Watershed Assessment and Modeling

(7/15/2019 12:00:00 AM)

Watershed Assessment and Modeling

Why a Watershed Approach?

A watershed is the area from which runoff resulting from precipitation collects and drains through a specified location. Watershed boundaries are determined mainly by topography (with some possible variance in ground water flow paths) and are the most basic unit of natural organization and management in the landscape. Watershed management plans may be very targeted – for example, focused on pollution abatement or on management of specific species – or more general, with the goal being to balance multiple uses and objectives. Whereas water resource management in the past often focused on single issues or processes, in recent decades there has been increased recognition of the value of managing watersheds with a more holistic or integrated approach (White, 1998). Integrated watershed management has been defined as the “process of managing human activities and natural resources on a watershed basis, taking into account social, economic, and environmental issues, as well as community interests in order to manage water resources sustainably” (MVCA) (Mississippi Valley Conservation Authority, 2018).

The Watershed Inventory

The movement of water through a watershed is a function of physical attributes and processes on the landscape, mitigated by land use and land cover. Effectively evaluating management alternatives requires understanding the ecosystem processes and anthropogenic factors that govern responses to different management actions. A first step in integrated watershed management is the watershed inventory, or the process of ‘getting to know your watershed’ (Heathcote, 2009).

A watershed inventory requires a review of features and processes over a range of spatial and temporal scales. At the largest spatial and temporal scale, one should catalog and understand factors such as regional geology, climate, and resulting landforms in the basin of interest. These physical factors determine attributes such as soils properties, infiltration, and runoff, which in turn dictate water quality, quantity, timing, and movement through a watershed, including both surface water and groundwater. These attributes provide the physical template for biological communities that exist within a watershed.

In addition to naturally occurring physical, chemical, and biological processes, human activities affect processes occurring within a watershed. These human dimensions may include factors such as land use practices, social and economic systems, or valued features and activities (Heathcote, 2009). The human dimension of watershed processes consists of two primary components: (1) the constructed (or built) environment and (2) human social systems overlain on both the natural and built environment. Inventorying the built environment, for example, may consist of evaluating the types and distribution of different land-use activities in a watershed. Steps involved in such an exercise may include determining land-use categories across a region or landscape, mapping the areal extent of each activity, and perhaps assessing how land use activities may change or evolve through time.

Assessing human activities as part of the watershed inventory should be targeted to management questions of interest. For example, land-use activities affect both the type of pollutants within a watershed, and the load or mass of pollutant delivered to waterways. Different agricultural practices may contribute specific types of pollutants to riverways or waterbodies within the watershed. Other human activities, such as aquaculture for example, come with a unique set of management opportunities and challenges that should be characterized at a level sufficient to address the management questions of interest.

The watershed inventory may rely on the collection of primary data (collected by the investigator), secondary data (originally collected for other purposes), or a combination of the two. In some instances, a watershed inventory may rely heavily on collection and synthesis of existing data (‘data mining’), while in other situations assessment of available data may reveal critical data gaps that should be filled in order to make informed decisions of watershed management. Thus, in addition to providing data and knowledge to inform management decisions, the watershed inventory serves as a systematic assessment of available data, allowing determination of whether additional resources should be invested in collection of data and where greatest information needs are. For example, the watershed inventory may identify additional sampling needed to better quantify in-stream or groundwater flow, water quality trends, or attributes of aquatic ecosystems. When designing a measurement program or sampling campaign, it is essential to design the program in a way that can support specific, decision-relevant questions. For questions related to industrial effluent or nutrient load, for example, key considerations may include designing a sampling program to adequately assess the temporal variability in parameters or systems that are heterogeneous.

The fundamental understanding of a watershed gained through the watershed inventory provides the building block for creating predictive models of a system that supports evaluation of the effects of proposed management actions.

Evaluating Effects of a Proposed Action

Developing a watershed management plan often involves identifying and choosing among alternate management paths or strategies. In complex systems, the outcome of an action or a combination of different actions may not be readily apparent. Often an assessment framework may be needed for systematically evaluating or predicting the effects of proposed management actions. Several different strategies or methodologies exist for environmental assessment that may be appropriate for use in developing a watershed management plan.

Initial screening assessments are intended to identify which of the possible management alternatives are feasible. Of those that are feasible, additional review may be needed to determine which results in best overall performance in terms of meeting specified objectives. For example, the sequence of devising a management strategy may include (1) identification of management objectives, (2) evaluate decision outcomes, and (3) prediction of decision outcomes, and (4) assessment of outcomes (U.S. Fish and Wildlife Service, 2008). Assessment of outcomes may include quantitative metrics or statistics, or a more qualitative approach, such as a ranking system. Predicting outcomes may require use of a modeling approach to anticipate results from natural processes, or a combined set of actions.

Modeling fundamentals

In watersheds and other complex systems, the outcome of a proposed set of actions or combination of actions may be uncertain or difficult to predict. Models provide a structured approach for evaluating system response to management actions and to test hypotheses related to system dynamics. Models are representations of processes or systems, and may take the form of a conceptual, physical, or numerical model. Simply stated, ‘models are conceptions of physical reality that result in qualitative or quantitative predictions’ (Darby and Van de Wiel, 2003). All models are simplifications of reality and come with a series of assumptions and limitations that dictate applicability and realism. Some models may not be suitable for rigorous predictions, but rather for evaluating hypotheses, or for comparative analysis of different simulated outcomes.

When is a model appropriate or necessary for application to a water resource management question? The US EPA identifies three situations when a modeling approach may be valuable (U.S. Environmental Protection Agency, 2017):

·        To scope or quantify a problem

·        Predict how conditions are expected to change over time

·        To evaluate alternative management options

The first type of model – the scoping model - may be used to quickly estimate the extent and severity of a problem. For example, the objective may be to compare levels of stress, prioritize areas or sources of impairment, examine trends, extrapolate monitoring data, or evaluate direction of system responses. The scoping model approach is often employed to build fundamental understanding of a water-quality problem. In the second situation – prediction through time – a model may be used to forecast future conditions resulting from a specified set of conditions, either natural or resulting from human actions. The third situation – evaluating alternative management options – may be a used to compare relative effects of different courses of action proposed in a watershed.

In the Mekong River Basin, an array of different modeling approaches has been used to evaluate the effects of different combinations of proposed management actions. These previous studies or assessments have spanned a range of spatial and temporal scales. For example, at the scale of a river basin, a variety of approaches have been used to evaluate the effects of dam construction on sediment dynamics in the Mekong (for example: Mekong Delta Study; Kondolf et al., 2014). Smaller scale studies have focused on local rice production, farming methods, and sea level rise (for example, Chapman and Darby, 2016). These models range from conceptual to more quantitative, and in the large-scale Mekong Delta Study, multiple approaches are used to simulate critical processes. A commonality of the studies is that each model is targeted to the management question of interest.

A key consideration at the onset of any modeling exercise is to determine the appropriate modeling tools and data needs to match the questions at hand. For example, when developing hydrodynamic models, practitioners must determine the dimensionality of the model, the time period over which the model will be run, or whether the model state is steady versus unsteady. Additionally, as is the case in large, complex river systems tackled in the Mekong Delta Study and the Missouri River case study, described below, the modeling approach may entail selection of higher-resolution models nested within larger-scale lower-resolution models.  

Missouri River Basin: An IWM Case Study

Background

The Missouri River is the longest river in the United States, extending from its headwaters in the northern Rocky Mountains to its confluence with the Mississippi River in the central United States (Jacobson and Galat, 2006). The river drains more than 1,300,000 km2, and the watershed includes both mountainous headwaters and parts of the Great Plains of the US, an area where land use is dominated by grazing and agricultural production. The Missouri River has been subject to over two centuries of extensive river engineering. The river is home to six large mainstem dams that form the largest reservoir system in North America. The dams have substantially altered the natural flow regime. Though the effects of flow regulation on river hydrology vary depending on location in the watershed and the contribution of tributaries, in general, regulation has decreased the magnitude of annual floods and increased summer base flows.

In addition to changes in the natural flow regime, the channel has been substantially altered. Historically, the Missouri River had a broad, shallow, braided channel that migrated across its floodplain and was well connected to adjacent wetlands and riparian habitats. In the interest of navigation and flood control, the United States (U.S.) government channelized much of the lower 1,200 km of the river, designing a self-dredging navigation channel that was much deeper, narrower, and swifter than what had existed historically. To accomplish these channel changes, much of the lower river has bank revetment and river control structures such as wing dikes. The result is a river that is easier to maintain for navigation, but the natural riverine habitat is very different than what existed prior to basin-wide development and river engineering.

Missouri River Integrated Watershed Management

There are eight purposes, or uses, of the Missouri River that have been officially authorized by the U.S. Congress: flood control, navigation, irrigation, hydropower, water quality control, water supply, recreation, and fish and wildlife. As is the case in rivers and watersheds around the world, the different water uses in the Missouri River basin are sometimes in conflict with one another. In addition to the authorized purposes, there are three federally-listed threatened or endangered species in the Missouri River that must be considered in management decisions. These species are two birds (piping plover and least tern) and one fish species (pallid sturgeon). The U.S. Army Corps of Engineers, the federal agency responsible for most of the river engineering on the Missouri River, is required to consider the effects of management actions on these three species.

The Missouri River Recovery Program (MRRP) is a program whose goal is to replace lost habitat and to identify actions that will avoid jeopardizing the continued existence of the three species of concern (U.S. Army Corps of Engineers, 2013). The multiagency team that implements the recovery program considers how management actions will affect not only the fish, birds, and other animals that rely on and live in the river, but also the authorized purposes, cultural resources, and tribal interests (Jacobson et al., 2015). As part of the adaptive management process on the Missouri River, multiple federal agencies and partners worked together to produce a Draft Environmental Impact Statement (DEIS). The DEIS identified six different alternatives (an alternative defined as a combination of management actions) that meet the objectives of the MRRP, including a preferred alternative. This preferred alternative was selected based on number of factors, such as economic and environmental effects, and after a period of public input.

Detailed numerical and statistical models have been used to inform understanding of how different management actions may affect target species within the basin, and affect other authorized uses and stakeholder interests. For example, to evaluate the effects of different flow regimes and channel restoration practices, hydrodynamic models have been used to simulate drift and dispersal of larval fish in the Missouri River Basin. The results of these hydrodynamic model feed into population models that are used to simulate the effects of different management actions on the abundance and distribution of native sturgeon in the basin (Jacobson et al., 2016). A rigorous modeling process has also been used to predict the effects of different management actions on the authorized purposes and human considerations, thus providing stakeholders with an understanding of how management actions will affect their interests or livelihoods.

Management of the Missouri River is very complex and, at times, contentious. The implementation of a robust science program in the basin has been an essential step in providing the information necessary for government officials and stakeholders to make informed decisions.  

 VIFEP   (USAID workshop)