20190201 11282700 FALSE s100_wm_v4 002 476342.968800 532663.968700 4873579.000100 5051678.991600 1 0.651 Projected GCS_North_American_1983 NAD_1983_UTM_Zone_10N <ProjectedCoordinateSystem xsi:type='typens:ProjectedCoordinateSystem' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' xmlns:xs='http://www.w3.org/2001/XMLSchema' xmlns:typens='http://www.esri.com/schemas/ArcGIS/10.5'><WKT>PROJCS[&quot;NAD_1983_UTM_Zone_10N&quot;,GEOGCS[&quot;GCS_North_American_1983&quot;,DATUM[&quot;D_North_American_1983&quot;,SPHEROID[&quot;GRS_1980&quot;,6378137.0,298.257222101]],PRIMEM[&quot;Greenwich&quot;,0.0],UNIT[&quot;Degree&quot;,0.0174532925199433]],PROJECTION[&quot;Transverse_Mercator&quot;],PARAMETER[&quot;False_Easting&quot;,500000.0],PARAMETER[&quot;False_Northing&quot;,0.0],PARAMETER[&quot;Central_Meridian&quot;,-123.0],PARAMETER[&quot;Scale_Factor&quot;,0.9996],PARAMETER[&quot;Latitude_Of_Origin&quot;,0.0],UNIT[&quot;Meter&quot;,1.0],AUTHORITY[&quot;EPSG&quot;,26910]]</WKT><XOrigin>-5120900</XOrigin><YOrigin>-9998100</YOrigin><XYScale>450445547.3910538</XYScale><ZOrigin>-100000</ZOrigin><ZScale>10000</ZScale><MOrigin>-100000</MOrigin><MScale>10000</MScale><XYTolerance>0.001</XYTolerance><ZTolerance>0.001</ZTolerance><MTolerance>0.001</MTolerance><HighPrecision>true</HighPrecision><WKID>26910</WKID><LatestWKID>26910</LatestWKID></ProjectedCoordinateSystem> Linear Unit: Meter (1.000000) 20190201 11290800 20190201 11290800 FGDC CSDGM Metadata 1.0 150000000 5000 FGDC ESRI ArcCatalog 9.3.0.1770 en This data set is a spatial framework to be used for reporting the quantity of certain ecosystem components, here called ecosystem services, present in the central area of the floodplain of the Willamette River in Western Oregon lying between the confluence of the Coast and Middle forks of the Willamette in the south to the Columbia River confluence in the north. In this framework, the central area, the "pragmatic" floodplain, is subdivided into units, called slices. The slices lie at right angles to the principal axis of the Willamette mainstem with one slice every 100 meters of mainstem axis. The slices are uniquely identified by a sequential number beginning with 101 at the Columbia confluence and ending with 22907 at the confluence of the Coast and Middle forks. The number scheme is described in the Supplemental Information section. The quantities of ecosystem services are provided as attributes of the slice polygons and also through an accompanying spreadsheet (http://ise.uoregon.edu/slices/main.html). This framework is intended to support the assessment, planning, and monitoring of conservation and restoration activities within the Willamette River floodplain. These activities may include reconnecting historical channels that have been blocked and allowing the river to flood certain locations from which it is now normally excluded. Since restoration activities that would require disruption of significant constructed assets, roads, residential and commercial areas, and others, are infeasible, the spatial extent of this framework is narrower, primarily in the section between Eugene and Albany, than the historical floodplain. Users of this framework will compare past and present day baseline quantities of ecosystem services to both the planned and achieved results of conservation and restoration projects. As delivered in December 2016, Floodplain forest area is reported for the ca. 2010. Channel length and area are reported for ca. 2010. The channel length, channel area, and floodplain forest quantities are also reported at the locations and in the amounts simulated for the year 2050 in a scenario produced as part of the Pacific Northwest Ecosystem Research Consortium, called Conservation 2050 (http://oregonstate.edu/dept/pnw-erc/). We anticipate that users may add other measures of ecosystem function as their specific locations of interest make relevant, including the results of monitoring their projects over time. Via the ISE web site (http://ise.uoregon.edu/slices/main.html), we provide access to four types of information, each of which uses the slices as a reporting unit for processes and patterns that are critical to native ecosystem function. These four types of information are: 1) a series of 20 pdf maps showing slice boundaries and slice numbers superimposed on contemporary air photographs; 2) an Excel spreadsheet that reports amounts of key processes and patterns by slice and how they vary over time; 3) a digital map in both ArcGIS geodatabase and shapefile formats. 4] this metadata as a pdf. Using the slices framework consists of finding the portion of the floodplain (i.e. north, middle, or south) in which you are interested, and opening the relevant pdfs, spreadsheet or ArcGIS file that best suits your purposes. The pdfs are a series of 20 layered maps, each combining an air photo with taxlot boundaries, major road names and 1 km and 100m slice boundaries and numbers and thematic map layers. Together, they cover the entire pragmatic floodplain of the Willamette River. Construction of preliminary 100m slices using segment AML script The 100 meter polygons are designed to be regular spatial subdivisions of a preexisting map that subdivided the Willamette River historic floodplain into 1 kilometer spatial reporting units called "slices" developed by the Pacific Northwest Ecosystem Research Consortium (http://oregonstate.edu/dept/pnw-erc/), Existing Conditions, Theme Group, River Slices , SLICES_v21. The 1km slices were created by a combination of automated and manual steps that relied on a series of scripts written in the ArcInfo command language AML. The principal script in this suite is called segment.aml, originally developed by John Gabriel at Alsea Geospatial, Inc. in 1999, and modified for use in creating this 100m slice coverage, see copy included below. Primary inputs to the script are the name of a line coverage representing the principal axis of the River's mainstem and parameters defining the interval to be used in locating points along the axis and the length of lines subsequently to be drawn by the script at right angles to the axis through each of the points. The axis coverage NUAXIS was provided by the Department of Fisheries and Wildlife at Oregon State University. Prior to creating the 100m slices map, we created a 1km slices map and ensured that its lines were spatially coincident with those in SLICES_v21. Subsequent to the development of the 100m slices coverage, a revised 1km slices coverage, called V21E was created that, primarily for the reach south of Albany, reduced the width of the floodplain from the historic area of inundation to the narrower "pragmatic" floodplain defined as the area for which the 100m slice spatial reporting units are intended to be used. The output of the segment script is a series of arcs intersecting the main stem centerline at 100m intervals. ============= segment.aml ============= /*Segment.aml /*John Gabriel, Alsea Geospatail, Inc. john@ageospatial.com /*2/27/99 /*adapted from Interval.aml by Patti Haggerty at EPA /*writes an event table so a line of points can be created /*at user defined intervals along the route /*This aml was used to set up "slices" of floodplain along the /*simplified floodplain axis, for analysis of landscape and river /*change in the Willamette Valley, Oregon, 1999. /*====================== /*Part 1. Set intervals to use divide the stream /*======================= /*If error bailout &severity &error &routine bailout &echo &br &sv scratch '' &terminal 9999 precision double /*------------- /*query user for coverage & interval at which to break stream &sv cover [getcover * -line 'Select a coverage'] /*prompt user for a line &sv cov [entryname %cover%] &if [null %cov%] &then &stop No coverage selected. Aml terminated! &messages &pop /*display messages in a pop up window &sv int [response 'Enter an interval distance to divide the stream (defaut 1000m)' 1000] &if %int% le 0 &then &stop Interval distance must be greater than ZERO ~ aml terminated!] &sv event [substr %cov% 1 3]eve /*name the event table /*query user for width of arc use to discect the floodplain &sv width [response 'Enter width of floodplain (default is 1000 m)' 1000] /*query user for background coverage while stream line is being segmented &sv back [getcover * -all 'Select a background coverage'] /*query user for starting point to begin slices &sv d [getchoice -pairs 'Upper Left' UL 'Upper Right' UR 'Lower Left' LL 'Lower Right' LR -var where] &messages &on display 0 /*set the display to zero /*------------- /* tell the user the names of the route and event table &type COVER: %cov% &type ROUTE SYSTEM: %cov%.rte &type EVENT TABLE: %event% /*------------- /*route the coverage and calculate the route statistics &if [exists %cov%.ratrte -info] &then &sv d [delete %cov%.ratrte -info] &if [exists %cov%.secrte -info] &then &sv d [delete %cov%.secrte -info] build %cov% line renode %cov% arcroute %cov% rte # # # %where% /*Where sets the start point routestats %cov% rte all ~ /*------------------ /*In ArcPlot determine the number or routes ap asel %cov% route.rte &sv num = [extract 1 [show select %cov% route.rte]] &if %num% le 0 &then &stop Route file contains Zero Routes resel %cov% route.rte rte# = %num% &sv evkey [show select %cov% route.rte 1 item rte#] &sv length [show select %cov% route.rte 1 item measurelength] &sv low [show select %cov% route.rte 1 item lowmeasure] &sv high [show select %cov% route.rte 1 item highmeasure] /*Loop to write routes to scratch file &sv scratch [scratchname -prefix out -file] &do i = 1 &to %num% &by 1 &sys touch %scratch% &sv fileunit = 1 &sv j %low% /*added 8/3/99 /*write the information for the event table by processesing the routes &do i = %low% &to %high% &by %int% /*8/3/99 this measure should be where the point is starting at 0 and incrementing &sys echo %evkey% %j% >> %scratch% /*added 8/3 &sv j [calc %i% + %int%] /*added 8/3 &end /*&sys echo %evkey% %high% >> %scratch% /*&if [write %fileunit% %i%] = 0 &then /* &type Record %fileunit% successfully written to %scratch% /*&else &stop Record %fileunit% NOT written to %scratch% (high) /*&end &sv cls = [close -all] clearsel %cov% route.rte &end /*end loop that processes each route and writes to a scratch file q /*arcplot /*cat the file for error checking and pause &sys cat %scratch% tables /*start tables and create an event table &sv df [iteminfo %cov%.ratrte -info rte# -definition] &if [exists %event% -info] &then kill %event% define %event% rte#,%df% point 4 12 f 1 ~ add from %scratch% q /*quit tables /*================================== /*Part 2. Create a coverage of arcs with nodes at the user selected /*interval /*=================================== /*a. establisth the source for the event information eventsource add point %cov%_e %event% info linear rte# rte# point /*b. create a coverage of arcs with nodes at the interval &if [exist %cov%_2 -cover] &then kill %cov%_2 all eventpoint %cov% rte %cov%_e %cov%_2 pecision double build %cov%_2 point /*build %cov%_2 arc ae disp 9999 2 ec %cov%_2 de label ids draw &pause q /*quit ae &type quit ae /*==================== /*Part 2 add arcs perpendicular to label points /*==================== /*create nodes from the points /*&if [exists %cov%_pnt -cover] &then kill %cov%_pnt all /*&type built point cover %cov%_pnt /*process each of the node and draw the arcs in ae &sv done 1 &sv label1 1 ae /*start ae mape %cov% /*set the map extent to the original coverage ec %cov%_2 /*set edit coverage to the point coverage de arc label /*set draw environment bc %cov% 4 bc %back% 5 be arc draw &do &until %done% eq 0 ef label /*set the edit feature to point sel %cov%_2# = %label1% &sv done [show number selected] &if %done% ne 0 &then &do &sv labeln [show select 1] &sv xstart [extract 1 [show label %labeln% coordinate]] /*x coord &sv ystart [extract 2 [show label %labeln% coordinate]] /*y coord &sv label2 %label1% + 1 sel %cov%_2# = %label2% &sv done [show number selected] &if %done% ne 0 &then &do &sv labeln [show select 1] &sv xmiddl [extract 1 [show label %labeln% coordinate]] /*x coord &sv ymiddl [extract 2 [show label %labeln% coordinate]] /*y coord &sv yend %ymiddl% + %width% &sv deltax [abs [calc %xstart% - %xmiddl%]] &sv deltay [abs [calc %ystart% - %ymiddl%]] &sv linelen [invdistance %xstart% %ystart% %xmiddl% %ymiddl%] &sv r_angle [radang [invangle %xstart% %ystart% %xmiddl% %ymiddl%]] ef arc &pushpoint 2 %xmiddl% %ymiddl% &pushpoint 2 %xmiddl% %yend% &pushpoint 9 0 0 add /*add the arc &pushpoint 1 %xmiddl% %ymiddl% rotate %r_angle% &pushpoint 1 %xmiddl% %ymiddl% &pushpoint 1 %xmiddl% %ymiddl% copy /*copy the arc &pushpoint 1 %xmiddl% %ymiddl% rotate 180 &sv label1 = %label2% draw &end &end &end draw &type aml completed &ret /*==================== /*bailout and clean up /*==================== /*bailout &routine bailout &message &off &sv close = [close -all] &if [exist %scratch% -file] &then &sys rm %scratch% &if [show program] ne ARC &then quit &type AML segment failed &type &return &error \~ &type Aml file: %aml$file% &type Last error: %aml$message% \~ &type Line of error: %aml$errorline% &ret /*clean up routine &routine cleanup &if [exist %scratch% -file] &then &sys rm %scratch% &if [exist %scratch2% -file] &then &sys rm %scratch2% &ret Manual edits and assigning of slice numbers Manual editing was required after the AML created the 100m arc coverage. Where the centerline was predominantly north to south (for example, slices 8101 - 10010), the AML created arcs were evenly spaced and aligned as expected. Where the centerline contained a vertex to accommodate variations in the shape of the river (for example, slices 6401 - 7310), the 100m arcs created by the AML were spaced and aligned in ways that were inconsistent with our nesting of 100m slices within 1KM slices. Manual edits were done where required to correct spacing and to distribute arcs where there is a change in channel orientation. The primary guideline was to have each set of ten 100m slices nested within its 1KM slice, i.e. the northern boundary of 100m slice 8301 is coincident with the northern boundary of 1KM slice 83 and the southern boundary of slice 8310 is coincident with the southern boundary of 1KM slice 83. The 100m arc coverage was converted to a geodatabase feature for manual edits in ArcMap. The 1KM coverage was used as a background for editing to align 100m slices and adjust spacing. A 100m parallel spacing is not possible in locations where the channel changes orientation over a short distance (for example slices 3501 - 3510, 4601 - 4610). In these locations each set of ten 100m slices was connected to a common vertex defined by its bounding 1KM slice. Where the common vertex occurs depends on the change in orientation (at 1KM slice 35 it is to the west, at 1KM slice 71 it is to the south, at 1KM slice 124 it is to the east). Where the bounding 1KM slice met at a vertex on the boundary (for example 1KM slices 72 and 112), the 100m arcs were snapped to this same vertex. Where the 1KM slices did not have a common vertex at the boundary (for example 1KM slices 71, 73, 122), the 1KM boundary arcs were extended to create a common vertex outside of the boundary. The 100m slice arcs were snapped to this newly created common vertex and then extended past the boundary on the opposite side. To create the 100m polygons, the arc coverage was clipped to the boundary polygon and then appended to the boundary polygon. The assigning of 100m slice numbers used the 1KM slice numbers as a starting place. For each 100m slice nested within a 1KM slice, the identification begins with the 1KM slice number and is followed by 1 - 10 (1 is northernmost, 10 is southernmost). For example the 100m slices within 1KM slice 35 are 3501 - 3510. The assignment of 100m slice numbers was accomplished with a combination of GIS operations (point on polygon overlay, joinitem) and manual inspection and editing. Where a slice is not contiguous (for example slice 22701 which has 3 polygons), the same slice number identifies all polygons of that slice. Channel complexity measures, ca. 2010, Conservation 2050 The Pacific Northwest Ecosystem Research Consortium (PNW-ERC) (http://oregonstate.edu/dept/pnw-erc/) quantified channel complexity in the Willamette River floodplain using two measurements, area of river channel, and length of river channel, using a 1km slice spatial reporting framework for ca. 1850 "historical" and ca. 1995 "current" conditions. Wetted features classified as contributing to these measurements employ the table below. Based on this approach we report channel length and area of wetted features within the pragmatic floodplain for ca. 2010 conditions and for a future condition based on the PNW-ERC Conservation 2050 scenario, using the 100m slice spatial reporting framework of this coverage, S100FM. The attribute fields for these measures are for ca. 2010 conditions CHLEN10 and CHAREA10 and for the Conservation 2050 scenario, CHLEN50 and CHAREA50 see the Entity and Attribute section. Active Channel 2010 Feature Codes VALUE DESCRIPTION 1 Main active channel 3 Soughs, alcoves where one end is connected to main channel 5 Side or secondary channel, both ends connect to main channel 6 Remnant river feature within known floodplain. Disconnected floodplain lakes, sloughs. No apparent connection to main river but within historical floodplain area based on flood extent coverage. 7 Remnant river feature outside known floodplain. Disconnected floodplain lakes, sloughs. No apparent connection to main river and outside historical floodplain area based on flood extent coverage. 8 Human created water features primarily inside the known floodplain 9 Human created water features primarily outside the known floodplain 10 Tributary to Willamette River 11 Islands Ca. 2010 Channel Complexity measures The spatial analytic process that produces these measures first creates maps that depict the two phenomena of interest, area of channel and length of channel, overlays those maps individually with the 100m slices map, extracts the attribute table from this overlay for summation in a spreadsheet program, and finally attaches the area and length values to the 100m slice polygons as the new attributes identified above. A similar process is used for the Conservation 2050 complexity measures as described below. Ca. 2010 channel length For determining channel length, the source dataset is the THAL2000 line coverage produced by the Oregon State University Dept. of Fisheries and Wildlife containing lines indicating the position of the Thalweg, position of deepest channel, for streams, alcoves, and other features created by moving water. The center lines of features other than the Willamette River mainstem and principal tributaries were revised based on the Summer 2009 NAIP air photo imagery. The revisions including changing the positions of secondary stream features, connecting these features to the mainstem where the imagery showed that this is the case, and changing the type classification of these features as needed. The net effect is to increase the amount of channel length ca. 2010 relative to ca. 2000. Ca. 2010 channel area The source data set is the polygonal coverage AC2KV3 produced by the Oregon State University Dept. of Fisheries and Wildlife. It depicts the active channel of the Willamette River ca. 2000, classifying features according to the table above. For the present work, the AC2KV3 map was clipped to the pragmatic floodplain boundary, and relying largely on the 2009 NAIP digital aerial photographs, revised to reflect the latest available location and classification of summertime wetted features within the study area. These edits primarily were performed using the ArcInfo ArcEdit Version 8.3 program, with the NAIP 2009 photographs as visual reference. Additional data verification and editing was performed using ArcGIS9.3. When the 100m slices polygons were created, the geometrical complexity of the boundary of the pragmatic floodplain caused some of the slices to be broken into multiple polygons. This is the reason that there are more polygons in the present coverage than there are 100m slices. Conservation 2050 Channel Complexity measures Ca. 2050 channel length The PNW-ERC Conservation 2050 scenario identified and mapped locations in which historical channels were assumed to be restored via reconnection to the existing Willamette River. For the present work, the PNW-ERC CONS2050 restored channel locations, together with the 2009 NAIP imagery were used as source data for the creation of new channel center lines ca. 2050. These lines were added to the THAL2000 coverage with an attribute designation that identified them as the source of the s100fm channel complexity attributes for Conservation 2050 scenario. Because these added channel delineations are intended to represent the results of restoration projects lying largely within agricultural lands, their placement was, to the best of our ability, made consistent with the likely outcomes of agreements negotiated with landowners concerning the purchase of ecosystem services. Thus, the source datasets were used as general placement guides while the precise locations of restored channel were responsive to considerations of landowner preference. Restored channel, remnant water features present in the 2009 NAIP imagery but not present in the PNW-ERC Conservation 2050 map were delineated. In locations in which these new features were not more than 100 meters from the Willamette mainstem, or features connected to the mainstem, a new centerline was drawn to connect them to the mainstem network. Ca. 2050 channel area In locations in which the restored channel center lines pass through preexisting (remnant), but disconnected channel features, primarily those identified with feature CODE values of 6 in the table above, channel area is calculated from the area of the remnant features whether preexisting, or newly created from the NAIP imagery. In locations in which new channel lines were created on the basis of the PNW-ERC Conservation 2050 scenario map but were not present in the NAIP imagery, channel width was simulated by buffering the center lines to widths matched to existing local channels of comparable flow. Ecologically significant features Some of the ca. 2010 wetted areas are constructed ponds and gravel pits, feature CODE values 8 and 9 in the table above. These features are deemed not to have significant ecological value in the Conservation 2050 scenario and do not appear in it. However, restored channels may exist on or near these features. The restored channel area at these locations is the amount of channel area created by the buffering operation described above. Floodplain forest attributes Land use/ land cover data, derived from LANDSAT imagery, provide the foundation for floodplain forest circa 1990, 2000 and 2050. The source data for 1990 floodplain forest is land use/ land cover circa 1990 from the Pacific Northwest Ecosystem Research Consortium*. The data can be found at: http://www.fsl.orst.edu/pnwerc/wrb/access.html - existing conditions - datalayer - LULC ca.1990. Land use/ land cover classes 53-62, 86, 87, and 89** were selected from LULC 1990 to create a grid for 1990 floodplain forest. The source data for 2000 floodplain forest is land use/ land cover circa 2000 (http://www.fsl.orst.edu/pnwerc/wrb/access.html - existing conditions - datalayer - lulc2000). Land use/ land cover classes 53-54, 56-61, 86- 89,and 98 were selected from lulc2000 to create a grid for 2000 floodplain forest. Floodplain Forest ca. 2010 Although the determination of floodplain forest ca. 2000 was based on LANDSAT data, the map of floodplain forest ca. 2010 started with a LiDAR based delimitation of riparian vegetation produced by Jimmy Kagan, Director of the Oregon Biodiversity Information Center. A raster grid of Kagan's classes 9-17 (i.e. intermediate vegetation 2.5 feet and taller) was used as a starting point and then modified as follows: Territory was determined to be floodplain forest if it was 1) covered with natural vegetation identified from on-screen inspection of the 2009 NAIP aerial photos, and 2) regularly inundated based on a) River Design Group's 2-year flood inundation mapping for the main stem Willamette River and b) wet features included in the National Hydrologic Database (NHD+). In some locations, territory that is not included in the 2-year flood inundation footprint was mapped as floodplain forest. These are places that are surrounded by or adjacent to 2-year inundated territory, would likely be inundated in a 5 to 10 year flood event, and are dominated by natural vegetation in the NAIP aerial photography. In cases where contiguous patches of vegetated territory were partially included within a frequently inundated zone but also extended beyond that zone, topographic breaks, evident in the 10m LiDAR and NAIP aerial photos, were used to delimit the floodplain forest. Sand and gravel bars, primarily located in the main stem Willamette River, were mapped as floodplain forest. Even if not vegetated, it is presumed that those locations are potentially on a trajectory to become floodplain forest. 2 Year Flood Inundation Purpose: The regulated 2 year floodplain inundation layer illustrates areas of predicted inundation associated with a regulated 2-year flood event. These data were created by River Design Group, Inc. (RDG) to be used as a tool in identifying restoration opportunities on the Willamette River and are not intended for floodplain management or regulatory purposes. These data were delivered by RDG on July 7, 2012 and are used here with permission of RDG. Background: RDG is working with the Meyer Memorial Trust, Oregon Watershed Enhancement Board (OWEB), The Nature Conservancy (TNC), Oregon State University (OSU), and the University of Oregon (UO), to facilitate restoration opportunities on 160 miles of the Willamette River between Eugene and Oregon City, Oregon. The goal of the project is to use existing information to help guide restoration planning for the Willamette River. RDG used a LiDAR data set provided by DOGAMI and acquired by Watershed Sciences, Inc., as the topographic surface model. Stream gage data and USACE flood frequency analyses were used to calculate water surface trendlines. Trendlines were then used to create a 2-year regulated peak flow water surface layer that was overlaid onto the Willamette River LiDAR surface model and clipped to the pragmatic floodplain extent (i.e. boundary of significant infrastructure such as highways, residential areas, etc.). The resulting 2-year floodplain inundation layer illustrates areas of predicted inundation associated with a regulated 2-year flood event. Depth represents depth of water, in feet, above the topographic surface captured during LiDAR acquisition and does not necessarily reflect true water depth. Cold Water Refuges The data: The data for determining cold water were collected by Stanley V. Gregory (Department of Fisheries and Wildlife, Oregon State University) and his field crew from August 11, 2010 through August 27, 2010. The data extend the entire north/ south length of the slices framework and were collected in the Willamette mainstem, side channels and alcoves. The data were received as a text file with latitude, longitude and a suite of attributes for each location. The latitude/ longitude were used to create a spatially referenced point file (there are 872 data points) and the work presented here uses the maximum and minimum temperature attributes for each point. Overview: We use a definition of cold water derived from the state of Oregon's narrative temperature standard to identify cold water locations in side channels and alcoves. The comparison to determine cold water is between water temperatures in side channels and alcoves and the nearest upstream mainstem temperature point. The temperature calculation determines the difference between the mainstem maximum value and the side channel/ alcove minimum values. Any side channel or alcove point that is at least 2 degrees C colder than its nearest upstream mainstem temperature is determined to be 'cold'. Process: A) Identifying side channel/ alcove points and a mainstem reference point. A series of attributes were added to the maximum and minimum temperature attributes reported for each data point in the spatial data. The first added attribute assigns a unique numeric value to each group of side channel or alcove points. The points in each side channel or alcove were manually selected in ArcMap with the data displayed over the 2009 National Agriculture Imagery Program (NAIP) aerial image. The next added attribute identifies a mainstem reference point for each set of side channel or alcove points. This step also used the data displayed over the 2009 NAIP aerial image to manually select the mainstem point that is nearest to and upstream from where the side channel or alcove meet the mainstem. This process identified 59 groups of side channel/ alcove points, each with an associated mainstem reference point. B) Calculations to determine cold water. For each of the 59 side channel/ alcove groups, a temperature difference was calculated for each point in the group as follows: The minimum temperature for each point in the side channel/ alcove was subtracted from the maximum temperature for that group's mainstem reference point: temperature difference = mainstem reference maximum - side channel/ alcove minimum Note that the mainstem reference maximum is the same value for all points in a given side channel or alcove. If the result of the temperature difference is 2 or greater, the point is determined to be cold (i.e. the point is at least 2 degrees C colder that its mainstem reference point). A total of 71 points meet the 2 degree C temperature difference in the August 2010 data. C) Integrating cold water points with the 100m slice framework. Each of the 71 cold water points was associated with one of the 100m slices by intersecting the cold water point file with the 100m slice polygon file in ArcGIS. In some cases, a single 100m slice polygon contains 2 or more cold water points. So, although there are 71 cold water points, these points intersect only 46 of the 100m slices. We expanded the representation of cold to include 100m slices that are contiguous to and in the same side channel or alcove as one of the 46 identified cold water slices. The additional slices were manually selected with the 100m slice data displayed over the 2009 NAIP. The PDFs show the expanded representation of cold water slices. In the GIS attribute table for the 100m slices, the original 46 slices are identified with coldpt = 1, the manually selected slices are identified with coldpt = 2. Conservation 2050 The source data for 2050 conditions is the Conservation 2050 Scenario from the Pacific Northwest Ecosystem Research Consortium* (Chapter 7 in the Willamette River Basin Planning Atlas). (http://www.fsl.orst.edu/pnwerc/wrb/access.html - future scenarios - datalayer - conservation 2050). Land use/ land cover classes 53-54, 56-61, 86- 89, 98 and 101 were selected from the Conservation 2050 data to create a 2050 floodplain forest grid. *D. Hulse, S. Gregory, J. Baker. (Eds). 2002. Willamette River Basin Planning Atlas: Trajectories of environmental and ecological change. (2nd edition), Oregon State University Press, Corvallis, Oregon 97333. 180 p. Available at: http://www.fsl.orst.edu/pnwerc/wrb/Atlas_web_compressed/PDFtoc.html **The translation for the numeric LULC classes is: 53 Forest closed hardwood 54 Forest closed mixed 55 Upland forest semi-closed conifer 56 Conifers 0 - 20 yrs. 57 Forest closed conifer 21 - 40 yrs. 58 Forest closed conifer 41 - 60 yrs. 59 Forest closed conifer 61 - 80 yrs. 60 Forest closed conifer 81 - 200 yrs. 61 Forest closed conifer older than 200 yrs. 62 Upland forest semi-closed hardwood 86 Natural grassland 87 Natural shrub 89 Flooded/marsh 98 Oak Savanna 101 Wet shrub A gridded version of the 100m slices was overlaid with the floodplain forest grids (1990, 2000, 2010, and 2050) to determine the amount of floodplain forest in each 100m slice polygon at each time. The floodplain forest cell counts per polygon were joined back to the 100m slices coverage and converted to acres with each 30m X 30m grid cell = 0.2224 acres. 2-year flood inundation attribute (FLD_2YR) The regulated 2-year floodplain inundation data represents areas of predicted inundation associated with a regulated 2-year flood event. These data were created by RIver Design Grouop (RDG http://www.riverdesigngroup.com/) to be used as a tool in identifying restoration opportunities on the Willamette River and are not intended for floodplain management or regulatory purposes. These data were delevered by RDG on July 7,2012 and are used here with permission of RDG. RDG is working with the Meyer Memorial Trust, the Oregon Watershed Enhancement Board, Oregon State University, the University of Oregon and The Nature Conservancy to facilitate restoration opportunities on a 62 mile reach of the Willamette River between Eugene and Albany, Oregon. The goal of the project will use existing information to help guide restoration planning for the Willamette River. RDG used LiDAR data set gathered between August 2008 and February 2009, provided by the Oregon Department of Geology and Mineral Industries (DOGAMI) and acquired by Watershed Sciences, Inc., as the topographic surface model. Stream gage data and U.S. Army Corps of Engineers flood frequency analyses were used to calculate water surface trend lines. Trend lines were then used to create a 2-year regulated peak flow water surface layer that was overlaid onto the Willamette River LiDAR surface model and clipped to the pragmatic floodplain extent (i.e. boundary of significant infrastructure such as highways, residential areas, etc. The resulting 2-year floodplain inundation layer illustrates areas of predicted inundation associated with a regulated 2-year flood event. Depth represents depth of water, in feet, above the topographic surface captured during LiDAR acquisition and does not necessarily reflect true water depth. . Institute for a Sustainable Environment March 2013 s100fm_3 vector digital data http://ise.uoregon.edu/slices/main.html publication date September 2010 Complete Intermittent -123.302203 -122.579828 45.618820 44.014783 476438.678988 532758.999988 4873379.714438 5051479.487938 None Restoration Channel complexity Floodplain forest Monitoring Flood inundation Western Oregon Future scenario Available via ISE web site http://ise.uoregon.edu/slices/main.html The 2-year flood inundation data are not intended for floodplain management or regulatory purposes. ArcInfo Coverage David Hulse University of Oregon, Institute for a Sustainable Environment Professor (541) 346-3672 (541) 346-3626 dhulse@uoregon.edu 8:00 AM - 5:00 PM Email preferred. Institute for a Sustainable Environment Pacific Northwest Ecosystem Research Consortium September 2002 Second http://oregonstate.edu/dept/pnw-erc/ Version 6.2 (Build 9200) ; Esri ArcGIS 10.5.0.6491 s100_wm_v4 -123.303439 -122.581036 45.618656 44.014635 1 476438.678988 532758.999988 5051479.487938 4873379.714438 1 -123.302203 -122.579828 45.61882 44.014783 1 002 This framework is intended to support the assessment, planning, and monitoring of conservation and restoration activities within the Willamette River floodplain. These activities may include reconnecting historical channels that have been blocked and allowing the river to flood certain locations from which it is now normally excluded. Since restoration activities that would require disruption of significant constructed assets, roads, residential and commercial areas, and others, are infeasible, the spatial extent of this framework is narrower, primarily in the section between Eugene and Albany, than the historical floodplain. Via the ISE web site (http://ise.uoregon.edu/slices/main.html), we provide access to four types of information, each of which uses the slices as a reporting unit for processes and patterns that are critical to native ecosystem function. These three types of information are: 1) a series of 20 pdf maps showing slice boundaries and slice numbers superimposed on contemporary air photographs; 2) an Excel spreadsheet that reports amounts of key processes and patterns by slice and how they vary over time; 3) a digital map in both ArcGIS geodatabase and shapefile formats. Using the slices framework consists of finding the portion of the floodplain (i.e. north, middle, or south) in which you are interested, and opening the relevant pdfs, spreadsheet or ArcGIS file that best suits your purposes. The pdfs are a series of 20 layered maps, each combining an air photo with taxlot boundaries, major road names and 1 km and 100 m slice boundaries and numbers and thematic map layers. Together, they cover the entire pragmatic floodplain of the Willamette River. <DIV STYLE="text-align:Left;"><DIV><DIV><P><SPAN>This data set is a spatial framework to be used for reporting the quantity of certain ecosystem components, here called ecosystem services, present in the central area of the floodplain of the Willamette River in Western Oregon lying between the confluence of the Coast and Middle Forks of the Willamette in the south to the Columbia River confluence in the north. In this framework, the central area, the "pragmatic" floodplain, is subdivided into units, called slices. The slices lie at right angles to the principal axis of the Willamette floodplain with one slice every 100 meters of mainstem axis. Each slice is identified uniquely by a sequential number beginning with 101 at the Columbia confluence and ending with 22907 at the confluence of the Coast and Middle Forks. The number scheme is described in the Supplemental Information section. The quantities of ecosystem services are provided as attributes of the slice polygons and also through an accompanying spreadsheet (http://ise.uoregon.edu/slices/main.html).</SPAN></P><P><SPAN /></P></DIV></DIV></DIV> Institute for a Sustainable Environment RestorationChannel complexityFloodplain forestMonitoringFlood inundation <DIV STYLE="text-align:Left;"><DIV><DIV><P><SPAN>The 2-year flood inundation data are not intended for floodplain management or regulatory purposes.</SPAN></P></DIV></DIV></DIV> Construction of 100 m slices The 100 meter polygons are designed to be regular spatial subdivisions of a preexisting map that subdivided the Willamette River historic floodplain into 1 kilometer spatial reporting units called "slices" developed by the Pacific Northwest Ecosystem Research Consortium http://oregonstate.edu/dept/pnw-erc/ . The 1 km slices were created by a combination of automated and manual steps that relied on a series of scripts written in the ArcInfo command language AML. The principal script in this suite is called segment.aml, originally developed by John Gabriel at Alsea Geospatial, Inc. in 1999, and modified for use in creating this 100 m slice coverage. Primary inputs to the script are the name of a line coverage representing the principal axis of the River's mainstem and parameters defining the interval to be used in locating points along the axis and the length of lines subsequently to be drawn by the script at right angles to the axis through each of the points. The axis coverage NUAXIS was provided by the Department of Fisheries and Wildlife at Oregon State University. Prior to creating the 100 m slices map, we created a 1 km slices map and ensured that its lines were spatially coincident with those in SLICES_v21. Subsequent to the development of the 100 m slices coverage, a revised 1 km slices coverage, called V21E was created that, primarily for the reach south of Albany, reduced the width of the floodplain from the historic area of inundation to the narrower "pragmatic" floodplain defined as the area for which the 100 m slice spatial reporting units are intended to be used. The output of the segment script is a series of arcs intersecting the main stem centerline at 100 m intervals. Manual edits and assigning of slice numbers Manual editing was required after the AML created the 100 m arc coverage. Where the centerline was predominantly north to south (for example, slices 8101 - 10010), the AML created arcs were evenly spaced and aligned as expected. Where the centerline contained a vertex to accommodate variations in the shape of the river (for example, slices 6401 - 7310), the 100 m arcs created by the AML were spaced and aligned in ways that were inconsistent with our nesting of 100 m slices within 1 km slices. Manual edits were done where required to correct spacing and to distribute arcs where there is a change in channel orientation. The primary guideline was to have each set of ten 100 m slices nested within its 1 km slice, i.e. the northern boundary of 100 m slice 8301 is coincident with the northern boundary of 1 km slice 83 and the southern boundary of slice 8310 is coincident with the southern boundary of 1 km slice 83. The 100 m arc coverage was converted to a geodatabase feature for manual edits in ArcMap. The 1km coverage was used as a background for editing to align 100 m slices and adjust spacing. A 100 m parallel spacing is not possible in locations where the channel changes orientation over a short distance (for example slices 3501 - 3510, 4601 - 4610). In these locations each set of ten 100 m slices was connected to a common vertex defined by its bounding 1 km slice. Where the common vertex occurs depends on the change in orientation (at 1 km slice 35 it is to the west, at 1 km slice 71 it is to the south, at 1 km slice 124 it is to the east). Where the bounding 1 km slice met at a vertex on the boundary (for example 1 km slices 72 and 112), the 100 m arcs were snapped to this same vertex. Where the 1 km slices did not have a common vertex at the boundary (for example 1 km slices 71, 73, 122), the 1 km boundary arcs were extended to create a common vertex outside of the boundary. The 100 m slice arcs were snapped to this newly created common vertex and then extended past the boundary on the opposite side. To create the 100 m polygons, the arc coverage was clipped to the boundary polygon and then appended to the boundary polygon. The assigning of 100 m slice numbers used the 1 km slice numbers as a starting place. For each 100 m slice nested within a 1 km slice, the identification begins with the 1 km slice number and is followed by 1 - 10 (1 is northernmost, 10 is southernmost). For example, the 100 m slices within 1 km slice 35 are 3501 - 3510. The assignment of 100 m slice numbers was accomplished with a combination of GIS operations (point on polygon overlay, joinitem) and manual inspection and editing. Where a slice is not contiguous (for example slice 22701 which has 3 polygons), the same slice number identifies all polygons of that slice. Ecological attributes Channel complexity measures CHLEN10, CHLEN50, CHLEN50, CHAREA10 The Pacific Northwest Ecosystem Research Consortium PNW-ERC) http://oregonstate.edu/dept/pnw-erc/ quantified channel complexity in the Willamette River floodplain using two measurements, area of river channel, and length of river channel, using a 1 km slice spatial reporting framework for ca. 1850 "historical" and ca. 1995 "current" conditions. Based on this approach we report channel length and area of wetted features, within the pragmatic floodplain for ca. 2010 conditions and for ca. 2050 future conditions based on the PNW-ERC Conservation 2050 scenario, using the 100 m slice spatial reporting framework of this shape file, S100_wm_v4. Additional detail concerning the development of these measures can be found in the section of the Technical Details Channel Complexity page at the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html. Two-year Inundation FLD_2YR Two-year floodplain inundation identifies areas of predicted inundation associated with a regulated 2-year flood event. Depth represents depth of water, in feet, above the topographic surface captured during LiDAR acquisition and does not necessarily reflect true water depth. These data were created by River Design Group, Inc. (RDG) to be used as a tool in identifying restoration opportunities on the Willamette River and are not intended for floodplain management or regulatory purposes. The data were delivered by RDG on July 7, 2012 and are used here with permission of RDG. Additional detail concerning the procedures used to produce the values for Two Year Flood Inundation can be found in the Technical Details 2 Year Flood ca. 2010 section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Floodplain forest measures FPF2010, FPF2050, PBF2010, PBF2010KM Floodplain Forest ca. 2010 is reported in acres for each 100 m slice in the field FPF2010. The representation of floodplain forest in the 100 m SLICES is derived from multiple datasets and two sources with different underlying spatial grains. In slices 1 – 7907, the primary data source is Landsat satellite data at a 30 m spatial grain. In slices 7908 – 22907, the representation was processed at a 6ft. grain using 2011 NAIP imagery and ca. 2009 LiDAR data. Additional detail concerning the procedures used to produce the values for Floodplain forest can be found in the Technical Details section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Percent bank forested measures PB2010, PBF2010KM Percent Bank Forested ca. 2010 is reported for each 100 m slice and also for each 1 km slice. In the GIS data, Percent Bank Forested is reported in the field PBF2010 for the 100 m slices and the field PBF2010KM for the 1 km slices. Percent Bank Forested ca. 2010 is calculated per 100 m slice to be the area of floodplain forest within 120 feet (one site potential tree height) of the low water bank divided by the total area within 120 feet of the bank. For each 1 km slice, the Percent Bank Forested is calculated to be the area of floodplain forest within 120 feet of the low water bank divided by the total area within 120 feet of the bank. Additional detail concerning the procedures used to produce the values for Percent Bank Forested can be found in the Technical Details percent Bank Forested section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Cold water Refuges CLDO_2015 Cold water refuges are reported for the 100 m slices. The value of 1 indicates that between 2011 and 2016 at least one data point collected in the corresponding 100 m slice met the definition of cold water refuge (explained below). The value of zero means that data collected in that 100 m slice did not meet the definition of cold water OR that no data were collected in that 100 m slice. Cold water refuges in the Willamette River occur in sloughs and side channels, where subsurface water emerges and exchanges slowly with the mainstem river. Cold water refuges are defined here as locations where, in the months of July and August, the slough or side channel temperature is 2 degrees C colder than the daily maximum temperature of the associated mainstem Willamette River and the concentration of dissolved oxygen is 4.0 mg/L or greater. The 4.0 mg/L value represents a concentration at which native fish can survive but this should not be confused with optimal dissolved oxygen for native fish. Additional detail concerning the procedures used to produce the values for Cold Water Refuges can be found in the Technical Details/Cold Water Refuges section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Juvenile Spring Chinook Habitat measures JCRES2010, JCCON2010 Two attributes are associated with juvenile Spring Chinook habitat, JCRES_2010 reports the number of acres in each 100 m slice identified for restoration, and the attribute JCCON_2010 reports the number of acres in each 100 m slice identified for conservation. Additional detail concerning the procedures used to produce the values for Cold Water Refuges can be found the Technical Details/Juvenile Chinook section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Native fish measures PCNAT2010, SABUN2010 The attribute field PCNAT2010 reports the percentage of native fish out of all fish captured in each 1 km slice (i.e. the value 93 represents 93 percent native fish). Salmonid abundance, the number of Salmonids captured in each 1 km slice is reported in the GIS attribute field SABUN2010. The percent native and salmonid abundance are reported in the first 100m slice (lowest number, ending in 01) for the entire associated 1 km slice. For example, salmonid abundance in 100 m slice number 21601 is 51; the value of 51 reports the number of salmonids captured in 1 km slice 216 (100 m slices 21601 - 21610). Additional detail concerning the procedures used to produce the values for Native fish can be found in the Technical Details Native fish section of the ISE Slices Web site: http://ise.uoregon.edu/slices/main.html . Chris Enright Institute for a Sustainable Environment, University of Oregon Research Associate en FGDC Content Standards for Digital Geospatial Metadata FGDC-STD-001-1998 local time David Hulse Institute for a Sustainable Enviornment mailing address Eugene OR 07403-5243

5243 Dept. of Landscape Architecture

Unniversity of Oregon

USA (541) 346-3672 Professor (541) 346-3626 dhulse@uoregon.edu 20130404 September 2010 As required Available as external PDF. ISO 19115 Geographic Information - Metadata DIS_ESRI1.0 dataset Available as coverage and shape file via the ISE web site: http://ise.uoregon.edu 1.484 1.484 Arc coverage and shapefile Zipped tar file Coverage and shapefile in ArcInfo workspace Zip http://ise.uoregon.edu None. None Available via web page only. David Hulse Institute for a Sustainable Environment Professor (541) 346-3672 (541) 346-3626 dhulse@uoregon.edu 8:00 AM - 5:00 PM Email preferred. October 1, 2010 002 withheld Local Area Network 1.484 Shapefile 0.651 Vector Simple FALSE 2602 FALSE FALSE GCS_North_American_1927 NAD_1927_UTM_Zone_10N coordinate pair meters 0.000000 0.000000 North American Datum of 1927 Clarke 1866 6378206.400000 294.978698 NAD_1927_UTM_Zone_10N EPSG 7.8.2(3.0.1) 2602 s100_wm_v4 Feature Class 2602 100m flood plain slices Internal FID FID OID 4 0 0 Internal feature number. Esri Sequential unique whole numbers that are automatically generated. Shape Shape Geometry 0 0 0 Feature geometry. ESRI Coordinates defining the features. AREA 19 18 Double Area of feature in internal units squared. ESRI Positive real numbers that are automatically generated. AREA 0 0 PERIMETER 19 18 Double Perimeter of feature in internal units. ESRI Positive real numbers that are automatically generated. PERIMETER 0 0 JF JF Integer 5 5 0 This field is used in the management of attributes obtained from external data sources. ISE. FP_KM FP_KM Integer 5 5 0 Beginning from the confluence of the Columbia River, segments of floodplain 1 km apart are identified by sequential numbering. ISE. FP_100m FP_100m Integer 5 5 0 100 m slices are enumerated in the north to south, upstream direction. ISE. FP_KM01 FP_KM01 Integer 5 5 0 Identifies the 1KM slice for attributes that are reported only at the kilometer extent. CHLEN10 CHLEN10 Double 19 0 0 The ca. 2010 centerline length in meters of wetted features connected to the Willamette River mainstem. ISE. CHLEN50 CHLEN50 Double 19 0 0 The ca. 2050 centerline length in meters of wetted features connected to the Willamette River mainstem under the PNW-ERC's Conservation 2050 scenario. ISE. CHAREA10 CHAREA10 Double 19 0 0 The ca. 2010 area in square meters of wetted features within the pragmatic floodplain. ISE. CHAREA50 CHAREA50 Double 19 0 0 The ca. 2050 area in square meters of wetted areas within the pragmatic floodplain under the PNW-ERC's Conservation 2050 scenario. ISE. FPF2010 FPF2010 Single 6 5 2 Floodplain forest area in acres derived from land use land cover classes from a ca. 2000 land use land cover. representation. ISE. FPF2050 FPF2050 Single 6 5 2 Floodplain forest area in acres derived from land use land cover classes from PNW-ERC's 2050 Conservation scenario land use land cover in slices 1 - 7901 at 30 m grain, and ca. 2010 6 ft Landuse-Landcover in slices 7902 - 22902. ISE. PBF2010 PBF2010 Integer 5 5 0 Percent Bank Forested ca. 2010 is reported for each 100 m slice. ISE. PBF2010KM PBF2010KM Integer 5 5 0 Percent Bank Forested ca. 2010 is reported for each 1 km slice. ISE. FLD_2YR FLD_2YR Single 6 5 1 Area in acres of predicted inundation associated with a regulated 2-year flood. ISE. CLDO_2015 CLDO_2015 Integer 5 5 0 Locations of cold water in side channels and alcoves. ISE. JCRES2010 JCRES2010 Single 6 5 2 JCRES_2010 reports the number of acres in each 100 m slice identified for restoration. ISE. JCCON2010 JCCON2010 Single 6 5 2 JCCON_2010 reports the number of acres in each 100 m slice identified for conservation. ISE. PCNAT2010 PCNAT2010 Integer 5 5 0 Of all fish captured in each 1 km slice, PCNAT2010 reports the percentage that are native fish. ISE. SABUN2010 SABUN2010 Integer 5 5 0 Salmonid abundance, the number of Salmonids captured in each 1 km slice is reported in the GIS attribute field SABUN2010. ISE. 20190201 The accuracy of the quantities reported for the ecosystem services was checked by multiple approaches: 1. Fifty 100m slices were selected at random and all of the attribute values were manually calculated from the source maps and compared to the reported values. 2. All instances of specific spatial operations were examined. For example, all locations in which new channels were created in the Conservation 2050 scenario were examined to determine if new wetted area of the correct amount was created; all slices in which ca. 2050 wetted area was less than the ca. 2010 wetted area were isolated and checked to ensure that the reason for the decline was the removal of the constructed ponds and pits which are deemed not to have ecological value in the Conservation 2050 scenario, see additional note below. 3. Due to the complexity of the floodplain boundary, some slices are discontinuous, comprising multiple polygons. Since the attribute values are reported for each polygon, the correct value for a particular slice consists of the sum of the values for each of the polygons that make up the slice. All slices consisting of more than one polygon were isolated and the total values compared to the values reported in the accompanying spreadsheet, which are reported at the slice level of spatial aggregation. Additional accuracy considerations: The ArcInfo spatial overlay operations used to derive areal values introduced uncertainties of 70m2 or less in the totals reported for the 100m slices. Source feature placement and classification The quantities of ecosystem services reported for each 100m slice are determined by spatial overlay operations. The source map for channel length, for example, contains arcs some of which represent ca. 2010 channel and others that represent the Conservation 2050 channels. A similar situation exists with respect to wetted area. The accuracy of the reported values therefore depends on the correct classification of the source features. Feature placement is also critical. Using the late Summer 2009 National Agricultural Imagery Program imagery, the originating source maps were first revised to reflect ca. 2010 locations of channels and wetted areas, then further revised to depict restored features under the ERC Conservation 2050 scenario. In preparation for this work, choices were made that defined features of interest and rules for delineation. As work progressed, in-house reviews were augmented with reviews by an expert outside of the production group familiar with the River, its floodplain, the objectives of the research, and the intended uses of the framework. Multiple examinations of the features of this type were conducted, examining the placement and classification of each feature in each of the 100m slices, until no errors were detected. The 100m spatial framework in this coverage is aligned with a previous one kilometer spatial reporting framework covering the same locations. The definitiins of channel complexity, channel length and channel area, and the definition of floodplain forest are the same as those used in the PNW-ERC work cited elsewhere in this metadata. The spatial framework is space filling within the defined floodplain boudaries. All instances of measured phenomena available in source data are reported within the spatial framework. Framework geometry & slice numbering: This 100m slice framework is an enhancement to a previous 1km framework (http://www.fsl.orst.edu/pnwerc/wrb/access.html), hence the arcs of the 100m slices must align exactly with the those of the 1km framework where they are coincident. Further, the 100m slices must subdivide the 1km slices evenly. Turns in the mainstem axis create the principal challenges to proper alignment. Through these turns slice arcs cannot be parallel. Where necessary manual editing of arc placement was used to ensure even subdivision of each 1km slice, and to ensure that polygon topology was correctly constructed. No instance of misalignment greater than 0.1m has been detected. The 100m slices are uniquely labeled in such a way that the 1km slice in which each is embedded is also identified in the label field, FP_100M. The value 109, for example, denotes the ninth 100m slice is the first kilometer slice. Using ArcMap to display them, each of the 100m slices were examined by two people to ensure the correctness of these identifiers. Unknown See supplemental information section. Dataset copied. withheld 20110420 11281400 Metadata imported. withheld 20130320 14301000 The 100 m spatial framework in this coverage is aligned with a previous one kilometer spatial reporting framework covering the same locations. The definitions of channel complexity, channel length and channel area, and the definition of floodplain forest are the same as those used in the PNW-ERC work cited elsewhere in this metadata. The spatial framework is space filling within the defined floodplain boundaries. All instances of measured phenomena available in source data are reported within the spatial framework. Framework geometry & slice numbering: This 100 m slice framework is an enhancement to a previous 1 km framework (http://www.fsl.orst.edu/pnwerc/wrb/access.html), hence the arcs of the 100 m slices must align exactly with the those of the 1 km framework where they are coincident. Further, the 100 m slices must subdivide the 1 km slices evenly. Turns in the mainstem axis create the principal challenges to proper alignment. Through these turns slice edges cannot be parallel. Where necessary manual editing of edge placement was used to ensure even subdivision of each 1 km slice, and to ensure that polygon topology was correctly constructed. No instance of misalignment greater than 0.1 m has been detected. The 100 m slices are uniquely labeled in such a way that the 1km slice in which each is embedded is also identified in the label field, FP_100M. The value 109, for example, denotes the ninth 100 m slice is the first kilometer slice. Using ArcMap to display them, each of the 100 m slices were examined by two people to ensure the correctness of these identifiers. The geometry of the 1 km and 100 m slices was analyzedfor polygonal closure. The separation of slices at their intersection with the floodplain axis was enforced at 100 meters. Geometry of spatial reporting units The accuracy of the quantities reported for the ecosystem services was checked by multiple approaches: 1. Fifty 100 m slices were selected at random and all of the attribute values were manually calculated from the source maps and compared to the reported values. 2. All instances of specific spatial operations were examined. For example, all locations in which new channels were created in the Conservation 2050 scenario were examined to determine if new wetted area of the correct amount was created; all slices in which ca. 2050 wetted area was less than the ca. 2010 wetted area were isolated and checked to ensure that the reason for the decline was the removal of the constructed ponds and pits which are deemed not to have ecological value in the Conservation 2050 scenario, see additional note below. 3. Due to the complexity of the floodplain boundary, some slices are discontinuous, comprising multiple polygons. Since the attribute values are reported for each polygon, the correct value for a particular slice consists of the sum of the values for each of the polygons that make up the slice. All slices consisting of more than one polygon were isolated and the total values compared to the values reported in the accompanying spreadsheet, which are reported at the slice level of spatial aggregation. Additional accuracy considerations: The ArcInfo spatial overlay operations used to derive areal values introduced uncertainties of 70 m2 or less in the totals reported for the 100 m slices. Source feature placement and classification The quantities of ecosystem services reported for each 100 m slice are determined by spatial overlay operations. The source map for channel length, for example, contains arcs some of which represent ca. 2010 channel and others that represent the Conservation 2050 channels. A similar situation exists with respect to wetted area. The accuracy of the reported values therefore depends on the correct classification of the source features. Feature placement is also critical. Using the late Summer 2009 National Agricultural Imagery Program imagery, the originating source maps were first revised to reflect ca. 2010 locations of channels and wetted areas, then further revised to depict restored features under the ERC Conservation 2050 scenario. In preparation for this work, choices were made that defined features of interest and rules for delineation. As work progressed, in-house reviews were augmented with reviews by an expert outside of the production group familiar with the River, its floodplain, the objectives of the research, and the intended uses of the framework. Multiple examinations of the features of this type were conducted, examining the placement and classification of each feature in each of the 100 m slices, until no errors were detected. Description of attribute validity assurance process