Search

Additional file 2: Fig. S3. Accessibility workflow: raster- and network-based methods. Fig. S4. Accessibility to health facility, raster- vs. network-based method comparison A. Under six hours walking time to access healthcare estimations between raster- and network-based methods and B. mapped infrastructures. Fig. S5 Accessibility to health facilities in urban context, raster- vs. network-based method comparison A. population estimation B. extent of under 6 walking hours access to health facilities. Table S1. Comparison of mapped health facilities in OSM with field observation: facilities inventory, localisation comparison and damages inventory.

Additional file 3: Performance and computational environment.

Additional file 1: Data availability and completeness. Fig. S1. Indicator comparing population building densities from Ohsome Quality Analyst to evaluate the completeness of OSM data. Fig. S2. Comparison of mapped health facilities in OSM with field observation: facilities inventory, localisation comparison and damages inventory.

The estimation of the value of georeferenced spaces is challenging because the value of a space depends on many factors such as recreation potential, sociability, cultural points of interest, etc. To account for these personal values, interviews and surveys can be conducted but this is costly and elaborate, and this requires a user action. Recent approaches use social media to attribute positive or negative associations of users to a georeferenced space by analysing the semantic linkages. This approach is promising but at the same time prone to biases due to a user selection bias and the interpretation of the semantics. In this paper, an alternative method based on the systematic analysis of georeferenced paths of people is proposed, which does not require direct user interaction. Underlying the assumption that people are willing to take a longer and more time-consuming path if the detour has a personal value for them, this work proposes to estimate these personal values systematically and partly automated by analysing the pathways taken. The authors suggest to conduct a first feasibility study analysing the GPS positions of people to obtain information and find patterns of the personal preferences of places, and matching them to interview or social media derived information on platial preferences.

To reduce the negative environmental impact of settlement development, a range of policy instruments are available. Empirical studies have highlighted advantages and disadvantages of different instruments, including tradable development rights (TDRs) and zoning. We add to this knowledge by developing a land-use modelling framework that can be used to assess the potential impact of different policy scenarios on the spatial pattern of settlement development and the potential loss of ecosystem services. We compare the impact of four scenarios that are inspired by different policy instruments. Our results show that limiting the urban growth in certain municipalities or regions has a high impact on ecosystem services. Thus, the definition of sending and receiving areas as employed in TDR programs could be very useful. However, possible trade-offs will have to be taken into consideration, in particular between the protection of fertile agricultural soils and the protection of valuable nature areas.