Ecosystem dynamics unfold into the future but are understood by examining the past. A forward looking ecology, which assesses a broad range of possible future ecosystem states, is the complement of long-term, historical approaches to ecology. Together they are the ecology of the long now. The "long now" of ecosystems includes historical influences that shape present ecologies, and the future consequences of present events. As a step in testing theories by their consequences, prediction is widely used in ecology. Ecologists have developed, criticized, and improved many predictive theories. Ecologists also have developed many empirical relationships that are potentially useful in forecasting. Eutrophication is an example of a problem for which ecologists created fundamental understanding, predictive capability, and new options for management. Ecologists frequently justify their research funding through appeals to improved predictability. This goal is sometimes attainable and in any case motivates a considerable body of insightful research. However, in many cases of environmental decision making, what ecologists cannot predict is at least as important as what can be predicted. It is important to assess the full range of changes in ecosystems that may plausibly occur in the future, and the implications of these changes. The paper discusses some ways that ecological information can be used to improve understanding of the future consequences of present choices.
