis, compiling and producing data by social-science methods (questionnaires, interviews, participatory GIS), and analysing and handling Earth Observation data (including software such as ENVI or Google Earth Engine for analysis). The University of Leeds offer a wide selection of short courses for PhD students, and he/she will also be able to take NERC-sponsored short courses delivered by universities across the UK.
The student should have a strong interest and high grades in spatial analysis including GIS and using statistical methods. Previous experience with remote sensing desirable but not mandatory. An aptitude for programming will be an advantage.
D.A. Barbosa, C.C., Atkinson, P.M. and Dearing, J.A., (2015) Remote sensing of ecosystem services: a systematic review, Remote sensing of ecosystem services: a systematic review. Ecological Indicators, 52, 430-443. (doi:10.1016/j.ecolind.2015.01.007).
van Berkel, D.B. & Verburg, P.H. (2014) Spatial quantification and valuation of cultural ecosystem services in an agricultural landscape. Ecological Indicators, 37, 163-174.
Plieninger, T., Dijks, S., Oteros-Rozas, E. & Bieling, C. (2013) Assessing, mapping, and quantifying cultural ecosystem services at community level. Land Use Policy, 33, 118-129
C.I. Seresinhe, T. Preis & H.S. Moat, (2015) Quantifying the Impact of Scenic Environments on Health, Scientific Reports 5, 16899
Related undergraduate subjects:
- Natural resource management
- Remote sensing
- Spatial ecology