Gordon Grant
USDA-Forest Service Pacific Northwest Research Station, Corvallis

To begin, watershed analysis is not an original idea. The notion that we ought to know something about landscapes before we start changing them reaches back at least as far as the ancient Chinese art of geomancy - siting buildings so that the gods smile propitiously on them and they don't fall down. More recently, the roots of watershed analysis can be found in the writings of certain conservationists - John Powell, Clarence King, Gifford Pinchot - who expressed the notion that the landscape is to be understood and human actions should be designed with that understanding in mind.

From these writings grew the concept that a watershed represents a reasonable and relevant demarcation on the landscape for land-use planning. The Tennessee Valley Authority and other early river basin planning strategies provide experience as important as more recent examples. If we look at the TVA, we see an important idea opposed by those who viewed it as unwarranted government interventionism in private affairs. We need to consider that experience when we consider the brave new world of watershed planning and management in which we are now engaged.

The concept of watershed analysis is closely related to the issue of cumulative effects. For several decades, we have grappled with how land use activities on federal, state, and private lands interact to affect hydrologic and ecologic processes. Many of our current problems - endangered species, declining salmon populations, forest fragmentation - are evidence of our inability to deal effectively with cumulative effects. In a sense, watershed analysis has evolved as a kind of pro-active analysis of cumulative effects, conducted prior to developing management plans rather than in response to predetermined action.

Historians talk about ideas emerging simultaneously in different places. A document of our current phase of history would note that the idea to use watersheds as units of planning and analysis has evolved in many government and private agencies seemingly independently. The process may not have been truly independent, but rather, in a Biblical paraphrasing, a story of begats: first there was PACFISH; PACFISH begat SAT; SAT begat FEMAT; FEMAT begat WA (or at least prescribed it).

Ideas have evolved through the process. However, the basic concept remains that watershed analysis is a mechanism to address inconsistencies between the current scales of planning and the direction to implement ecosystem management. These inconsistencies arise on Forest Service lands, for example, as Congress sets both the commodity output levels, through specified timber targets, and environmental direction, through legislation such as NEPA and NFMA. Forest Plans attempt to implement these targets and direction. As we know, the courts have found that the commodity outputs and environmental direction are fundamentally incompatible, and Forest Plans have been challenged as inadequate.

Attempts at regional scale conservation of owls, salmon, old-growth and various species, required that landscape planning and management be more spatially explicit at scales such as physiographic provinces, river basins, and watersheds. Regional scale conservation strategies, such as PACFISH or FEMAT, called for landscape or watershed analysis as a way of focusing conservation strategies to specific landscapes.

So watershed analysis presents a very simple idea, that a comprehensive and systematic analysis of a landscape can and should inform landscape management. Originally conceived as a method to tailor riparian management, WA has quickly broadened (theoretically, at least) to address a full range of terrestrial and social objectives as well. The current expectation for watershed analysis is that it will do more than develop effective riparian reserves; it will provide the analytical framework to accomplish landscape design.

As many of you who participated in FEMAT and some of the other regional conservation strategies know, the integration of terrestrial and riparian issues, as well as social expectations, has proven difficult. In addition to its technical application, WA is expected to be an interagency process and forum. It is expected to involve the public in some way not yet fully understood. Management activities will be placed in the context of variability of historical disturbance regimes. And there is a great deal of uncertainty as to how planning is going to proceed in the face of what looks like a reconstituted planning universe.

Beyond all these expectations for WA is the big question: how do we use the information from watershed analysis to make better land-use decisions? WA was developed to meet a specific set of objectives that reflects many of the dominant issues that we face in the months and years ahead. The political context in which watershed analysis has emerged establishes new benchmarks against which watershed analysis will be judged, rightly or wrongly. There is not only the legal mandate to analyze cumulative effects, an issue that watershed analysis intrinsically addresses, there are also legal requirements imposed by the Endangered Species Act and other legislation to design protection schemes for riparian and other organisms. Unless we can reach agreement about how we, as a community of scientists, managers, and the larger public, use WA to design effective environmental protection, then WA will never achieve its potential.

Some of the most challenging objectives of WA are also the most exciting. Watershed analysis could provide a common framework for evaluating, planning, and managing watersheds. It could carry us beyond defining protection schemes to designing landscapes to meet varying objectives. Perhaps the most ambitious goal of WA is to serve as a basis for interagency and multi-user interactions and agreements regarding land-use decisions. How do we do watershed analysis? What is the technical framework, sequence of tasks, relationship with planning? The examples presented at the workshop represent a number of different approaches emphasizing different objectives driving land-use/landscape planning. Most stratify the landscape into analysis units, and then examine how watershed and ecosystem processes are distributed through these units. Most of the examples follow the analysis with a synthesis of individual components that connects the landscape units into a blueprint to guide management activities.

How will we evaluate these examples? One reference point may be to ask how well each addressed a set of key questions - the pillars of WA (above, right). A fundamental product of WA is an analysis of landscape process, condition, structure, and change that allows us to determine what human activities are fundamentally incompatible with the landscape.

In conclusion, we see there are some salient historical issues that brought us to this juncture. These case studies represent different fledgling approaches to a complex problem. They come from different landscapes with different objectives, they had different players involved, they were funded to different degrees, they have different institutional investments behind them, and they have different outcomes and products attached to them. It will be interesting to see how well we can compare, contrast, and glean useful information from them. Along the way, we may learn something equally important: how to learn from each other.