The Willapa Alliance
Salmonid Habitat Surveys
1995-1996
The Guidebook to
Salmon Habitat Restoration Data
in the
Bear, Palix, Naselle, Nemah
Watersheds
INTRODUCTION
Late in 1996 the Willapa Alliance was contracted through this grant by the Pacific Conservation District to develop, coordinate, and execute the technical components of the project. The Pacific Conservation District provided all of the survey field crew labor and assisted with crew supervision and coordination as part of the larger salmon restoration program which they manage.
A Geographic Information System (GIS) database was created for managing the data collected through the survey. GIS was chosen because it is an extremely powerful data storage, analysis and presentation tool with many applications. Also, many state agencies and local GIS consultants like Mark Scott currently store a wide range of natural resource data for the region in GIS format. This improves the compatibility and functionality between existing databases and the survey data.
Purpose
The primary purpose of the project was to provide the data needed to better understand and characterize the current extent and condition of salmonid habitat in the Willapa Bay Basin. Specifically, the project sought to provide the database needed to address two questions. First, what is the current condition and availability of salmonid habitat in the Willapa Basin? Second, what are the specific salmonid habitat restoration needs in the Basin?
THE SURVEY METHODOLOGY
Given the time and resources available to meet the objectives of the survey, four of the eight major watersheds of the Willapa Basin were selected to be surveyed in this stage of the project. Field surveys were carried out continuously over a one year period between November 1,1995 and November 11,1996.
The primary data collection involved the use of a field survey form and methodology developed by the combined expertise of the Willapa Alliance. The methodology was developed with the assistance of Dr. Peter Bisson of the USFS and personnel from the WDFW. The methodology is based upon a combination of methods used by the WDFW, the WDNR and Northwest Indian Fisheries Commission Stream Typing Program. The methodology is substantially different in nature and content than those of the reference formats and was specifically tailored to meet the goals and needs of the program. It was designed to address the two primary questions which the survey was asked to answer. It was also designed with the characteristics of the survey crew in mind, individuals who have recently been displaced from the salmon fishing industry due to the decline in salmon populations. These individuals generally have significant experience in field natural resource work, but little to no formal training in the biological sciences or in habitat survey methods.
The field survey entailed walking stream channels and recording channel characteristics, riparian zone characteristics, and habitat and other features found within 200 feet on either side of the stream. Survey data are sequentially recorded on several mediums to hold the data, each intended for use with a variety of future applications. A field methodology incorporating the use of a specialized data collection form provides a template and a way of standardizing the interpretation of a variety of in-stream and riparian features. The methodology is designed to provide a mechanism for increasing the level of objectivity in the interpretation of field data.
Extent of the Survey
The extent of stream survey coverage has been limited to areas presently accessible by anadromous fish. Stream surveyors walked up stream channels to the points where a natural obstruction to anadromous fish existed including increases in gradients beyond 20%, waterfalls, and channel widths less than 2 feet. Surveys were continued above barriers created by human activities and barriers which are not completely impassable. The full extent of any one watershed within the Willapa Bay system has not been surveyed primarily due to stream channel obstructions that are impasses to anadromous fish. The survey covers 223 linear miles of mainstem stream channels and primary tributaries in the Naselle, Palix, Nemah and Bear river watersheds. A total of 30 miles was surveyed in the Palix Watershed, 73 miles in the Nemah Watershed, 88 miles in the Naselle Watershed, and 32 miles in the Bear Watershed.
As discussed earlier, the stream survey did not cover several watersheds in the Willapa Basin for a variety of reasons. Watersheds which were judged to be most important relatively for producing salmonids, watersheds in which data on habitat conditions were most lacking, and watersheds which could be surveyed in the available time frame were prioritized for the survey. The North watershed and the Willapa watershed remain un-surveyed because of the limitations of the resources available to carry out the project. Additionally, some minor systems and tributary stream channels were not surveyed within the watersheds covered, in particular in the Naselle drainage. Many tributaries often extend beyond blockages to anadromous fish and were therefore also not surveyed. Abrupt increases in channel gradients limits the extent of nearly all surveyed tributary channels. These un-surveyed stream reaches may have significant importance due to their distribution and potential for contribution to ecological productivity.
Data Collection
Field work was carried out continuously throughout the year by teams of Pacific Conservation District stream surveyors. To collect stream data, surveyors entered the stream channel to be surveyed and followed the channel continuously until reaching the end of the survey. The surveyors obtained direct measurements of stream channel features while simultaneously recording the features distance from the beginning of the survey. An interpretive survey guide contains instructions on how to make distinctions between categories of observed features. A four page standardized form is used which has spaces to input observed responses (See Appendix H Survey Form).
Most data are recorded at 200 foot increments along the stream, which are then aggregated into 1000 foot summations. Segments are the basic linear components that the stream data are expressed in and directly correspond to 1000 linear feet sections of stream channel and riparian zone. Segments are connected to represent the extent of hydrographic system covered by the survey. Segments form linear spatial components that connect to form a spatial network of mainstem and tributary river channels. A measuring device called a hip chain was employed to establish field measured locations. Hip chain measurement were used to establish the location of individual stream segment locations relative to the beginning of the survey which is at a known point. Hip chain readings were also used to keep track of the exact location of certain features encountered on the survey including confluence points with tributary stream channels, potential passage barriers and other fixed point features. Compasses were used to determine stream aspect. 100 foot measuring tapes were used to calibrate the estimation of stream feature estimates such as width and depth. Clinometers were also used to calibrate the estimated measurements of stream gradient. Estimation was employed in concert with direct measurement once surveyors achieved a reasonable level of calibration. Calibration was checked regularly by the surveyors as they collected data.
Navigation while in the stream was accomplished using the United States Geological Survey, 1:24000 Quadrangle, 7.5 Minute Series maps. Stream segments to be surveyed were marked in the office by the survey coordinator. Survey crews used these maps, hip chains, compasses, and other supplementary maps to navigate. Upon completion of a survey, field crews roughly mapped out the extent of their survey on the base maps and submitted the maps with their data sheets for processing.
Data Resolution
Resolution is a characteristic that can describe the level of detail inherent to data contained in this survey. The degree of accuracy of the data in this database is limited by the scale at which the data was collected and is represented. Resolution can be expressed in units of linear feet, and/or square feet. All of the stream data contained within this set has been aggregated to 1000 foot linear units of resolution. This means that the overall continuity of the data is no finer than 1000 feet along any stream channel. Stream channel features contained in this data are measured in no units smaller than inches. In this case, the level of detail is high enough so that the data has the capacity to remain adaptable to a variety of future applications. All stream survey data in this set can be compared to most other data at a variety of scales common to GIS and GPS applications given its relatively high level of resolution. The primary factor limiting the resolution of this data is the base map on which the data is projected.
Replicability of the Data
This stream survey data set contains static measurements of dynamic geomorphologic and biological features. Data obtained in the survey represent interpretations of measured features. Measurements taken are either estimated or physically measured. The data are representative of the conditions and circumstances at the time of collection only. Recorded values in this data set are subjected to the parameters established by the dynamics of the natural and developed landscape and are episodically subject to change. The same data could never be obtained again as they were under the conditions and circumstances at the time of the original survey. A one time only stream survey cannot assess the long term variability inherent in stream systems caused by seasonal changes and the dynamic nature of streams. A survey of the same stream segments at a later date with the same methods would add to the range of known variability of the features found along those surveyed sections of stream channel. Reproduction of the data at a different time would also provide a basis for the measurement of habitat restoration and improvement.
THE DATABASE
Data Set Phases
There are three sequential forms of the stream survey data that result from post stream survey data processing. The data in this database has been transformed into Phase III data. The Phase I database is a verbatim copy of the original field notes with subtle paper to database transformation changes completed as needed. Phase I database is checked and corrected only for completeness. The Phase II database is a result of establishing a GIS link with the Phase I data and the completion of additional quality control measures. Data quality control for Phase II revolves around two aspects, spatial accuracy and content accuracy. Phase II data has been checked for anomalous record values and field checked and corrected for both data content and spatial accuracy. Appendix E describe this correction process in greater detail.
The Phase III database includes the data (spatial and content) enhancements from Phase II and goes one step further. Stream data are overlain with other spatial data layers to extend the database beyond the area physically surveyed. This further enhanced data set has the capacity to include combinations of other spatial data sets. This process enables the original stream survey data to be compared with other spatial data such as, soil types, geography and geology, land ownership, zoning, land covers, and other associated data and imagery.
The Phase I Data Set
The stream survey data was originally written on paper field survey forms for later reference, analysis and storage. In this form, the data can be exchanged or distributed by photo copies of the originals. A paper copy storage system was established which contains stream survey forms grouped by the individual watershed. The paper forms are further differentiated into main channels and tributaries of the main channel. The primary benefit of maintaining the original field notes is that the highest level of detail is preserved in this format. Transferring the original paper data to the computer database required abbreviations to conserve space on the data base version (see Appendix B). Original field notes relay a more complete sense of the interpreters approach and perspective on the day the survey was carried out. Notes commonly refer to the many obstacles encountered including weather and terrain. Events during the day are also captured and preserved directly into the pages of the notes themselves.
Digital Data Base Storage System
The data base file is a computer storage system that contains the same data as the original stream survey documents. The software Dbase was used to build the survey database. A digital file takes the original data a step further in user accessibility and analysis capabilities than the original paper counterpart. The data base format allows all of the features described in the original field survey forms to be quickly searched with user defined parameters. Parameters used in the data base are restricted by the extent and resolution of the data. The computer can sort and search for more than one variable at a time. Results from database tabulations can be graphically charted for ease of illustration or further processed with statistical methods.
The survey database is a two dimensional grid that contains places to store data. This grid is actually made up of two components: fields on the horizontal axis (at the top); and records along the vertical axis. The grids size along the horizontal direction is controlled by a database file structure that contains the fields (Appendix A). Fields are chosen to represent and contain specific values to responses that are on the original field notes. They vary in width and type. For example, a stream name contains characters that form words like BEAR. The field type in this case should be character and should be wide enough to hold all of the letters in BEAR. A numeric field type can contain numeric values to responses found on the original stream survey form. For example, the width of a stream channel is expressed in feet. Rather than spelling out 'Thirty Five and a Half Feet' the numeric field can accept a value of 35.5 to express the width of a stream.
The Process Used to Create the (Stream Habitat Data) Digital Database
1)
The first step involved in creating the data base was to
generate a data file structure (see Appendix A) that is based on the original
field survey form. The data file structure created was then used to hold
data from responses given on the original data form. The final data file
structure is 746 spaces wide and contains a combined total of 1186 river
segment records.
2)
A unique ID numbering system was created and each segment
was designated its’ own number.
3)
The contents of the original field notes used in the survey
were entered into the data file structure.
4)
The resulting data base collection of individual data files
were assembled into four main data files; Naselle, Nemah, Bear and Palix.
5) Data files were checked for error.
Segment identification numbers
One critical step in the development of the data base was the establishment of the unique identification numbering system used to label individual stream survey segments. Each survey sheet and corresponding stream segment is designated one unique code. A seven digit coding system has been applied to all stream survey field notes in this stream survey. This code serves as the address for each segment.
The following is an explanation of the identification numbers from left to right:
(1-2) The first two digits are character abbreviations for the stream name (See Appendix D).
(3-4) The third and fourth places are the field segment number designations for the main stem river segments (1-99) in ascending order in increments of one except the mainstem of the Naselle that exceeded 100 segments and required the use of an extra character.
(5-7) The last three positions give identification to tributary stream channels.
The Utility of the Unique Identification Number
Unique coding of individual river segments allows for increased ease of maneuverability with respect to data query and is required for use by a GIS. This coding system makes use of nested codes that can reveal characteristics about the data through the identification number itself. Combinations of river segment selections based on the available attributes contained within the stream survey data base allow the river segments to be identified without the use of a GIS.
The unique identification number makes use of a configuration where main channel record numbers can be easily distinguished from the tributary record numbers. A sequence of three zeros in the last three positions indicates the segment record number is a mainstem channel segment. Mainstem coding can also relay location up the mainstem relative to the origin of the survey. These data are contained in the third and fourth positions of the segment identification number and relays the distance up the segmented river network in thousands of feet from the original starting point of the survey. Tributary channel identification numbers also carry mainstem segment numbers with them. This configuration allows for quick determination of the location of the confluence of the tributary with the mainstem. The “ID” number also gives reference to the relative sequence of segments surveyed and the location of tributaries.
The Data Entry Process
The original stream data were manually entered into the database using PC based ARC/INFO software at the Willapa Alliance. Stream surveyors assisted with the data entry and provided crucial insight because of their intimate familiarity with the hidden details and other elements of the original data. The data were entered as objectively as possible with the intent of fully and completely relaying the contents of the original field notes into a digital database file.
Quality Control for the Phase I Data
Quality control for the transfer of data into the database from the original notes was established in a number of ways. The first level of quality control involves the content accuracy of the database file. The use of the database management software provided the capability to sort through records in a way that revealed typographical errors. Most errors occurred primarily as a result of human error in data entry. For each type of error, a subsequent action was taken to fix the problem.
Adjustments to the original data
In order to provide continuity amongst numeric values in the data set, calculated fields were added to contain values that are intended for relational purposes. Average substrate types, average gradient etc..
The Phase II Data Set
In the Phase II database a spatial component for each record (river segment) in the Phase I database is added. The spatial data layer then is no longer limited to being a record in a two dimensional database, but can be presented graphically to express the spatial element of the data.
The spatial data layer is nothing more than a graphical representation of the same distribution of streams surveyed in the field. Each stream channel is represented as lines in a digital format that contains the same stream data from the original stream survey data forms. The GIS is dependent upon the original data that is organized with the previously described unique identification numbers. These components are all merged into a layer that contains spatial features.
The power of the GIS database comes from the ability to utilize computer based computations to consider all of the individual parameters of the stream database set simultaneously in an infinitely accurate manner. It is also powerful in that the GIS can consider relationships between the designation, size, and distribution of individual features contained within spatial layers represented by their individual spatial coordinates. A unique and interactive perspective is therefore made possible by the use of GIS. GIS carries the ability to relay to the user of the system, the spatial analysis of the full extent of all surveyed streams contained within layers.
The Process Used to Create the Stream Survey Spatial Data Layers
1) ID numbers were established for each surveyed stream segment and written to original field documents. These are the same ID numbers used in the stream survey data base.
2) Survey segments were overlain on U.S.G.S. 7.5 minute quadrangles. Segments were approximately located according to descriptions and foot distances from known locations in field notes. 7.5 minute quadrangles were digitized to create spatial layers that represent the extent of the streams surveyed.
3) Line work (the representation of line features such as streams and roads) in the WDNR hydrographic layer was applied and edited to extract the extent of the streams that were surveyed. This line work was stripped off the original WDNR attributes.
4) Digitized stream layers and extracted WDNR hydrography were edited together to transfer surveyed stream segments to the WDNR layer. The new layer was edited several times to ensure that survey segments were sequenced and located accurately on this layer.
5)
The spatial files version of the stream survey database
(arc attribute table) and the stream survey data files were linked by the
established unique “ID” number that appears in both sets of files and is written
on the original stream survey notes.
6) Four spatial data layers were produced that each represent the extent of the survey in each of the four watersheds covered.
7)
Data layers were checked for spatial and content errors
as part of the creation process. Spatial errors were corrected by using GPS. Content errors
were identified and checked with field visits. (See Appendix E)
Origin and Scale Limitations of the Stream Survey Spatial Data Layers
The original source of the hydrographic spatial data layer used for this project was the WDNR which was made available by Interrain Pacific of Portland, Oregon. This spatial stream layer was originally created at a scale of 1:24000 and contains the most complete set of digital linework available for the region. All original attributes were deleted to make room for attributes specific to this stream survey data set. The linework contained within the WDNR layer was edited to fix inherent errors in the layer. For example, edge matching from tiled hydrography layers was not completely processed. The original WDNR hydrography layer contained mis-matched and/or missing stream segments before it was edited for use in this project. In some cases lines were missing and had to be joined to provide continuity in the line network. Linework contained within the final stream data layers has been corrected from the original WDNR hydrographic layers.
The level of interaction between different features is dependent on the scale at which the data are collected and represented. Scale differs from resolution in that scale proportions the spatial data for illustration purposes. Baseline spatial data sets used for this survey are intended for use at a maximum scale of 1:24000. At this scale, individual river segments can be compared to other data of approximately the same scale. The 1:24000 scale is large enough to illustrate the data contained within each segment.
Projection of the Stream Survey Spatial Data Layers
The level of detail and location accuracy is an important component of GIS. Location and spatial accuracy with respect to perspective is a major concern for GIS databases. The round world is often illustrated with flat maps. A spherical object like the earth can be conformed to a flat map by means of projection. This is a way to transfer the image of the round earth onto a flat map.
Many GIS layers use a projection system called Universal Transverse Mercator (UTM) to keep track of spatial data layers. This database uses the UTM coordinate system. The coordinate systems used in these databases contain measurements in units of meters. The spatial data layers can be re-projected into a different projection that may use units of feet, making the integration of different data sources much easier. The stream survey spatial data set was projected in the Washington State Plane South projection system with all river segments contained within measured units of meters. The state plane projection system produces spatial coordinates in State Plane distances from it's origin. This projection system is also commonly used at city and county levels as a measure of location. The stream survey spatial layers also use the North American Datum (NAD) 1927. Any layers used for comparison should match the above parameters.
Quality Control for the Phase II Data
Spatial accuracy is related to the exactness at which stream channel segment locations contained within the spatial data set are represented. The spatial data set has two possible sources of spatial error. One error effects the river channel segments internally with respect to length. The other is the relative location of adjacent segments along extent of the surveyed river channels. Both errors were introduced to the system during the field data collection process.
The first kind of error occurs as a result of the method of segment measurement in the field. Ideally, the segments are 1000 feet in length on the ground. Segments are field measured with hip chain string. The string can stretch so the segment can potentially be longer than expected. The string can get wrapped around obstacles which can have the effect of shortening the segment. Since there was no way of capturing the exact path the stream surveyors took in the original survey, the survey path was fit to an established pathway that exists in the linework of the 1:24000 scale WDNR hydrographic layer. The lines, tributary intersections and other reference points in the original WDNR spatial data layer representing the surveyed stream segments did not follow the exact path which the surveyors took, but was very close. It can be assumed that stream segments in the stream survey spatial data layers are not exactly 1000 feet in length, but each represent 1000 feet of surveyed stream fit into the proper reference points. It is also very difficult to get exact field measurements of the segments because of sinuosity of the stream. Even if hipchain and river channel sinuosity were a direct match, the data would not match the third sinuosity introduced into the equation by the 1:24000 scale WDNR spatial base layer.
The original stream survey data was fit to the 1:24000 scale WDNR hydrographic layers. In order to rely upon a spatial data layer of river survey channel segments, there must be as many references to the location of the beginning and ending points of segments as possible. In this case, there were no field recorded coordinates to generate an exact location for the segments even though segments were flagged in the field. Spatial data sets for this project were tested for spatial accuracy in two separate stages.
The first involved identifying all the surveyed streams and laying out all the surveyed segments on a U.S.G.S. 1:24,000 quadrangle. The original field notes were thoroughly read for details concerning the location of reference points on the map. Segments were hand drawn and fitted onto the base map. This step helped determine the relationship between all stream segments and their true locations. This was also an aid to stream segment identification and finding gaps and overlaps in the survey. A number of fitting problems associated with the arrangement of stream channel segments were encountered. A lack of location detail to the original field notes in some cases lead to a larger location error. One key to fitting the segments was the description of measured distances to bridges and tributary intersections within the segment. In some cases segments had no written location references. Also the distance into the river system can be unreliable due to a varying degree of sinuosity.
In the second stage, both types of errors were minimized by close interpretation of stream survey notes and the implementation of GPS ground truthing. GPS employs hand held electronic instrumentation connected to orbiting satellites to provide exact location data. A GPS was used in this project to aid in the determination of the location of stream channel segments using real world coordinates to verify survey segment start and end points. Suspect river channel segments were field corrected. A list of spatially corrected stream survey segments appears in appendix F.
In addition to the spatial correction, content error checking and corrections were made to create the Phase II database. Data content suspected of being inaccurate based upon professional judgment was systematically searched for in the database and identified. In addition, some data fields were assessed for errors by sorting through the entire database with an automated tool. The six calculated fields contained in the survey data set were checked by calculating values of the known constituents of the calculated field. Calculated fields in this data set are typically fields that contain averages based on other fields. For example, the “substrate type” by percentage contains six calculated fields that represent an average of substrate type over five 200 foot sections of a segment. The substrate type calculated fields are broken into representative average boulder, cobble, gravel, sand, sediment and bedrock for each 1000 feet. Since the substrate type was recorded every 200, there must be an average calculated for the total 1000 foot segment. Another type of content error encountered was incomplete data entry for fields in the data sheet that required a response for every record. Also, in some cases repetitive responses to the field forms were skipped by the stream surveyors, and subsequently not added to the database in the phase I version of the data.
Data correction was completed by making field visits to verify the data as needed. A complete log of data checking and corrections is presented in Appendix E.
Upon completion of the final data correction steps, a random sampling of 120 records (10%) was performed to check for the occurrence of all content and spatial errors. The sampling showed 100% accuracy at a 10% sample size.
Phase III Applications
GIS Analysis
Analysis of spatial data is one of the primary applications for a GIS. Spatial analysis frequently incorporates the overlay methods used in Arc/Info because of the ease of relatability between spatial data layers. Once an overlay has created a new data set, questions can be formulated to query the new data layer to obtain results that otherwise could not be done. The overlay process allows point line and polygon features from other thematic layers to be combined into one single layer. For example, a watershed (polygon) layer can be overlaid (combined) with a stream (line) layer in such a way that the total lengths of streams can be tallied per watershed. Spatial analysis involves selecting and displaying images to making maps and data files of real world features. This is done by selecting feature attributes. Once feature attributes are selected, there is a pattern and color that can be assigned to the selected feature for display purposes on the computer screen and maps.
Combining stream survey data with other spatial data layers provides added information to stream survey attribute tables. Phase III data tables contains added peripheral information such as land ownership, land use and adjacent vegetation types. These elements are incorporated into phase III stream survey data layers . Nothing is added or deleted from the original extent of linework, with the exception of new attributes. These new fields added for phase III data make graphical buffers possible, include, for example; width to depth ratio is based on units of feet with respect to depth and width. The resulting polygon layer simply indicates the relative proportion between numeric values.
There are literally thousands of combinations of data layers that can occupy the same extent as the stream survey linework and share attributes. There is a limiting factor to adding thousands of attributes, and that is a software limitation of the number of fields that can be in a data set. There are trade off between speed of query and content. One way around this is a look up table that contains amenable data in small pieces that can be joined on as needed.
Phase III Graphical Buffers:
Graphical buffers used in stream habitat mapping provide a new way of illustrating complex data sets. Buffering is a Arc-Info process that delineates a polygon from line work. Polygon widths correspond to numeric values contained within the data set. The wider the line, the higher the numeric value contained within the data set. Graphical buffers are created from individual fields contained within the data set. There are now over 30 graphical buffer layers for each line coverage created from the phase III data set that represent the relative proportion of data for the extent of each of the four watersheds that were surveyed.
Graphical buffers are representations of the original stream survey data set that have been averaged over a 1000’ stream survey segment (as described in phase II). Discontinuity within buffers is a result of null data within a field, for example a riparian forest will periodically disappear when the stream survey goes through a meadow. In many cases data had to be averaged by combining data from right to left sides of a stream channel. Buffers based on field averages not found in phase II are contained in the phase III data set.
The nature of graphical buffers is such that line widths can in some cases overwrite on to adjacent stream segments that contains lesser value. Buffer widths for fields with high numeric values commonly overprint onto adjacent buffered areas, this is especially true when it comes to large woody debris. Buffers exhibiting these characteristics can draw attention by being larger others.
Combinations of these layers produce a continuous representation of data throughout the survey area. Note that trade off in dominance between graphical layers are a common occurrence. If continuity is preserved with respect to line width, (for example individual substrate types adding up to 100% then ideally all combinations of substrate buffer types should remain continuously as wide as any other, that is not the case, since some substrate types predominate over others and get effectively covered by lesser dominant layers. This phenomena creates a situation whereby buffer widths do not remain continuous, and should not necessarily remain continuous by the dynamic nature of the data.
Some of the individual graphical buffer (phase III) coverages are as follows:
1) General Substrate Types
Boulder
Cobble
Gravel
Sand
Bedrock
Sediment
2) General Forest Types
Average Riparian Age (left)
Average Riparian Age (right)
Average Riparian DBH (right)
Average Riparian DBH (left)
Average Riparian Height (right)
Average Riparian Height (left)
Deciduous Riparian Forest Type (left)
Coniferous Riparian Forest Type (right)
Average overhead canopy cover
3) In Channel Features
Coniferous woody Debris GT 12”
Deciduous woody Debris GT 12”
Average number of clumps (in channel)
Average number of root wads (in channel)
4) General Flow Types
Average Glide
Average Pool
Average Riffle
5) Miscellaneous
Average Gradient
Ordinary high water depth
Ordinary high water width
Average wetted width
Average wetted depth
Average width to depth ratio
Basic Applications
The database provides the data needed to complete basic salmonid habitat analysis to answer the two primary questions which the survey was designed to assist in answering. First, basic mathematical analysis of this data can be used to produce estimates of salmonid habitat extent by location. With this analysis, carrying capacity estimates or fish production estimates can be completed. The analysis process can be greatly facilitated through the application of GIS based tools.
Secondly, the database can also be used to identify and prioritize specific and general habitat management needs, specifically restoration. Criteria to gauge habitat restoration need can be developed and then used to direct a query of the database to identify specific locations that may benefit from restoration. These identified locations may then be further investigated in the field to determine what restoration actions may be most appropriate. Potential restoration sites may then also be compared against one another to prioritize treatments.
Appendix A: Stream Survey Data Structure
Id: Unique identifier stream for stream segmentation.
Stream: Name of stream surveyed.
Tributary: Name of water body that surveyed stream flows into.
County: Name of the county that the stream segment falls within.
Surveyors: (See Appendix C) Initials of the stream surveyors who surveyed the segment.
Date: Date survey was completed.
Mouth_loc: Location by township, range, and section of the downstream confluence point of the stream.
Start_loc: Nearest landmark to the start of the survey.
Seg_num: Sequential number assigned to each segment surveyed that corresponds to tot_seg_num. Segments are 1000' in length.
Tot_seg_num: Total number of surveyed segments per stream.
Rain_amt: Amount of rainfall in inches that occurred before the survey.
Rain_date: Date of last rainfall.
Tidal_inf: (yes/no) influenced by tidal activity.
Assoc_wet: (yes/no) Associated wetlands to the stream segment.
Dike_wet: (yes/no) Man-made structure that separates stream from the wetlands.
Comp(1-5): Compass readings at 200 foot increments along the segment.
OHWD: Ordinary high water depth in units of feet/inches. This field contains the average wetted depth for the 1000’ segment.
OHWW: Ordinary high water width in units of feet. This field contains the average wetted width for the 1000’ segment.
WD: Wetted depth of the stream channel at the time of the survey. This field contains the average depth for the 1000’ segment at the time of the survey.
WW: Wetted width at the time of the survey. This field contains the average width of the 1000’ segment at the time of the survey.
OHWD(1-5): Ordinary high water depth of individual 200 foot increments.
OHWW(1-5): Ordinary high water width of individual 200 foot increments.
WD(1-5): Wetted depth of individual 200 foot increments, at the time of survey.
WW(1-5): Wetted width of individual 200 foot increments, at the time of survey.
Off_channe: Total number of off channel areas per 1000 foot segment.
Grad(1-5): Stream channel gradient, averaged over 200 foot segments.
PCT_POOL: Linear percentage of (pool) flow regime observed within the segment.
PCT_RIFF: Linear percentage of (riffle) flow regime observed within the segment.
PCT_GLID: Linear percentage of (glide) flow regime observed within the segment.
AVG_BOULDE: Average percentage of the substrate comprised of boulder sized material for the segment.
AVG_COBBLE: Average percentage of the substrate comprised of cobble sized material for the segment.
AVG_GRAVEL: Average percentage of the substrate comprised of gravel sized material for the segment.
AVG_SAND: Average percentage of the substrate comprised of sand sized material for the segment.
AVG_SED: Average percentage of the substrate comprised of sediment sized material for the segment.
AVG_BED: Average percentage of the substrate comprised of bedrock material for the segment.
BOULD(1-5): Average boulder substrate for individual 200 foot increments.
COBBLE(1-5): Average cobble substrate for individual 200 foot increments.
GRAVEL(1-5): Average gravel substrate for individual 200 foot increments.
SAND(1-5): Average sand substrate for individual 200 foot increments.
SED(1-5): Average sediment substrate for individual 200 foot increments.
BED(1-5) Average bedrock substrate for individual 200 foot increments.
SED_SOURCE: Individual sediment source adjacent to stream channel.
SED_SEV: Severity of impact sediment source has on adjacent stream channel.
SED_LOC: Number of feet into the segment that the sediment source can be found at the time of survey.
SED_SOUR(2-4): Additional space for sediment source per segment.
SED_SEV(2-4): Additional space for sediment severity per segment.
SED_LOC(2-4): Additional space for the location right/left in feet from the beginning of the segment.
OHC_PCT: Overhead canopy cover percent averaged over the segment.
CON_L: Total percentage of the forest found within 200’ of the stream channel comprised of coniferous trees, on the left side of the segment.
CL_AGE: Field estimated age of coniferous trees within 200 feet of the stream channel on the left side for the segment.
CL_DBH: Field estimated diameter breast height of coniferous trees within 200 feet of the stream channel on the left side for the segment.
CL_HGT: Field estimated height of coniferous trees found within 200 feet of the stream channel on the left side for the segment.
CL_COV: Field estimated density of vegetation types within 200 feet of the stream channel on the left side for the segment.
DEC_L: Percentage of the forest found within 200’ of the stream channel comprised of deciduous trees, on the left side for the segment.
DL_AGE: Field estimated age of deciduous trees found within 200 feet of the stream channel on the left side for the segment.
DL_DBH: Field estimated breast diameter height of deciduous trees found within 200 feet of the stream channel on the left side for the segment.
DL_HGT: Field estimated height of deciduous trees found within 200 feet of the stream channel on the left side for the segment.
DL_COV: Field estimated density of conifers within 200 feet of the stream channel on the left side for the segment.
SHRUB_L: Density of shrubs within 200 feet of the stream channel on the left side of the stream.
SHRUB_L_DOM: (Yes/No) Is the shrub layer the dominant vegetation on the left side of the stream channel for the segment?
CON_R: Total percentage of the forest found within 200’ of the stream channel comprised of coniferous trees, on the right side of the segment.
CR_AGE: Field estimated age of coniferous trees found within 200 feet of the right side of the stream channel for the segment.
CR_DBH: Field estimated diameter breast height of coniferous trees found within 200 feet of the right side of the stream channel for the segment.
CR_HGT: Field estimated height of coniferous trees found within 200 feet of the right side of the stream for the segment.
CR_COV: Field estimated density of vegetation types within 200 feet of the stream channel on the right side for the segment.
DEC_R:: Percentage of the forest found within 200’ of the stream channel comprised of deciduous trees, on the right side for the segment.
DR_AGE: Field estimated age of deciduous trees found within 200 feet of the right side of the stream channel for the segment.
DR_DBH: Field estimated diameter breast height of deciduous trees found within 200 feet of the right side of the stream channel for the segment.
DR_HGT: Field estimated height of deciduous trees found within 200 feet of the stream channel for the segment.
DR_COV: Field estimated density of vegetation types within 200 feet of the stream channel on the right side for the segment.
SHRUB_R: Density of shrubs found within 200 feet of the stream channel on the right side of the stream
SHRUB_R_DOM: (yes/no) Is the shrub layer the dominant vegetation on the right side of the stream for the segment?
CONGT12: Total number of pieces of coniferous woody debris in the stream channel that are greater than 12 inches in diameter.
DECGT12: Total number of pieces of deciduous woody debris in the stream channel that are greater than 12 inches in diameter.
CONLT12: Total number of pieces of coniferous woody debris in the stream channel that are less than 12 inches in diameter.
DECLT12: Total number of pieces of deciduous woody debris in the stream channel that are less than 12 inches in diameter for the segment.
NUM_ROOT: Total number of instream root wads/plates..
NUM_CLUMP: Total number of instream individual woody debris clumps.
LAND_USE: Average general use of the land for the segment. (F=Forest, A=Agriculture, S=Suburban, U=Urban)
NUM_DIV: Total number of stream channel diversions that occur within the segment.
POLLUTION: Type of pollution sources that occur within the segment. (A=Animal waste, I=Irrigation Runoff, D=Industrial runoff, D=Industrial Discharges, S=Sewage)
ROAD_X: Number and type of road crossings in the segment stream channel.
ROAD_DIS: Field estimated distance to nearest road from the stream channel for the segment.
WAT_CLAR: Observed water clarity at the time of survey (CL=clear, MOD=moderate, CD=cloudy).
TYPE(2-5): Type of existing/potential barrier to fish within the segment. Types include: beaver dams; man-made dams/reservoirs; water falls; culverts; log/debris jams.
PASS(2-5): Field estimated passibility of fish through the type of stream channel barrier (P=passable, I=impassable, U=uncertain).
Height(2-5): Field measured height of type of stream channel barrier. The height is measured according to the type of stream channel barrier. i.e., culverts use bottom lip.
DIAMETER: Field measured diameter of culvert type of stream channel barrier. There is no other application of diameter to any other type of stream channel barrier.
LENGTH(2-5): Field measured length of the type of stream channel barrier.
GRADIENT(2-5): Field measured gradient of either falls or culvert type of stream channel barrier.
LOCATION(2-5); Field estimated location in feet from the beginning of the segment, related to the type of stream channel barrier.
SPECIES(2-5): Field identified name of the species of fish observed within the segment. Fish are placed into the categories of: Coho; Chinook; Chum; Steelhead; Cutthroat; Sculpins; and Unknown.
ADULT_LIVE, [A_L(2-5)]: Total number of field observed live adult species of fish found within the segment.
ADULT_DEAD, [A_D(2-5)]: Total number of field observed dead adult species of fish found within the segment.
JUV_LIVE, [J_L(2-5)]: Total number of field observed alive juvenile species of fish found within the segment.
JUV_DEAD, [J_D(2-5)]: Total number of field observed dead juvenile species of fish found within the segment.
REDDS,(2-5): Total number of field observed redds related to the particular species of fish found within the segment.
COMMENTS: A place where additional descriptive details about the segment is abbreviated and written. Comments relay useful reference points, tributary names and features found within the segment. Comments are abbreviated from their original form in field notes.
Appendix B: Abbreviations Used in Comments:
CB=car body
CP=Crab Pot
TL=Tributary Left
TR=Tributary Right
COHO=Coho
CHIN=Chinook
CUTT=Cutthroat Trout
STHD=Steelhead
RMZ=Riparian Management Zone
CCUT=Clearcut
CL=Clear
CD=Cloudy
Massfailu (mf)=Mass failure
BDam=Beaver Dam
@=at
GLBT=
Den=dense
Int=Intermediate
Spa=Sparse
BR=Bridge
Cobl=Cobble
STRVEG=Stream veg
DNS=Did Not Survey
SVEG=Streamside Vegetation
WD=Woody Debris
Past=Pasture
Past(L&R)=Pasture on left or pasture on right
PAC=Pacific County
C=Culvert
RD=Road
Appendix C: Stream Surveyors:
EJ=Erik Johnson
MH=Matt Harder
JR=Jeff Rudolph
MD=Mark Drage
TM=Tim Moore
DM=Dave Maki
BL=Bob Lake
Appendix D: Stream Names with Abbreviations:
Bear River=BR
Cannon River=CR
****Canyon
South Palix=SP
Middle Palix=MP
North Palix=NP
North Nemah=NN
****Finn Creek=NN (tributary)
****North Fork (Williams Creek)
Clear Water Creek=CC
Seal Slough=SS
Green Head Slough=GH
****Naselle
Appendix E: Suspect Stream Channel Segments That Were Field Checked
CR08000- Segment 8 no compass plus channel characteristics.
CR35000- CR40000 incomplete surveys.
CR 42000- Age, DBH, and height of conifers average 20 yr. ,10 DBH, 100' height.
Correction: Actual 40 yr., 70' height.
SOUTH PALIX
Tributary Extensions- Township Range 13N/10W section 35 SE, Washington Stream Catalog Trib .0431( T2R ) ends at segment 6. Looks passable, extend survey. 13N/10W section 26 (SF Palix) extend survey above segment 19, fish presence @ B-500 RD. 13N/10W section 26 Stream catalog Trib .0430,(Trib to S.F. Palix) extend above previous end @ segment 4.
WILLIAM CREEK
WM02000- Stream vegetation (SVEG) on both sides are conifers listed as 60 yr. old , 20 DBH, 160' height.
Correction: Actual 18 DBH, 110' height. WM05000- SVEG conifer listed as 60 yr. old, 15 DBH, 170' height.
Correction: Actual 18 DBH, 110' height.
WM07000- Check land use for area ( Residential).
WM09000- Check % glides throughout segment( 60% riffle, 30% glide, 10% pool).
WM20030- Trib 2L no count on Instream structure, comments say too numerous. Field reviewed this stream, coho spawning above first bridge crossing.
Correction: Extend survey.
North Fork Williams- Ended at WM 31040.
Correction: extend to at least next road-xing.
NN48050- Incomplete.
NN71020- Trib 12L right side conifer 40 yrs.,15 DBH, 170' height.
Correction: Actual 80' height.
NN12030- No channel characteristics.
Correction: Reviewed on 11/18/96, measurements OHWD 1', OHWW 7', WD 3', WW5.
NN26090- Seg 9 Finn Creek log jam 20' high @ 324' called impassable.
NN44010- Impassable culvert @ 498' ( 26' height, 36' diameter, 1' gradient, 40' length ) Check culvert.
NN44030- TR @ 555' listed as good flow, DNS.
Extensions- Township Range 12N/9w sections 29,28,21 (Finn Creek, see map) possible not previously surveyed.
SN19010- SN19090 Glides = 100% on most segments.
SN19071- SN19072 SVEG conifers both sides listed as 100 yr. old, 45 DBH, 190’ height.
SN19073- Ended @ culvert 200', called culvert passable w/ wetland above.
Note: First three segments listed no access for review.
SN45020- SVEG conifers r.s. listed as 40 yr. old, 14 DBH, 180' height. ( This area has patches of mature hemlock approx. 75 yr., 20 DBH, 150' height, otherwise surrounding conifers ones replanted in 1972.
SN45040- Percent glides 100% with substrate 36% cobble, 40% gravel,24% sand.
Correction: Reviewed on 11/22/96, glides presently @ 10%.
SN36000- Small feeder trib l.s @ 10' mark “noted” as DNS but crew mentioned they could be wrong about fish use, field review this stream.
Correction: ( Checked with Jeff Rudolph, stream needs no survey).
SN50000- Survey ended @ 800' w/ impassable log jam ( OHWW 15', wetted width 12). @ 300' stream forked, crew stayed left.
BR42000- Trib right @ 813' no survey, ck. for habitat?
Note: TR @ 1000' did survey
BR58000- Drainage @ 485', surveyed?
BR68000- End of survey, check for passage at description above. At end 890' large BDAM 80' x 200' x 30' deep, ck. for passage.
Note: Some segments in question for review were logged in Windows @ TWA.
NS02010- Sediment source cow xing severity (M) @ 950' ( possible fence project).
NS22020- SVEG conifer l.s. @ seg 2 25 yr., 12 DBH, 125' height. Conifer r.s. 25 yr., 10 DBH, 100' height.
Correction: Actual 50 yr., 90' height both sides.
NS41100- Seg 10 culvert @ 360' (location not in data) called impassable.
Correction: checked with Jeff Ruldoph, this stream called correctly.
NS43300- NS43500 SVEG conifers both sides (seg 3) 40 yr., 12 DBH, 100' height.
Correction: Deciduous 40 yr., 14 DBH, 120' height.
NS48010- SVEG conifers both sides 40 yr. 18/16 DBH, 150' height.
NS48020- NS48030 SVEG conifers both sides 40 yr., 17 DBH average, 140' height average.
NS48050- SVEG conifers both sides 40 yr., 16 DBH average, 150' height.
NS48060- SVEG conifer right side 30 yr., 10 DBH, 120' height.
NS48070- SVEG conifers both sides 40 yr., 16 DBH, 145' height average.
NS48080- SVEG conifers both sides 40 yr., 14 DBH, 150' height.
NS48090- SVEG conifers both sides 30 yr., 14 DBH, 130' height average. Check TL @ 470' for fish use.
Correction: Checked this stream, too steep for passage).
NS48032- SVEG conifers both sides 60 yr., 30 DBH, 170' height. (9L-1R).
NS48033- SVEG conifers both sides 40 yr., 18 DBH, 150' height.
Correction: Last 10 segments (NS433500-NS48033) conifers consistent average @ 60 yr., 100' height.
NS50010- Culvert @ 360' called no passage (4' height, 3' diameter, 60' length, 2% gradient).
Correction: Checked for passage, impassable at this time.
NA52000- Survey ended @ 35' into segment 53, check cascade for passage.
Correction: Checked with Jeff Rudolph, this area too steep for fish passage upstream of segment end point.
AR30000- TL @ 370' with 5' falls. Field reviewed this stream, had little potential habitat. No future survey needed.
AR24040- Incomplete survey
Appendix F: Stream Segments Ground Truthed with GPS
Stream Stream Segment Date Rover file Visited By
Middle Nemah Seg 40 11/26/96 R112622A JR
Middle Nemah Seg 42 11/26/96 R112623A JR
Middle Nemah Seg 42 11/26/96 R112623A JR
Middle Nemah Seg 4 (+25) 11/26/96 R112620A JR
EF Naselle Seg 81 11/21/96 R112119A JR
EF Naselle Seg 83 11/21/96 R112120A JR
EF Naselle Seg 84 11/12/96 R112121A JR
North Nemah (6L) Seg2 11/26/96 R112619A JR
North Nemah (6L) Seg 4 11/26/96 R112620A JR
North Nemah Seg 72 11/19/96 R11191B JR
North Nemah Seg 73 11/19/96 R111918B JR
North Nemah Seg 35 11/12/96 R111222A JR
North Nemah Seg 55 11/12/96 R111222B JR
North Nemah Seg 23 11/21/96 R112200A JR
Finn Creek Seg 6 11/26/96 R112618A JR
Finn Creek Seg 9 11/1/96 R110123A MS
Alder Creek Seg 3 11/1/96 R110117A MS
Alder Creek TL@800’ 11/1/96 R110117B MS
Alder Creek T2L 11/1/96 R110117C MS
Alder Creek Seg 6 11/1/96 R110117D MS
Cannon Seg 34 (35) 12/11/96 R121118A MD
Cannon 260’ Seg 32 (33) 12/11/96 R121119A MD
Palix (1L) Seg 2 12/11/22A R121122A MD
Palix (1L) Seg 4 12/11/22B R121122B MD
Palix Seg 3 (TL) 12/4/96 R120400A JR
Palix Seg 6 12/4/96 R120400B JR
Palix Seg12 12/4/96 R120400C JR
Williams Creek Bridge on NF 11/12/96 R111221A JR
Williams Creek Seg 18 11/2/96 R110200A MS
Appendix G: Database File Structure and STATISTICS
· Bear.DBF 168 records containing data for the Bear River and Greenhead Slough.
· Naselle.DBF 469 records containing data for: Alder Creek; Naselle (including E fork Naselle); Savage Creek.
·
Nemah.DBF 390 records containing data for: Clearwater
Creek; Finn Creek; Middle Nemah; North Nemah; South Nemah; and William’s Creek.
· Palix.DBF ****
Associated file size in Kilobytes:
Bear.DBF 132Kb
Naselle.DBF 356Kb
Nemah.DBF 297Kb
Palix.DBF 126Kb
Appendix H: Survey Form
Appendix G: Data Base Documentation