两种自然保护区设计方法——数学建模和计算机模拟

传统的自然保护区设计方法是打分法和Gap分析法,这两种方法简单易行但可靠性不高；地理信息系统(GIS)在保护区设计领域的应用也为人熟悉.关注近年来快速发展而国内使用不多的两种方法——数学建模和计算机模拟.数学建模主要用来从一组备选地块中选择一部分组成自然保护区,包括线性和非线性模型,用启发式算法或最优化算法求解.启发式算法具有速度快、灵活等优点,但解通常不是最优的,不能保证稀缺资源的最优化利用.最优化算法运算效率低,变量较多比如数百时就可能遇到计算困难,但解是最优的.预计两种算法都将继续发展.计算机模拟主要用于保护区评价、功能区划分、预测特定环境比如空间特征和气候变化对物种的影响等,多用启发式算法,与其它软件结合把结果以图画显示出来.两种方法特别是计算机模拟均要求保护区设计者有较强的专业知识.讨论了两种方法面临的问题和新的研究方向,至少包括:1)基础数据依然需要完善；2)一些新的因素比如动态性和不确定性如何在模型中考虑并与其它因素结合；3)气候变化预景下模拟参数如何评估和调整；4)如何协调保护与发展的关系；5)方法的实际应用需要研究者与决策者之间建立交流机制；6)多领域专家和相关利益方应有机会参与保护区设计.

Designing nature conservation reserves using mathematical modeling and computer simulation: a review

Traditional methods for nature conservation reserve design include scoring and Gap analysis, which are simple and easy to use, but the results of these two methods lack reliability. Geographic Information System (GIS) is another widely used method in designing nature reserves nowadays. This paper focuses on two other methods which have been developed and gained popularity in the past two decades, but not used as widely in China. These two methods are mathematical modeling/programming and computer simulation. Mathematical modeling aims to select an optimal subset from a large set of potential sites to assemble a nature reserve which is expected to protect a set of targeted species while satisfying some specific biological and/or ecological constraints. The problem can be formulated as linear and nonlinear optimization models, and solution methods vary depending on the type of formulation. When formal optimization methods fail to solve these models, due to the model size or degree of nonlinearity, heuristics are employed instead of formal optimization. Heuristic methods are computationally convenient and flexible in finding multiple solutions, but the solutions can be significantly suboptimal; therefore this approach does not ensure an optimal allocation of scarce conservation resources. Formal optimization, on the other hand, provides the best possible solution to the problem. However, solving a reserve selection problem to an exact optimum can be computationally challenging and modelers can easily encounter computation difficulties when the number of variables in the model is large. Computer simulation, the second method reviewed in this paper, is used mainly to evaluate a nature reserve in terms of its functionality, delineate functional areas, and predict the impact of specific circumstances such as spatial attributes or climate change on species' persistence. As the solution methods in computer simulation, generally heuristics are employed and the results are displayed in graphical form or even in animation by integrating the simulation software with other software. Results in such forms may look attractive but cautions should be taken in terms of model selection, parameter valuation, and interpretation of the results. Model validation is generally required in computer simulation and the results of a simulation process should be interpreted from a statistical perspective. Both methods, especially computer simulation, require and benefit from user's judgment, expertise, and especially ecological and biological knowledge. Using linear integer programming to guide optimal allocation of conservation resources is extremely important and valuable in China's nature reserve planning and design process, considering the conflict between conservation and development. In this review, relevant issues and potential new research directions are discussed, including: 1) original data needs to provide a basis on which these design methods can work; 2) some more challenging factors such as dynamics, uncertainty, and particularly spatial coherence of the reserve areas; 3) reevaluation and adjustment of model parameters under the projected climate change scenarios; 4) coordination issues of nature conservation and economic development; 5)communication mechanisms between scholars and decision-makers in applications of these design methods in real world reserve planning and design practices; and 6) participation of experts from multiple disciplines and other concerned parties in the reserve design process.