Autor(es):
Castro, José 
; Carvalho-Ribeiro, Sónia ; Pinto-Correia, Teresa
Data: 2013
Identificador Persistente: http://hdl.handle.net/10174/10235
Origem: Repositório Científico da Universidade de Évora
Assunto(s): Landscape Preferences
Descrição
This paper presents an expedite method to upscale land cover pattern preferences as
expressed by tourists based on landscape metrics as indicators. Ten preferred land cover
patterns previously considered in the Landscape Preferences Spatial Framework (LPSF)
(Ribeiro et al, 2013) were searched and evaluated over the study area (Alentejo, Portugal)
using PLAND (% of landscape of each land cover) maps built with FRAGSTATS moving
window analysis. For each one of the preferred land cover patterns resulted one regional map
that was further analysed through multiple regression predictions built upon class and
landscape level FRAGSTATS metrics such as PLAND, LPI, TE, AREA_AM, PAFRAC,
SHAPE_AM, ENN_AM, CONTAG, IJI, AI, SHDI, SIDI, SHEI e SIEI. The results show that
landscape preferences could be an useful tool at strategic landscape planning level, as they
express stakeholders interests, as well as at operational landscape planning level helping
outdoor recreational zoning. These results were examined by local landscape experts that
Where to walk in Alentejo? Mapping outdoor recreation based on land cover pattern preferences by tourists 24/01/14, 12:07
http://www.iale2013.eu/print/where-walk-alentejo-mapping-outdoor-recreation-based-land-cover-pattern-preferences-tourists Página 2 de 2
assessed and in some cases validated the meaningfulness of the analysis here conducted
selecting various areas for each landscape preferences. Finally results were refined by a
logistic regression of the recognized landscapes areas against the most significant landscape
metrics for each case.
Throughout the paper selected metrics and regression coefficients of each preferred
landscape are discussed. Overall, it can be said this is a valid methodology to map regional
scale landscapes preferences based on local (municipality) preference surveys but further
work needs to be done to refine the approach namely by incorporating other landscape
metrics beyond PLAND.
