A recent United Nations (UN) study concludes that degradation of the world's ecosystems over the past century "…could have devastating implications for human development and the welfare of all species.” (United Nations, et al., 2000, p. 6). The report calls for an "ecosystems" approach to environmental problems that would foster "sustainable development," development capable of meeting the needs of today without sacrificing resources needed by future generations. The ecosystems approach would take a holistic view of problems; include the public in the management of ecosystems; integrate social and economic information with environmental information to address the ecosystems “information gap”; and bring urban planning into ecosystem management, as urbanization places severe pressures on ecosystems today. This paper argues that ecosystems management constitutes a "wicked" problem (Rittel and Webber, 1973) or a "mess" (Ackoff, 1999), and that Singerian inquiring organizations (Churchman, 1971; Courtney, Paradice and Croasdell, 1998; Richardson, Courtney and Paradice, 1999) can provide a framework for ecosystems management. A study of infrastructure decision making in Houston, Texas (Lomax, et al., 1998) is presented as an example of using the Singer model in ecosystem management. The objective of the study is to develop decision support systems for Houston's infrastructure systems, including roads, sewers, water supply, and drainage. A very broad perspective on infrastructure decision making is being taken. The relationships among built, human and natural environments are being considered. Preliminary results show that infrastructure decision environments are indeed wicked, that the Singerian model does provide a useful framework in which to investigate them, and that decision factors tend to cluster into political, need, economic, and environmental categories. The wicked domain presents special problems for DSS development. However, an effective DSS may be help alleviate the present lack of under development.