普陀山岛旅游生态安全发展趋势预测

科学地预测海岛目的地旅游生态安全发展趋势,对促进海岛旅游经济和生态环境协调发展具有重要的理论意义和实践价值。基于可持续发展的视角,建立了由承载力-支持力-吸引力-延续力和发展力(CSAED模型)子系统构成的普陀山旅游生态安全指标体系,并在灰色系统GM(1,1)模型和RBF神经网络模型比较选优的基础上,对普陀山岛旅游生态安全发展趋势进行了预测。研究结果表明:(1)和灰色系统GM(1,1)模型相比,RBF神经网络模型的Pearson相关系数和误差均方根值更优,可更精确地拟合普陀山岛旅游生态安全发展趋势;(2)2015—2020年,普陀山岛旅游生态安全指数的RBF神经网络模型预测结果由0.7017增加至0.8135,安全等级由比较安全上升至非常安全。研究结果可为维护普陀山岛旅游生态安全提供决策依据。

Developmental trend forecasting of tourism ecological security trends: the case of Mount Putuo Island

The study of tourism ecological security is a core problem in the research of sustainable tourism development. It has important theoretical significance and practical value for coordinated development of island tourism economy and the ecological environment. It can be used to scientifically predict island destination development trends of tourism ecological security. Tourism destination can be regarded as an organism with a complex ecosystem. To our knowledge, once the function is disordered, the destination must be considered threatened. The purpose of this paper was to construct a tourism ecological safety index system based on the five subsystems, including "Carrying Capability", "Supporting Capability", "Attraction Capability", "Evolution Capability", and "Developing Capability" (Known as the CSAED model). Based on the sustainable development perspective, and the comprehensive, dynamic principle; this paper uses Mount Putuo Island in Zhejiang Province as an example, and constructs a tourism ecological safety index system based on the subsystems of carrying capacity, attraction capability, evolution capability, and development capability (CSAED model). The paper used the Grey system GM (1, 1) model and the radial basis function (RBF) neural network model to forecast tourism ecological safety in Mount Putuo Island. The results showed that:(1) both the Pearson correlation coefficient for the RBF neural network and the root mean squared error were better than the Grey System GM (1, 1) model. They also exhibited a better linear fit and a higher precision of prediction. This paper used Grey Relational Analysis to select the main driving factors; and used the results of linear and nonlinear analysis to build equations for trend extrapolation. In addition, based on the results of principal component analysis, the RBF neural network model appeared to provide a new research area for tourism destination ecological security. One of the key issues was that the tourism ecological security situation of Mount Putuo Island became better, because the index of the RBF model from 2005 to 2014 predicted results from 0.3568 to 0.6475. It appeared that the security level increased the sensitivity level, critical level, and the general level. Additionally, the index of the RBF model from 2015 to 2020 predicted results were from 0.7010 to 0.8135, the security level increased from the relatively safe grade to very safe grade. However, it is well known that the ecological system on the island will be influenced by several factors, including natural, social, and economic, among others. In terms of Mount Putuo Island, during the period of the forecast, it may be affected by typhoons, which may affect the vulnerability of the tourism industry. As such, it may lead to deviation from forecasted results. In short, it is suggested that the perspective of natural ecosystems be considered in future research, which would help to construct a better tourism ecological security index system through better mathematical algorithms to enhance the progress of a RBF neural network model. Consequently, the results could provide critical suggestions to scientifically protect tourism ecological security in Mount Putuo Island.