Data for: Scaling up nature-based solutions for climate-change adaptation: potential and benefits in three European cities
The dataset contains GIS data and JPEG maps of nature-based solution scenarios and related benefits in three case-study cities partners of the H2020 project Naturvation (https://naturvation.eu/): Barcelona (Spain), Malmö (Sweden), and Utrecht (the Netherlands). The data were produced as part of the research described in the article “Scaling up nature-based solutions for climate-change adaptation: potential and benefits in three European cities”, published in Urban Forestry & Urban Greening (doi:10.1016/j.ufug.2021.127450). The dataset is structured into three main folders, one for each city. Each folder contains six raster maps of land cover under different scenarios, a vector map with the results of the assessment of the selected benefits at the local level, and a sub-folder with the benefit maps printed in JPEG format. The six scenarios include the current condition (Baseline - LC); four scenarios that simulates the full-scale implementation of one specific type of nature-based solutions: installing green roofs (GreenRoofs - GR), de-sealing parking areas (ParkingAreas - PA), enhancing vegetation in urban parks (Parks - PK), and planting street trees (StreetTrees - ST); and a scenario considering the contemporaneous implementation of all four types of nature-based solutions (GreenDream - GD). The simulated full-scale implementation is based on space availability and technical feasibility: other constraints to the implementation of nature-based solutions are not considered. The five benefits assessed include two benefits related to climate change adaptation, i.e. heat mitigation (HM) and runoff reduction (RR), and three co-benefits, namely carbon storage (CS), biodiversity potential (BP), and overall greenness (OG). The vector maps and related JPEG prints show the results of the assessment at the block level. Blocks are based on a modified version of Urban Atlas polygons obtained by removing streets and railroads. Maps have coordinate reference system UTRS89 - LAEA Europe (EPSG:3035) and cover the whole administrative territory of the respective city, excluding the sea. Raster maps are provided in Geotiff format, UInt 16, with a resolution of 1 m. The legend includes eight land cover classes: water (0), trees (1), low vegetation (2), impervious (4), agriculture (5), buildings (10), green roofs (11), vegetation over water (13), permeable parking areas (14). The attribute tables of the vector maps store the value of the selected benefits for each block, together with the links to the original Urban Atlas polygons. Scenarios and benefits are identified by their two-letter codes as reported above. The printed JPEG maps of benefits have a common legend, to allow for comparison between cities.
Steps to reproduce
The current land cover was classified through a segmentation algorithm (using the mean shift segmentation framework in the Orfeo-toolbox) and a tree-low vegetation classifier (trained following the method by Abdi, 2020). The procedure combined five datasets: Normalized Difference Vegetation Index (NDVI), Normalized Digital Surface Model (NDSM), Urban Atlas data on water and agriculture, and vector maps of building footprints and coastline. Land cover scenarios corresponding to the implementation of nature-based solutions (NBS) were developed by defining a set of rules to identify suitable areas for NBS implementation, and translating them into GIS algorithms that modify the maps of the current land cover. Spatial data to model land cover transitions in the scenarios were retrieved from the Urban Atlas and from Open Street Map. Heat mitigation, i.e. the potential of NBS to lower high (summer) temperatures, was assessed using the heat mitigation index as calculated through the InVest - Urban cooling model v 3.8.7 (Sharp et al., 2020), a spatially-explicit proxy-based model. Runoff reduction was assessed through the runoff retention index calculated by the InVest - Urban flood risk mitigation model v.3.8.7 (Sharp et al., 2020). The model applies the Curve Number method developed by the USDA and the index expresses the percentage of stormwater that is retained in the analysed area. Carbon storage was modeled as a function of land cover, assuming a steady state (no sequestration or decomposition), using values per unit area of each land cover class retrieved from the literature. Biodiversity potential was calculated for each block using the method by Radford and James (2013), which expresses the potential as a function of land cover (structural) diversity and share of green area. Overall greenness was calculated as the share of green (i.e., water, trees, low vegetation, vegetation over water, and agriculture) in the 500-m buffer around each point of observation. The index was computed for points randomly placed at a minimum distance of 10 m from each other. The final value is the average of the points within the analyzed area. For more information on data and methods, please refer to section 2 and to the Supplementary Material of: Cortinovis C., Olsson P., Boke-Olén N., Hedlund K. (2022). Scaling up nature-based solutions for climate-change adaptation: potential and benefits in three European cities. Urban Forestry & Urban Greening. doi:10.1016/j.ufug.2021.127450