Universidad Politécnica de Madrid, Escuela Técnica Superior de Ingenieros de Minas, Calle Ríos Rosas 21, 28003 Madrid, Spain.
Sci Total Environ. 2012 Dec 1;440:140-53. doi: 10.1016/j.scitotenv.2012.07.071. Epub 2012 Aug 25.
The services of natural ecosystems are clearly very important to our societies. In the last years, efforts to conserve and value ecosystem services have been fomented. By way of illustration, the Natural Capital Project integrates ecosystem services into everyday decision making around the world. This project has developed InVEST (a system for Integrated Valuation of Ecosystem Services and Tradeoffs). The InVEST model is a spatially integrated modelling tool that allows us to predict changes in ecosystem services, biodiversity conservation and commodity production levels. Here, InVEST model is applied to a stakeholder-defined scenario of land-use/land-cover change in a Mediterranean region basin (the Llobregat basin, Catalonia, Spain). Of all InVEST modules and sub-modules, only the behaviour of the water provisioning one is investigated in this article. The main novel aspect of this work is the sensitivity analysis (SA) carried out to the InVEST model in order to determine the variability of the model response when the values of three of its main coefficients: Z (seasonal precipitation distribution), prec (annual precipitation) and eto (annual evapotranspiration), change. The SA technique used here is a One-At-a-Time (OAT) screening method known as Morris method, applied over each one of the one hundred and fifty four sub-watersheds in which the Llobregat River basin is divided. As a result, this method provides three sensitivity indices for each one of the sub-watersheds under consideration, which are mapped to study how they are spatially distributed. From their analysis, the study shows that, in the case under consideration and between the limits considered for each factor, the effect of the Z coefficient on the model response is negligible, while the other two need to be accurately determined in order to obtain precise output variables. The results of this study will be applicable to the others watersheds assessed in the Consolider Scarce Project.
自然生态系统的服务对我们的社会显然非常重要。在过去的几年中，人们一直在努力保护和评估生态系统服务。例如，自然资本项目将生态系统服务整合到全球日常决策中。该项目开发了 InVEST（生态系统服务和权衡的综合评估系统）。InVEST 模型是一个空间综合建模工具，可用于预测生态系统服务、生物多样性保护和商品生产水平的变化。在这里，InVEST 模型应用于地中海地区流域（西班牙加泰罗尼亚的 Llobregat 流域）的土地利用/土地覆被变化利益相关者定义的情景。在所有 InVEST 模块和子模块中，本文仅研究了供水模块的行为。这项工作的主要新颖之处在于对 InVEST 模型进行了敏感性分析 (SA)，以确定当模型的三个主要系数 Z（季节性降水分布）、prec（年降水量）和 eto（年蒸散量）的值发生变化时，模型响应的可变性。这里使用的 SA 技术是一种称为 Morris 方法的逐个变量 (OAT) 筛选方法，应用于 Llobregat 河流域分为的 154 个子流域中的每一个。结果，该方法为每个子流域提供了三个敏感性指数，这些指数被映射到研究它们的空间分布。通过分析，该研究表明，在所考虑的案例中，在考虑的每个因素的限制范围内，Z 系数对模型响应的影响可以忽略不计，而其他两个因素需要准确确定，以便获得精确的输出变量。本研究的结果将适用于 Consolider Scarce 项目评估的其他流域。
