WDM 7_23 Lecture | Understanding Preferences for Rural Landscapes
12 October 2023 - October 12 @ 17:30
12th October 2023
Understanding Preferences for Rural Landscapes
by Cathal O’DONOGHUE, University of Galway, Paul KILGARRIFF, Luxembourg Institute of Socio- Economic Research, and Mary RYAN, Teagasc
General subject area | Landscape preference modelling
Abstract | The purpose of this study is to understand public preferences for different landscapes. Our aim was to categorise landscapes in terms of identifiable attributes and to assess the public’s preference for different types of landscape.
The first part of this study undertakes a choice experiment, utilising a survey of landscape preferences. We use a statistical methodology to extract individual preferences for different rural landscapes as defined by images. Using Ireland as a case study, we conducted a survey consisting of 430 individuals, asked for their preferences through the use of 50 rural landscape photographs.
Initial results show that individuals prefer managed agricultural landscapes that contain stonewalls and hedgerows and a negative preference for rough unmanaged landscapes, industrial peat harvesting and wind turbines. Heterogeneity is also found between urban and rural dwellers. With increasing levels of development and urban expansion, it is important to preserve not only protected habitats but also landscapes that are highly valued by local residents and society as a whole. In order to protect the latter, we need to establish what society values and where these areas are located.
In this study we decompose landscape photos into individual attributes and quantify the preferences for each attribute. Using this set of environmental attributes, we use open-source spatial data to present each attribute spatially for the wider landscape. For example, a high Normalized Difference Vegetation Index (NDVI) score is used for green pasture, stonewalls in areas that have limestone bedrock and agriculture present. By only using European-wide data, the methodology outlined is scalable when combined with a regional choice experiment to reflect local preferences. Ireland is divided into a grid of 400m cells. Each cell is assigned landscape attributes on the basis of the characteristics that can be measured within the grid.
Utilising our the preference model estimated, we can generate a public preference (both in terms of the national population and in terms of the local population) for the visible landscape characteristics. Of interest is the different between the public’s view and expert views. We compare in two dimensions, preferences for landscapes compared with expert based Landscape Character Assessment (LCA) and with the biodiversity potential of the landscape.
The results show the areas that are highly valued by individuals do not correspond to the areas that are highly valued in expert-led studies such as the (LCA). In terms of biodiversity, we find that the public have similar preferences for biodiverse landscapes with experts, except in the case of peatland. This highlights a potential issue for future planning and environmental policy and a need for improved education in relation to the biodiversity value of peatland. If provision is only made to protect the landscapes valued by environmental experts, landscapes that are highly valued by society are left unprotected and vulnerable to being damaged or lost.