Assessing scenic beauty of hilly and mountain villages: An approach based on landscape indicators

•A method allows predicting scenic beauty of hilly and mountain villages.•Using depth values as part of the predictors.•Predicting the scenic beauty of villages with Support Vector Regression algorithms.•Nonlinear model outperformed linear regression in our comparison.•Depth values play an important...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Ecological indicators 2023-10, Vol.154, p.110538, Article 110538
Hauptverfasser: Long, Keliang, Wang, Nanxi, Lin, Zhongxiao
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A method allows predicting scenic beauty of hilly and mountain villages.•Using depth values as part of the predictors.•Predicting the scenic beauty of villages with Support Vector Regression algorithms.•Nonlinear model outperformed linear regression in our comparison.•Depth values play an important role in assessing the scenic beauty. With the growing demand for cultural ecosystem services around the world, many countries and regions have recognized the importance of the scenic beauty of villages and have introduced relevant policies. Many less modernized hilly and mountain villages will be redesigned. It is important to develop a model to predict the scenic beauty of these villages. However, only limited studies have been conducted to evaluate the scenic beauty of villages. This study aimed to explore the procedures for assessing the scenic beauty of mountain and hilly villages and to investigate the relationship between scenic beauty and landscape indicators. We first summarized the characteristics of these villages. Based on these characteristics, a series of indicators such as depth values and landscape metrics were used as independent variables, while photo scores were used as dependent variables, and different prediction models were developed with multiple linear regression, principal component regression, and support vector regression. Statistical metrics and the generalization ability of different models were compared. The results indicated that the inner areas of Xitou villages had the lowest scenic beauty scores, and people preferred natural scenery. Among landscape indicators, landscape diversity metrics and shape metrics have a significant impact on scenic beauty. Landscapes with a combination of multiple elements and not overly complex shapes may be more attractive. There was a significant negative correlation between mean depth and scenic beauty, implying that people prefer open space. It is essential to consider the depth value when evaluating scenic beauty in regions with strong 3-dimensional elements. The comparison between the three models revealed that the complex nonlinear algorithm performed and generalized better than linear models in villages where these areas are smaller and more sensitive to subtle changes. The method can be applied to other hilly and mountain villages in China, helping designers and decision-makers to output more reasonable planning and management schemes to fully utilize the local landscape and enhance tourism c
ISSN:1470-160X
DOI:10.1016/j.ecolind.2023.110538