约束线方法在生态学研究中的应用

Constraint line methods and the applications in ecology

生态系统格局和过程往往受到多个因子的共同影响,故此反映两个生态学变量相互关系的散点图常常会表现为有边界的散点云。基于数据均值或中值分析的传统统计方法不适用于散点云数据的分析。散点云所表征的不是两变量之间的相关关系,而是限制作用关系。约束线(包络)方法为提取散点云边界、理解限制变量对响应变量的作用,以及预测响应变量的潜在最大值提供了有效手段。该文对应用约束线方法研究生态学问题所取得的成果进行总结与归纳,介绍了约束线概念的发展历程、提取方法,从物种分布、种群行为及作物产量优化三个方面总结了约束线方法的优点及适用性,概述了当前约束线方法应用研究面临的问题与挑战,指出约束线方法应结合其他统计方法,实现对生态过程的准确理解,此外,还应重视约束线方法的尺度依赖性。最后,该文展望了约束线方法在研究生态系统服务关系和土地系统优化等方面的应用前景。

With increasing data availability in the big data era, many traditional statistical analyses based on the mean or me- dian are insufficient or inappropriate to elucidate the complex patterns of variation. This is particularly the case when multiple factors are involved and the bivariate scatter occurs as scatter clouds. In such circumstances, con- straint line （or envelope） method could be an alternative and effective tool to extract the data boundaries, thus im- proves our understanding of the complex relationships between limiting factor and response factor. Here, we syn- thesize the major fmdings and achievements in the field of applying the constraint line method in ecology. Spe- cifically, we first describe the history and development of the constraint line method. We then discuss the tech- niques to establish the constraint lines with examples, and discuss the applications and implications of the con- straint lines in species distribution, population performance, and optimization problem. We suggest simultaneously application of both constraint lines and regression techniques to the same datasets to achieve a comprehensive understanding of ecological process and underlying mechanisms. Such combined methods should be used with special attention to the role of spatial heterogeneity and scale dependency. We also discuss in detail the potential applicability of the constraint line method in studying the linkages between ecosystem services, and land system design.