Chapter 5

Watershed Assessment, Modeling, Planning and Evaluation

5.1 Introduction

The defining features of a watershed include drainage divides, hill slopes, floodplains, terraces and deltas. 
The boundary of a watershed is based on the drainage divides, which take into account both hydrologic and
geomorphic processes.  Other determining factors include climatic regime, topography, underlying geology,
morphology, soils, and vegetation.

Watershed assessment is a necessary component of a monitoring program in order to determine what

degraded or impaired areas may exist in the watershed and why. Several characteristics of a watershed are
taken into consideration during the assessment process including land use, land cover, and hydrology. Land
use and cover is considered in a historical as well as current perspective to determine the types of activities
that have occurred in the watershed and their potential as sources of pollution. It is important to consider the
natural and cultural resources of the watershed as well as the human activities. This information may be
obtained from various sources including topographic maps to determine drainage area and land features as
well as land use data.

Much has been said about the need to use "holistic" perspectives that consider the entire watershed when
contemplating stream restoration options. Unfortunately, political, programmatic, and jurisdictional
boundaries seldom correspond with watershed boundaries and restoration projects focus on specific sites.
Without a comprehensive reach or watershed assessment, selected restoration measures often ignore
underlying problems at a broader scale and are either ineffective or not cost effective relative to other
measures (Figure 5.1).

Figure 5.1: Watershed conditions can dictate processes that affect restoration potential in reaches far downstream

A reconnaissance and assessment of watershed character is necessary to:

 Assess watershed conditions to determine the causes and nature of impairment
 Determine feasibility of using restoration or other management options to meet objectives

In some cases, ecological restoration is the most effective response to impairment; in other cases, restoration
may be one among many candidate tools for achieving objectives. To determine the appropriate actions, it is
necessary to collect, compile, analyze, and interpret environmental data rapidly to facilitate management
decisions and resultant options for preservation and control or mitigation of impairment. The technical note of
watershed assessment considers watershed and reach reconnaissance techniques that possess the following
principal elements:
 Cost-effective
 Facilitate comparisons among sites
 Quick, yet scientifically valid
 Easily presented to the public
 Environmentally-benign procedures

Assessing a watershed to understand its current condition, and how it got there, is usually the first step taken
in developing a strategy toward improving and protecting the watershed’s condition. A relevant watershed
assessment addresses the sources of watershed impacts rather than just their symptoms, which key to
achieving effective watershed protection and restoration.
An assessment is far more than an encyclopedic collection of information about the watershed—it must
analyze why the watershed is in its current condition. Watershed assessments can be relatively comprehensive
or be focused on several specific issues. However, watershed practitioners must make a choice between a
broad assessment and a focused assessment after thoroughly discussing the advantages and disadvantages of
each in light of the watershed’s needs and the purpose of the assessment.

A watershed assessment is: “a process for analyzing a watershed's current condition and the likely causes
of these conditions.”

A watershed assessment report is: “a report documenting the findings of the watershed assessment
process.”

Within a watershed common zones, often used for management purposes, are: 1) upland (land above the zone
inundated by floods or above the transition between riparian and terrestrial vegetation), 2) riparian (vegetated
area between the waterbody edge and the upland area), and 3) water body (any stream, river, abandoned
channel, pond, lake, wetlands, estuary, or ocean). The term “watershed” is not synonymous with terms such as
“stream” or “riparian corridor” or another single feature of the watershed.

5.2 Watershed Assessment: What It Is and What It Is Not

Coming to a common understanding of what watershed assessment is—and what it is not—is important for the
users. “Watershed assessment” definitions include the following:

 The analysis of watershed information to draw conclusions concerning the conditions in the watershed.
(Nehalem River Watershed Assessment, Washington, 1999).
 A process for evaluating how well a watershed is working. (Oregon Watershed Assessment Manual,
WPN, 1999)
 A process that characterizes current watershed conditions at a coarse scale using an interdisciplinary
approach to collect and analyze information. (NCWAP & CDF 2002)
 The translation of scientific data into policy-relevant information that is suitable for supporting decision
making and action at the watershed level. (Watershed Academy, U.S. EPA).

Despite their differences, what is common to each definition is a process composed of actions—analysis,
integration, translation—that leads to the interpretation of information about the watershed’s current condition.
It is critical that the watershed assessment effort lead to a better understanding of watershed processes and
conditions and why the watershed is in that condition. That way, the assessment can serve as a compass to help
direct further actions.
Your assessment should move beyond a simple description of what a watershed looks like, or what historical
activities took place in the watershed. While those are some of the building blocks of an assessment, your
assessment must connect past and current human activities and land uses (causes) to watershed processes and
current condition (effects). (Watershed processes refer to the natural processes, such as hydrologic and nutrient
cycles, that influence the waterways’ conditions). With an understanding of dominant watershed processes and
potential causes of watershed condition, watershed practitioners can propose solutions to problems. Without this
understanding, proposed solutions may address only the symptoms. A successful watershed assessment leads to
the implementation of actions that benefit watershed processes and conditions—the ultimate “performance
measure”.

What an Assessment Is
 The scientific interpretation of watershed information and data, leading to conclusions about watershed
condition
 An objective problem-solving tool that identifies the potential causes of problems
 An objective problem-solving tool that identifies the potential causes of problems
 A tool to help identify available data or information gaps
 Analysis and findings that can be used to develop appropriate actions
 A component of a watershed management package that leads to planning, implementation, evaluation,
and additional monitoring
 A product that is useful for its audience

What an Assessment Is Not

 Monitoring and data collection only
 A list of data only
 A consolidation or summary of existing information only
 Historical conditions or “baseline” conditions only
 An identification of symptoms of problems only
 A plan
 An endpoint

The following steps are basic process for conducting a watershed assessment, planning and evaluation
– 1) Organize the assessment team
– 2) Define the purpose and develop a plan for the assessment
– 3) Collect data and information
– 4) Analyze the data
– 5) Integrate and report the data to inform decision-making

Step 1: Organize the assessment team

Your organization has decided that it wants to do a watershed-scale assessment. In some cases, there are
conditions and impacts in your watershed that have raised concerns. You are interested in finding out why
these impacts have occurred. In other cases, your watershed may be relatively pristine and you want to
maintain its character. You are interested in assessing conditions and analyzing potential sources of
degradation so you can prevent or minimize future problems.

Assemble the Team and Committees

No one has all the expertise required to do an assessment, not consultants, agencies, academics, or watershed
groups. Accordingly, your assessment team should include people with a wide variety of expertise and
interests.
 Create a Realistic Schedule

It’s important to be realistic about how much time it takes to perform a watershed assessment, but estimating
time required can be challenging. Experience has shown that simpler assessments performed in-house with
sufficient expertise and information may take four to eight months, More complicated or comprehensive
assessments or assessments where the process is not under tight scheduling control can take as long as 36
months. Use milestones to stay on track. Here are some sample milestones (adapted from Coastal
Conservancy 2001):
􀂃 Start-up
􀂃 Initial project team meeting (define approach)
􀂃 Public meeting #1 (review issues, concerns)
􀂃 Technical Advisory Committee (TAC) meeting #1 (review strategy)
􀂃 Begin assessment
􀂃 TAC meeting #2 (mid-progress review)
􀂃 Draft assessment complete
􀂃 Review results—TAC and Public Advisory Committee
􀂃 Release revised draft to public
􀂃 Revise and deliver final assessment

Involve the Community

Those who will be making decisions using information contained in the assessment should be included,
consulted, or at least considered when designing an assessment. From start to finish, the assessment should
make clear how and why various steps were taken. This approach has the benefit of getting all-important buy-
in—stakeholders and decision-makers are more likely to trust the assessment’s conclusions if they understand
the reasons various approaches were taken or they were involved in gathering data and information for each
step.

Steps
􀂾 Assemble the assessment team and committees.
􀂾 appoint a coordinator and seek contractors, if necessary.
􀂾 Encourage community participation through public meetings, the media, and outreach to other relevant
local organizations (e.g., Farm Bureau, resource conservations districts, etc.).

Once you have your watershed team assembled, you can actually begin the work. One way of organizing a
watershed assessment is to break it into four main parts:
• Defining the problem and planning the assessment
• Collecting information and data
• Interpreting results: data analysis and synthesis
• Preparing the report

Step 2: Define the purpose and develop a plan for the assessment

The first formal phase of a watershed assessment consists of clearly identifying the issues of concern,
identifying the purpose of the assessment, developing a conceptual diagram of the key components of the
watershed, and developing a plan for carrying out the assessment.

I. Formulate the relevant questions and goals

Watershed assessments may be motivated by one or more influences:

 to evaluate watershed conditions from a neutral perspective, i.e., with no prior assumptions;
  to address identified watershed issues or problems;
 to meet a particular purpose, e.g., identify conditions that need to be improved in order to increase
drinking water quality;
 To meet a particular goal, such as educating the public about natural and human features of the entire
ecosystem and assist in planning and decision-making.

For many assessments, one or more issue-based questions usually drive the process. The question may be as
generic and general as, “What is the condition of our watershed, and why is it that way?” More specific
questions might be along the lines of, “Why did the salmon stop spawning in our stream? Why did such a big
flood come from such a small storm? Why can’t we drink the stream water anymore?” or “How can we protect
our pristine watershed from the degradation we see in neighboring watersheds?” Questions based on
observations and community concerns will direct the watershed assessment, which will in turn provide the basis
for addressing the important issues.

Issues to Consider
 If there are no fundamental questions or concerns guiding a watershed assessment, you may wish to make
explicit the perceived need for the assessment.
 The questions should be stated clearly enough to capture the prevailing concerns that led to wanting or
needing a watershed assessment.
 Clearly write out the questions and/or issues use them to guide future data collection and analysis.

II. State the purpose of the assessment

Watershed assessors should develop a clear statement of purpose. A “fuzzy,” or implied purpose statement that
never gets clarified, or an absent purpose statement can lead to bigger and bigger problems (such as getting off
target, or creating misunderstandings due to different expectations of the product) as the assessment process
continues.
It is also important to clearly identify who really wants the watershed assessment, and why they want it.
Otherwise, misunderstandings can occur. For example, the impetus may come from the local level—from a
cooperative group (e.g., a watershed council), a local agency (e.g., a resource conservation or water district),
or other private or public stakeholders—for a variety of reasons.

The comprehensive approach assesses the conditions of all processes and features in a watershed. The
advantages are that this broad approach gives an overview of the watershed’s condition, may expose
previously unknown problems in the watershed, and may identify the interconnections between various
problems or issues. On the other hand, “comprehensive” may sound desirable, but a focused product may
prove more useful.

In the focused approach, the assessment process chooses the most critical issues in the watershed, and then
focuses the assessment effort on these. The benefit to this approach is that it makes the assessment potentially
more useful for future decision making about specific problems or areas. Groups identify upfront the issues—
of all those possible—that most need to be addressed because the assessment cannot address all issues in
depth. The watershed’s problem(s) drive the assessment. The risks of this approach are that the focus can
become too narrow, miss critical issues, and overlook connections among problems/issues, resulting in a
failure to correctly identify the root cause of problems

State What the Watershed Assessment Will Be Used For

Assessments generally serve to inform certain functions:

 General watershed management planning with multiple purposes
 Regulatory concerns
  Restoration or enhancement planning
 Monitoring program development
 Management of areas at risk and practices resulting in risk
 Land use activities

III. Define the geographic boundaries of the watershed

Establishing the boundaries of your watershed assessment or the spatial limits of the area to be analyzed is a
critical early step. The only watersheds defined by nature are those with a low point at the ocean or a closed-
basin lake. All others (including those contained within a “naturally-defined” watershed) are defined by a
human choice of the lowest point. Agreeing on the assessment area at the outset so that everyone knows
exactly what piece of ground is under discussion can head off many problems and arguments.

IV. Develop a basic overview of the watershed (past and present)

When you defined the boundaries of the watershed, you identified the spatial scale of the assessment. You
also need to define the temporal scale. How far into the past and into the future do you plan to collect data?
Ideally, having data that extends over a period of many years is best because you can get the clearest picture
of the changes that have occurred. Also, year-to-year variability is inevitable. Having data that extends over
many years will permit you to distinguish between natural variability and a real alteration or change.

V. Identify the watershed processes and/or valued ecosystem components on which you will focus

“Watershed processes” refers to the natural physical, chemical, or biological processes that interact to form
the terrestrial and aquatic ecosystems (the water cycle, for example). “Valued ecosystem components” refers
to the things within the watershed that stakeholder’s value, such as fish, clean water, trees, or open space. In
other words, these components may be structural (population of a certain species of fish) or functional (such
as the return frequency of fire). It is not necessary in every case to directly measure a component or process.
Frequently, surrogates or indicators can be used to get an idea of the condition of the selected watershed
component. Looking at a simple example, you might not be able to directly measure the population of the
splittail fish, but by measuring different habitat and water quality characteristics (e.g., water temperature), you
can get a good idea of whether or not this fish could survive in these conditions.
There are many possible watershed processes and attributes. These include, for example, the distribution of
benthic macro-invertebrate communities, drinkable water, presence of a species of fish or a plant that is
important to the stakeholders, or, more generally, the overall riparian corridor or upland areas.
Some criteria that are often used to select the watershed processes or components that could be the focus of
the assessment are:
• Importance to the health and sustainability of the watershed;
• Related to the assessment’s purposes;
• Sensitive to those activities or factors you suspect might be causing changes in the watershed;
• Have societal value; in other words, are important to the community, region, or state.

Additional criteria might include watershed processes for which the natural variability is known; attributes
required by a regulation; and the availability of data, models, or knowledge about the particular endpoint.

VI. Develop a conceptual model of the relationship between human activities, processes and
conditions, and potential impacts on watershed condition.
A conceptual model is a graphical representation of potential relationships among the watershed’s
components and processes. Once you have identified the watershed processes you are most interested in, you
will need to think about how they are impacted by changes in regional and watershed processes and the stress,
or impacts that may result from human activities. The relationship between human activities, watershed
processes, potential impacts or sources of stress and the effects on ecosystem function are depicted in the
conceptual model. Watershed assessments typically focus on those alterations that are human-induced since
these are the ones we can influence.
Developing a conceptual model is an iterative process. The model you develop at the beginning of your effort
may look different from the one you finally adopt because you will modify it as your knowledge of the
conditions and processes within the watershed grows. The knowledge needed to develop a conceptual model
will come from familiarity with the particulars of your watershed as well as from general knowledge about
watershed science.

Figure 2 Simple conceptual model of an urbanizing watershed

Conceptual models can be developed in a variety of ways. These models will help you understand the
possible relationships that are important to consider when you collect and/or analyze data. The diagram in
Figure 2 is an example of a simple conceptual model of an urbanizing watershed. In this example, the
assessors were interested in the health of the benthic macro-invertebrate community in their local creek. They
were concerned that runoff from agricultural fields and new development was impairing the diversity and
viability of the insects.
The more complex model shown in Figure 4 is a refinement that reflects more detail as the group’s
understanding of the watershed components and processes increased.
Clearly, the knowledge required to draw accurate conceptual models can be significant. That is why having a
team of people with varied technical backgrounds is very helpful. These are just two examples of construct a
conceptual model. Regardless of how yours looks, the key point is that the conceptual model diagrams should
identify hypothesized relationships between human activity, changed conditions or processes in the
watershed, and the potential effects of these changed conditions on the selected watershed processes and/or
components. These relationships can serve as the basis for data collection and analysis.
 Figure 3: More complex conceptual model for an urbanizing watershed

Now that you have figured out what your assessment questions are, developing and implementing an
“analysis plan” for collecting and analyzing information and data is typically the next step. Data collection
and analysis constitutes the heart of the watershed assessment. The conceptual model or diagram you
constructed can serve as a guide. Accordingly, as you prepare for the analysis phase of your assessment, you
should identify the data and information that must be gathered and outline the process for organizing and
analyzing this material.
The watershed assessment focuses in part on the potential harmful effects of human activities on watershed
components and functions. These effects occur when human activities cause changes in the watershed’s
physical, chemical, or biological characteristics and processes.
• Physical changes include water temperature and flow rate, generation and transport of sediment, stream
channel shape and connectivity with the floodplain, erosion and incision of the stream bank, and any
other physical characteristic that makes up the habitat on which the watershed processes being evaluated
depend.
• Chemical changes include the introduction of pesticides, excess nutrients, oil/grease, effluent from
industry, or other contaminant to the targeted habitat
• Biological alterations that might be associated with harm could include invasive species and pathogens.
Step 3: Collect data and information

I. Determine the kind of data you need to collect

The data you collect should correspond in type and substance to the questions rose at the beginning of
assessment planning and refined in the conceptual model. Types of data include spatial or geographic data for
understanding things taking place on the landscape (such as land use) and water quality or quantity data for
understanding a waterway.

The conceptual model can serve as a guide to what data you need to collect. The following list includes key
classes of information that typically are useful.

• Data on human activities and land uses – the location, type, intensity, areal extent (acreage), and
proximity to or linkage to the waterways (such as via storm drains)
• Data on the physical, chemical, and biological properties and potential sources of impacts in the
watershed – in-stream and riparian habitat characteristics, water quality data, animal/plant population
abundance and diversity, etc.
• Data on alterations in watershed processes – changes in the hydrological cycle, nutrient cycling, etc.,
particularly as they are affected by past and current water and land use and by climate change.
• Data on potential effects of potential impacts on watershed functions

These and other data can come in a variety of forms: digital and non-digital spatial data, quantitative or
qualitative data, and anecdotal information. All these data types can be useful in a watershed assessment.
Specific types of data include quantitative water quality data, geomorphological surveys, biological surveys,
maps, and other similar data. These data may be presented in various formats and at varying levels of detail of
analysis. For example, data might be presented in spreadsheets, tables, and graphs; as spatial data within or
separate from a computer-based geographic information system; in internal agency memoranda; in field
surveys, in narrative or historic descriptions of a place and past processes and events, such as floods,
landslides, or contaminant spills; and products of computer models developed to illustrate specific processes
(e.g., storm-water runoff). Consider collecting any type of data that would be useful for the goals of your
assessment.
Anecdotal information may be one of the most difficult data types to record and store, but may also provide
knowledge about watershed processes that might otherwise be lacking. Common types of anecdotal data
include:

1) The extent of salmon runs in rivers now lacking these runs due to dams or other barriers,
2) Increasing turbidity of streams and rivers over time due to upstream activities,
3) Encroachment of roads and human structures into previously undeveloped landscapes,
4) Growth of nuisance vegetation (e.g., benthic algae, riparian weeds, or invasive exotic weeds), or
5) Increased rate of flooding in river valleys due to landscape modification.

Although not the same as quantitative information, anecdotal data can provide very useful information and
help you develop hypotheses about historic conditions in the watershed and the effects of human activities.
 II. Identify sources and collect existing watershed data and information

The following section references sources of watershed information and data that will be available.

Waterway data
Hydrology and Flooding Data
Riparian Vegetation and Wetlands Data
Physical Watershed, Channel, and Habitat Conditions Data

III. Identify sources and collect existing spatial data about the watershed

Landscape data are often collected for areas. They may also have been collected initially at individual survey
sites (e.g., soil or vegetation) and then subsequently generalized to areas. Currently, most contemporary data
about a watershed landscape is collected with a geographic reference point. In contrast, historic data may be
very valuable, but lack easily usable or identifiable reference points.
Non-digital Spatial Data
Digital Spatial Data and GIS

IV. Develop a system for archiving and managing your data

As data are collected, they should be organized in a manner that suits the questions being asked and the users’
needs. Because watershed assessment usually involves the collection of several different types of data (e.g.,
maps, water quality, and field surveys), consider developing file organizational systems for each type of data
that conform to a single standard for categories (e.g., wildlife habitat, water quality, land use). One way to
keep track of information collected is to make a database of the category types. If you will be collecting data
for aquatic and terrestrial systems and of various different types (i.e., from text to digital spatial data), then
keeping track of the types of data and the areas they cover will help in both organizing the data and describing
how much of the watershed they cover.
For most watershed assessments of modest scale, you don’t need to become a database expert, but you should
learn enough to choose an adequate structure for your data needs. Existing environmental data will be stored
in some sort of file structure or database, with the particular details dependent on the type of data, the agency
archiving the data, and the needs of data users.

V. Identify data gaps and collect new data, when needed

As you collect and organize your data, you will quickly identify important areas of concern for which you
have no or very little data. For example, you might be concerned with alterations in the hydrological cycle in
your watershed, but don’t have any data or information on stream morphology (% pools and their size, % fine
sediment, etc.) except for anecdotal information. If in-stream habitat is an important factor in your assessment
and you are not in a position to collect this data, you will have to identify it as a data gap in your report and
consider the uncertainty that results when you analyze the data you have. It will be a limitation in your
assessment – but there are limitations on just about everything anyone does so this may not be a fatal flaw.
You might identify the lack of data on stream morphology as a priority when future funds become available
and explain how this information might help provide a more complete picture of watershed conditions.
Alternatively, if you have the resources and time to fill the identified data gaps, the uncertainty of the
assessment can be minimized with the addition of the new information. If your budget is limited, many times
there are less sophisticated methods to collect the same information; methods that volunteers or high school
students can learn with a short orientation. Typically these methods won’t be highly quantitative, but at least
they can provide you with a first approximation of the condition about which you have no data. Frequently,
watershed groups will partner with a local community college or university to collect selected types of data on
their watershed. In the best-case scenario, you will have funds available to collect the information you need to
fill in the data gaps.

Step 4Analyze the data

This section suggests ways you can move from the raw data you have collected to interpreting its meaning
and importance. You might be data rich, but information poor staring at a bunch of numbers that do not yet
tell a story. This will assist you in making your assessment more complete and accurate. In moving from raw
data to integration and interpretation, you may encounter a few stumbling blocks along the way. Suggestions
to overcome these problems can be found in this section.

I. Summarize and explore the data

Before beginning any formal statistical analysis, you should explore the data informally. This can be done
with descriptive statistics. Descriptive statistics refers to simple calculations that can be done on an Excel
spreadsheet and include calculating the mean or average value, calculating the standard deviation (or range of
variation), and making a frequency distribution if you have sufficient data points. For example, if you’ve
collected water temperature data once a month for three years, you might decide to summarize the data for
each month, based on the value you collected over the three years, by calculating the mean and standard
deviation. You can then construct a graph or table that reflects the average temperature each month.
A frequency distribution plot is another way to look at data variability. If you collected data on temperature
from 15 sites in the watershed in the month of September, you might plot the data to see how similar or
different the sites are.

Figure 4: Frequency distribution of data

Plotting the data in a frequency distribution (see Figure 4) gives you a visual picture of the variability in
temperature throughout the stream. It helps give more meaning to the average.
Overall, descriptive statistics give you a better feel for the data. These simple statistics are sometimes all that
is needed for the watershed assessment, especially if you have a small dataset

II. Perform/Decide if statistical analyses are needed or possible with the data available

Once you have summarized your data, you will need to determine whether it would be useful to perform a
statistical analysis to identify significant changes over time or between different places within the watershed.
Here are some questions to consider to get an overall feel for the data.
• Is the data of sufficient quality to use? Do the data meet appropriate official standards and practices for
collection? Are data collection methods documented adequately so that you can assess their quality?
• Are the gathered data and information useful for your needs?
• Do all the potential users and detractors of the watershed assessment accept the raw data?
• Do all the stakeholders support the choice of analyses?
• Are you thinking in ranges rather than single values for the data?
 • Are you making comparisons to natural variability, which requires determining or estimating baseline
and reference conditions?
• What statistical tests, if any, do you plan to use? Some screening level assessments do not necessarily
require statistics. Also, if the datasets you have collected are limited in scale (either temporally or
spatially) then they might not be suitable for statistical analysis.

You might want to analyze your data using more complex, multivariate methods. These methods permit you
to estimate which factors contribute the most to minimizing variability in the results. In most cases, those
factors that reduce variability in the data are usually the most important regarding meaningful relationship.
Principal components analysis is one method to determine, for example, which stressor out of six might
contribute the most toward the change in habitat that you might have observed.
Another way to approach data analysis is to do spatial analysis or time series analysis. An example of an
analysis over a spatial scale is the measurement of extent of development (e.g., human population or parcel
density) in watershed areas that erode more rapidly than other areas. An example of analysis over a temporal
scale is determining whether changes in water temperature over time are meaningful (or whether they just
reflect natural variability).
The methods used to analyze things that change over space are different from those used for things that
change over time. There is an extensive technical literature on how to measure each of these types of changes,
depending on what needs to be measured (e.g., analysis of trends over time). Two cautionary notes are that
most analyses involve assumptions about the nature of the process being analyzed and that sometimes
analysts have employed inappropriate tools, so copying an approach used elsewhere should be done with
caution. In general, it is wise to consult with someone knowledgeable in statistics to get an informed opinion
and recommendations.
Geographic information systems (GIS) were created to allow calculations for specific places on the earth. If
you have a GIS software program, you can carry out these calculations, too. Examples of common
straightforward analyses are densities of things within a certain area of the landscape (e.g., abandoned mine
density in a sub- watershed), intersection of lines of different types (e.g., roads crossing streams), and
summarizing data for an area (e.g., the number of people in a watershed).

III. Compare your data to standards, historical, and/or reference conditions

Another aspect of analyzing your data is comparing it to standards that are recognized as supporting the
normal functions of biota or watershed processes that you are evaluating. To objectively do this, compare
your data either to that from similar watersheds that are widely considered to have well-functioning processes
and good conditions OR to values for habitat conditions and water quality standards that are known to be
protective for the watershed processes on which your assessment focuses. One critical issue for this analysis
is that for many processes the standards for comparison will vary by bioregion and by habitat type. So don’t
expect one statewide standard to be available or to fit your needs.
The following information might be useful in identifying sources of information for making these
comparisons.

Water Quality Standards that Support Aquatic Life

Sediment Quality Standards that Support Aquatic Life
Habitat Conditions
Step 5: Integrate and report the data to inform decision-making

Once you have collected all the data needed or available to answer your watershed assessment questions, you
face the challenging step of incorporating or integrating all the information into a common analysis.
Information integration here means combining or linking information about various watershed processes and
attributes in a way that leads to conclusions about overall watershed condition and the possible causes. You
could integrate information for particular processes, like the movement of sediment from hill-slopes through
waterways until it is deposited and the impacts of that transport and fate, for example. You could also
combine multiple processes and potential impacts in a system using indicators for potential impacts (e.g., land
use), system stressors (e.g., water temperature), and impacts (e.g., aquatic biota). Without linking individual
processes (or separate disciplines or specialties), watershed assessments may fail to identify potential causes
of the watershed’s condition and important linkages among watershed processes.
In this section, we describe a variety of ways that you can carry out this step, depending on your needs and
available resources. There is no single ‘correct’ way to do this. We give several examples of approaches that
scientists and watershed partnerships have tried. None of them is necessarily right or always usable; they are
listed here to inform you of the range of choices. The methods range from relatively simple conceptual tools
to modeling tools.

The relative condition of watersheds and waterways can be expressed in a variety of ways, but it is commonly
measured using such indicators as drinking water standards, aquatic community composition, terrestrial and
riparian vegetation condition, and constraints on the free flow of water. A majority of watershed or waterway
monitoring and restoration projects are based upon definitions of “health” that are either explicit (e.g., water
quality standards) or implicit (often expressed as deviation from “historical condition”). Any risk or condition
assessment scheme designed to support monitoring or restoration programs should make these watershed
health definitions explicit so that stakeholders understand and support the relevance of the findings or
products of the assessment activities. Making these overall watershed assessments will require the
development of a scheme for integrating the information.

There are many possible ways to integrate information, from qualitative to highly quantitative, from informal
to formal. Many watershed partnerships have a group of experts from different disciplines evaluate
information and form professional opinions about watershed condition(s) and the potentially interrelated
causes of those conditions. Other watershed assessments rely on computer modeling for most of information
processing and then base conclusions on the products of these models. Some assessment programs develop
models that return evaluations of watershed condition as the product.

Models are often helpful in this process. When you developed a picture, or conceptual diagram, of your
watershed’s processes and influences, you were modeling, even if the picture was only in your head. A model
in watershed or environmental assessment is a scaled representation of a system, just as a model boat is a
scaled model of a real boat. The term “model” covers a lot of conceptual and computational territory. You
could model using only mental processes, or you could rely on a physical model intended to represent a
system, such as a watershed.

A model is: A representation of a system; Based on understanding the types and magnitudes of relationships;
Done mentally, visually, or with computers; An aid for evaluation and decision-making; Dependent on the
quality of inputs
A model is not: A replacement for understanding a system; Independent of experts; A substitute for good
science and field work; The answer

With this perspective in mind, there are a number of approaches you can take to analyze your data and
understand the cause-and-effect relationships at work in your watershed. They range from mental team
integration to simple and complex mathematical models.

Regardless of the model or approach used, the overall goal is to identify the link between the adverse effects
and their causes. Being able to attribute a cause to an effect is one of the major values of doing an assessment.
It also can be quite difficult. As noted above, historical, hidden, or multiple factors can be involved as causes
of a problem. Be careful in making assumptions about cause and effect, even when they might seem obvious:
stream bank erosion caused most of the sedimentation, housing development caused more frequent flooding,
or log jams blocked fish passage. Your data might show instead that roads caused most of the sedimentation,
channel aggradation (from sedimentation due to multiple causes) increased flooding, and culverts blocked fish
passage much more often than log jams.

Option 1: Team Mental Integration: Weighing the Evidence

Most watershed assessments involve convening a team of experts from several disciplines to discuss the data
collected and conclusions reached. The team mental integration method is really nothing more than the
assessment team and appropriate experts systematically reviewing the data and, using their best professional
judgment, assessing the impacts of various alterations in the watershed on the ecological endpoints on which
the assessment focuses. In many watersheds, a collection of true experts about the watershed may provide
more detailed and accurate knowledge about influential processes than the best computer model. This may be
partly due to the absence of adequate data, partly due to the lack of a model that truly represents the system,
and partly because expert knowledge is still pretty good compared to modeling. Watershed processes are
complex, and all models contain simplifying assumptions, some of which may preclude investigation of
relevant issues.

On the other hand, the team mental integration approach has certain limitations. There is not a single, widely-
accepted approach for evaluating the weight of the evidence for an assessment. Also, it may be difficult to
ascertain whether team members have sufficient knowledge to thoughtfully interpret the data. If your team
does not have the right qualifications, the insight gained from integrating their knowledge and information
will be limited. Competency is best measured by assessing the amount of formal training in one or more
scientific disciplines, field experience, the amount of watersheds like it, and the ability to see watershed
functioning from more than one perspective.

The suggestions in the following list address some of the potential benefits and pitfalls of the expert team
integration approach:
 Record whatever approach you use in a way that will allow a reader of your assessment, or a future
assessor, to understand exactly what you did. This means describing both the details of the data
considered and the analyses chosen and rejected, as well as providing a summary of the approach
your team took.
 The composition of your team determines the quality of your assessment. Include team members’
qualifications, experience, and training as part of the assessment so readers can assess for themselves
how much confidence to put in the conclusions drawn.
 Comparing professional judgment can be done in various ways, with the most common (and possibly
easiest) being to turn each set of information into rank values, using the criteria for establishing
cause-and-effect relationships.
  Because you will rarely get a group of experts together again to discuss your watershed, take
advantage of the opportunity and make sure they stretch their brains. Encourage them to think about
novel ways that data and knowledge about individual processes can be brought together. Record the
full spectrum of information, from speculation with little data to sturdy conclusions based on a lot of
data, analysis, and expertise.
 Find ways to express professional judgment graphically so people can see what the experts are
thinking. This will help make your analysis understandable to a wider audience.
 Promote diversity in your team by including members from a wide range of disciplines, backgrounds,
ages, and organizational origins. This is bound to lead to critical questions, a range of approaches, and
interesting discussions.

Option 2: Use Statistics

You might wonder how statistics relates to identifying cause-and-effect relationships. Statistics can help
identify associations, the magnitude of differences and other patterns. However, statistics alone cannot
determine causation from observational data. The correlation between two factors does not necessarily mean
that one caused the other. Looking for significant correlations (e.g., regression analysis, r-squared) between
various factors (such as percent of impervious surface vs. peak flows, or population vs. average annual flood)
with available data in your watershed could be performed by following the methods described in user-friendly
books based on watershed research (e.g., Leopold 1994; Gordon et al. 1992; Center for Watershed Protection
1998). However, a sound statistical approach can be difficult to apply in a non-research setting due to lack of
controls and inadequate data, funding, or resources. If the analysis you perform is inconclusive or the
uncertainty too great, reevaluate your procedures and your available data. Quite often, sample sizes are just
too small to provide definitive answers. In such cases, if there is no clear alternative means of analysis, don’t
be afraid to admit that you don’t know or that you are unsure or that there is a lot of uncertainty. It is quite
common to have an indefinite outcome from analysis of environmental data sets where tight experimental
control is not feasible or cost-effective. Don’t let it bother you; just be honest about the limits of the data and
analysis and carefully qualify any conclusions you develop.

Option 3: Relative Risk Model

The relative risk model (RRM) is another method for analyzing watershed data. The RRM is a simple
mathematical method for ranking stressors and altered conditions in a watershed and the likelihood that they
are associated with adverse impacts. It is a useful tool for prioritizing which factors appear to pose the
greatest risk to the ecological endpoints of interest. Most of the process for conducting a relative risk
assessment follows the suggested steps for any type of environmental assessment, including what has been
outlined in this Guide.
The RRM relies on identifying key stressors or altered processes in the conceptual model. Data related to
these factors is compared to data on reference conditions, as previously described. The integrative aspect of
the RRM is that stressors and sources of stress (land use/human activities) can be prioritized based on their
relative rank. To assign ranks:

 Each stressor (altered process or condition) is assigned a rank based on the difference between the
observed value (your watershed data) and the threshold above which an unacceptable effect is likely
to occur.
 Risk is calculated by comparing ranks for all stressors. The assumption here is that stressors with the
highest ranks are more likely than others to be linked to the adverse effects.

The result of this analysis is a ranking of the types of human activities that are likely to be linked to the
harmful impacts and/or a list of stressors likely to be linked to these impacts. This model does not prove cause
and effect, but suggests tenable hypotheses about likely causes and effects. In most cases, follow up studies
are needed to obtain more definitive data. However, the value of using the RRM is that it focuses efforts on
scientifically credible hypotheses that can lead to improved decision-making and management activities.

Option 4: Knowledge-base Models: Ecosystem Management Decision Support

One process for evaluating watershed condition involves using a new modeling approach designed both to
reflect inexact knowledge about natural processes and to be based upon expert knowledge of a system. This
approach is embodied in the software tool “Ecosystem Management Decision-Support (EMDS). EMDS has
been used in the North Coast Watershed Assessment Program (NCWAP) to evaluate restoration potential for
salmon habitat in several North Coast watersheds, to evaluate watershed condition and risk to that condition
in the Yuba River watershed (http://snepmaps.des.ucdavis.edu/snner/yuba/StateYubaLands.pdf) and to
prioritize restoration sites for mercury remediation in the Sacramento River basin
(http://www.sacriver.org/subcommittees/dtmc/documents/DTMC_MSP_App5.pdf).
The EMDS model is a computer-based model that can be used to compare the observed conditions to
reference values for a variety of watershed components and processes in order to assess the present
conditions. EMDS is an integrative approach to assessment, in that it combines data about a place or concern.
Because of this, the product can be one form of watershed condition assessment. At the same time,
imperfections in data knowledge about aspects of the watershed processes and features will be reflected in the
certainty of the assessment. This is true of any model.

Option 5: Assessing Cumulative Watershed Effects

Cumulative watershed effects (CWE) or impacts, refers to two or more individual effects that, when
combined together, make a significant, usually adverse change to some biological population, water quality,
or other valued environmental attributes, or that compound or increase other environmental effects (CEQA
Guidelines, Section 14 CCR 15355). Considering how the effects of human activities may combine to have
greater consequences than the individual effects is central to the watershed approach. Thinking about
processes and impacts in the watershed context usually involves combining individual, seemingly isolated
events. Evaluating CWE typically involves assessing the impacts that might occur in the future as a
consequence of certain human activities or changes in land use. This contrasts with the previously described
methods, which focus on analyzing present conditions that are the consequences of past activities. Following
are a few examples of different approaches you might consider if you want to analyze cumulative watershed
effects.

Step 6: Preparing an assessment Report

A critical component of watershed assessment is describing how you conducted the assessment, what you
found out, and how people can use the information to help them with decision-making.

A. Watershed Assessment Report

The Manual defines a watershed assessment report as: “a report documenting the findings of the watershed
assessment process.” There are several primary components to a well-written report:
• Concise and accurate descriptions
• Use of structural elements like sub-sections, pull-out boxes, appendices and indexing
• Visuals (photos, maps, charts)
• Clear distinctions between how you did something, what you found, and what it means
The report can be published in a variety of ways:
• On a CD with hyperlinks to relevant material on the CD and the Internet
• Online, with links to other online resources
more tangible

With electronic publication, consider associating the watershed assessment report with maps and other data
types on the same Web site. Maps can be served using Internet map server software. Data can be shared as
stand-alone tables for download, or as an online searchable database. Photographs of parts of the watershed or
issues of concern can be linked from a map or from the report itself.

B Report Evaluation

It is a good idea to build product review into your schedule and budget. Decision-makers and others using the
product may have more confidence in it if it has gone through review. The outline or framework for the
assessment, an interim draft report, and the final draft report can all be peer or expert reviewed.
Some evaluation criteria that can be used for this process are:
• The flow from assessment questions through data collection and analysis to findings and
recommendations makes sense and is explicit.
• Data are presented clearly and analysis methods are described.
• Conclusions are based on scientific and statistically valid approaches.

C. Make Recommendations

Many watershed assessments make recommendations for particular restoration, management, policy, or
monitoring actions that could or should be taken in a watershed. We recommend that you keep most action-
oriented recommendations in a watershed plan (e.g., a watershed management plan).
Here are examples of things that the Manual team considers appropriate assessment recommendations:
• How to deal with data or knowledge gaps
• What assessment findings can be linked to elements in a watershed management plan
• How the assessment could inform future watershed decision-making

Here are examples of recommendations that fit well in a watershed plan:

• Certain restoration actions in specific sub-watersheds would benefit watershed function.
• Changing specific land and water management policies and implementation would benefit
watershed function and condition.
• Prioritized actions and places for action
• Description of the relative benefits of carrying out specific management and land/water use actions
for watershed function and condition.

D. Use in Decision Making

A watershed assessment that is not used in decision-making has lost an important function. The type of
decision may range from “more needs to be learned about the system” to “land use designation or pollutant
discharge must be tied to watershed impacts. “ Decisions may be combined as planned actions in a watershed
management plan, or occur separately in different decision-making venues (e.g., local government). There are
many types of decisions that can be informed by watershed assessments; here are a few:
1) Restoration planning, from action at a single site to changes in permitted land-use within a watershed, is
best done in the context of watershed assessment. Natural and human processes will affect the efficacy of the
restoration action and should be taken into account by the restoration planner. In turn, a watershed assessment
intended to support restoration planning should make explicit the connections between watershed processes
and/or sub-watershed condition and potential restoration actions.
2) Regulation of human activities on the landscape or in waterways is a critical part of environmental
management. Regulation of these activities should be informed by watershed assessment when the activities
can cause watershed-wide impacts or originate from large portions of a watershed (e.g., non-point source
pollution). Examples include: permitted discharges from point sources, permitted discharges from diffuse
sources (e.g., under an agricultural discharge waiver), timber harvest plans, housing development planning,
parcel subdivision and zoning, road or highway construction or enlargement, water diversion and storage,
public lands grazing or logging, and channel or floodplain modification.
3) Land use planning is carried out by local agencies in California and affects how and where we impact the
environment. General plans, zoning ordinances, and parcel subdivisions are important land-use decisions that
could be informed by watershed assessment. General plans describe how much new development is desired
and where it will placed in a city or county, and therefore in a watershed. Zoning decisions show what kinds
of development – industrial, commercial, residential, agricultural – are permitted in specific areas and thus in
sub-watersheds. Subdivision of parcels by landowners, which must be approved by local governments, affects
future development patterns (including roads, water delivery, and sewage treatment). Aspects of watershed
assessment, such as water quality analysis, erosion modeling, and habitat degradation, are useful to inform the
where and how much of land-use decisions.

4) Water management is a fundamental driver of condition in many watersheds. Water may be stored,
diverted, or pumped from underground. How much of it is moved around, when it is moved, and where it
ends up can all affect the health of waterways. As California’s population grows and increases pressure on
surface and ground water supplies and as climate change increases the chance of dramatic shifts in weather
and the need for science-based water management increase. Watershed assessment can inform water
management by describing the natural volumes and timing of flows, showing locations of current and
restorable aquatic habitat, and making the links between surface and sub-surface water quality and quantity.
5) Monitoring of watershed conditions should both inform and be informed by watershed assessment.
Monitoring programs can be designed, or modified based on the findings in a watershed assessment.
Monitoring can be allocated to sub-watersheds already under pressure, or at risk of future pressure from
human activities. The type of monitoring occurring (e.g., water quality, aquatic biology, geomorphology)
should be dictated both by what you find in your assessment and what remained as questions about watershed
functioning. The location of sample sites, the sampling frequency, the parameters chosen, and the way the
data is analyzed can all be informed by your watershed assessment.