《Analyzing uncertainties and estimating priorities of landscape sensitivity based on expert opinions》
打印
- 作者
- 来源
- LANDSCAPE AND URBAN PLANNING,Vol.163,P.56-66
- 语言
- 英文
- 关键字
- Landscape sensitivity; Uncertainty; Expert modelling; Bayesian inference; FOREST; CONSEQUENCES; PREFERENCES; CRITERIA; MODELS
- 作者单位
- [Haara, Arto] Nat Resources Inst Finland, POB 68, FI-80101 Joensuu, Finland. [Store, Ron] Nat Resources Inst Finland, Silmajarventie 2, FI-69100 Kannus, Finland. [Leskinen, Pekka] European Forest Inst, Yliopistokatu 6, FI-80100 Joensuu, Finland. Haara, A (reprint author), Nat Resources Inst Finland, POB 68, FI-80101 Joensuu, Finland. E-Mail: arto.haara@luke.fi; ron.store@luke.fi; pekka.leskinen@efi.int
- 摘要
- The aim of this paper was to present an approach to estimate the priorities of landscape sensitivity criteria and to study uncertainties compounding that process. The data are based on expert judgment priorities of visual landscape sensitivity. The uncertainty model utilized was a Bayesian multi-criteria decision analysis (MCDA) model. The sources of uncertainty were classified into two categories: uncertainty caused by inconsistency of judge-specific assessments, and differences between judges in elicited priorities. In particular, we estimated the relative magnitude of different sources of uncertainty and considered the overall reliability of landscape sensitivity modelling. The results showed that the uncertainties in estimation of lands cape sensitivity criteria are large. Further analysis of the two uncertainty sources implied that the number of pairwise comparisons used to assess the landscape sensitivity criteria can be reduced to simplify the assessment task. This was the case since the assessment task involved multiple respondents. However, if there would be only one respondent, the inconsistency of the pairwise comparisons can be an important measure of uncertainty that would help decision makers to avoid over interpretation of the reliability of the priority estimates of the landscape sensitivity estimates. (C) 2017 Elsevier B.V. All rights reserved.