Mountain Meadows - Sierra Nevada - UC Davis [ds2671]

Brief Methods:

UC Davis compiled the best available meadow polygon layers into a single data layer. Data layers were collected from various agencies, individuals, and organizations. Data layer quality varied based on compilation methods and age; some layers were excluded due to poor data quality. A confidence rank (1 = low, 10 = high) was assigned to the remaining layers which were rasterized at a 10m resolution. The layers were combined and combined raster cells with a summed rank of 2 or less were excluded. Raster cells representing open water were also excluded. A majority filter was run on the resulting remaining cells to reduce boundary heterogeneity, which replaced cell values based on the majority of the eight neighboring cells. Individual meadow polygons were created through a raster to vector conversion that treated all contiguous cells as a single part meadow feature with boundaries smoothed using the Polynomial Approximation with Exponential Kernel (PAEK) method (20 m tolerance to reduce edge complexity). Polygons with an area less than 0.4 ha (less than 1 acre) were removed from the final meadow composite. Original IDs and other attributes were attached to the meadow polygons.

Compiled Sources

Layer Name	Source	Spatial Coverage
AMPH	Stanislaus National Forest	Sierra Nevada
AMR	American Rivers	Yuba and Mokelumne watersheds
CALVEG	USFS Region 5 Clearinghouse	Sierra Nevada
ENFa	Eldorado National Forest	Eldorado National Forest
INFa	Inyo National Forest	Inyo National Forest
INFb	Inyo National Forest	Inyo National Forest
INFd	Inyo National Forest	Inyo National Forest
INFe	Inyo National Forest	Inyo National Forest
INFf	Inyo National Forest	Kern Plateau
JMAA	Inyo National Forest	John Muir and Ansel Adams Wilderness
LAVO*	Lassen Volcanic National Park	Lassen Volcanic National Park
LNFa	Lassen National Forest	Lassen National Forest
LNFc	Lassen National Forest	Lassen National Forest
LNFd	Lassen National Forest	Lassen National Forest
NHD	USGS - National Hydrography Dataset	Sierra Nevada
NWI	USFWS - National Wetlands Inventory	Sierra Nevada
PNFa	Plumas National Forest	Plumas National Forest
SEKIa	UC Merced	Sequoia Kings Canyon National Park
SNFa	Sierra National Forest	Sierra National Forest
SQFa	Sequoia National Forest	Sequoia National Forest
STFa	Stanislaus National Forest	Stanislaus National Forest
STI	Stillwater Sciences	Sierra Nevada
TMUa	Lake Tahoe Basin Management Unit	LTBMU
TMUb	Lake Tahoe Basin Management Unit	LTBMU
TMUc	Lake Tahoe Basin Management Unit	LTBMU
TNFa*	Tahoe National Forest	Tahoe National Forest
TNFb	Tahoe National Forest	Tahoe National Forest
UCD**	UC Davis	Sierra Nevada
WEIX	USFS Region 5 Range	Sierra Nevada
WIFL	USFS Region 5 Clearinghouse	Sierra Nevada
YOSEa	UC Merced	Yosemite National Park

*Restricted layer use*

**Two layers**

Acknowledgements

We would like to acknowledge the following people for providing meadows data used in the Sierra Nevada Multi-Source Meadow Polygons Compilation (v 1.0):

Nathan Amboy

Oswaldo Angulo

John Babin

Eric Berlow

Kirsten Bovee

Cathy Brown

Coye Burnet

Leslie Chow

Janet Coles

Alan Doerr

Elaine Elliott

Shana Gross

Sally Hallowell

Anthony Hewitt

Heidi Hosler

Luke Hunt

Erin Lutrick

Amy Merrill

Steve Peaslee

Mark Schug

Michele Slaton

Ed Tallyn

Alissa Tanner

Deb Tatman

Heather Taylor

Kurt Teuber

Dave Weixelman

Lucas Wilkinson

Expanded Methods

Sierra Nevada Multi-Source Meadow Polygons Compilation

We used ArcGIS v10 (ESRI, Redlands, CA, USA) for all geoprocessing.

A total of 44 meadows layers were collected from various individuals, agencies, and organizations. Layers varied in number of features and feature topology (i.e., points, lines, and polygons), meadow type (e.g., wet meadow, dry meadow, fen, etc.), extent (local vs. study area wide), method of creation (digitized from aerial photographs, geolocated (i.e., GPSd), extracted from GIS layers, Feature Analyst), purpose, and accuracy. All layers were examined for completeness, attribution, and spatial precision. High quality data layers were selected for use in the composite data set, and in few cases selected features were extracted to represent restricted spatial extents. In other cases, selected point layers were used to generate representative polygons using high resolution aerial imagery (less than or equal to 1 m pixels). We deemed 32 meadows layers suitable for compilation (Compiled Sources Table).

We examined each layer, removed any duplicate features, and assigned a confidence rank based on source, method of creation, representativeness (feature accuracy), and spatial precision. Confidence ranks were assigned as either 1 (low) or 10 (high). Highly ranking layers were often created in the field by trained personnel using survey grade global position system receivers, whereas low ranking layers were often of large extent and high number of polygons (which decreased certainty that each individual polygon was an actual meadow) but whose polygons were generally accurate. All layers were reprojected into the NAD 1983 California Teale Albers coordinate system. Each layer was then rasterized (retaining the confidence rank) and snapped to a 10m digital elevation model of the study area to serve as geographic control. Raster cells representing open water from the National Hydrography Dataset (NHD) were removed, and resulting cells with ranks were summed. Only cells with a summed rank greater than 2 were retained for further processing. A majority filter was run on the resulting remaining cells to reduce boundary heterogeneity, which replaced cell values based on the majority of the eight neighboring cells. Individual meadow polygons were created through a raster to vector conversion that treated all contiguous cells as a single part meadow feature with boundaries smoothed using the Polynomial Approximation with Exponential Kernel (PAEK; ArcGIS) method (20 m tolerance to reduce edge complexity). Polygons with an area less than 0.4 ha (less than 1 acre) were removed from the final meadow composite, and lastly all remaining meadow features were assigned a unique identifier and cross-walked to original source layers.

Additional Attributes

Geology : Dominant rock type was attributed to the meadow polygons based on available state geology layers. Using the Identity (Analysis) tool in ArcGIS, the most abundant lithology in the map unit (ROCK_1) was identified for each meadow.

Vegetation : The LANDFIRE dataset was used to attribute generalized vegetation (GROUPVEG) to the meadow polygons.

Soils : SSURGO datasets were compiled for the entirety of the study area. Gaps were filled with compiled STATSGO data. Soil attributes include: COKEY (component key), Kf (soil erodibility factor), Clay_tot_r (representative value of total clay), map unit key (MUKEY), component name (COMP_NAME). Components were assigned based on the soil component with the highest representative value in the map unit in which the meadow was located. For each component, the clay and Kf values from the top-most horizon were assigned to each meadow polygon. Flow Length : We ran the flow length geoprocessing tool across a state-wide 30m DEM then ran Zonal Statistics for each meadow on that raster. This tool measures the number of cells from each location to the nearest ridgetop. The flow range is the range value from the zonal statistics and reflects the range of distances to ridgetop across the meadow. Flow Slope : A slope metric utilizing the range of elevation values (the rise) and the Flow Range previously described (the run). Output is effectively Range (Elevation) / Range (Flow Length to ridgetop) for each meadow. Catchment Area : The approximate area of the upstream catchment exiting through the meadow. This is obtained by taking the highest value of a flow accumulation raster within the meadow and multiplying it by the raster's cell size. Not every meadow uses the same raster for calculations. Most meadows utilized the NHDPlus Version 1 30m Flow accumulation rasters and were automatically assigned to the appropriate raster based on geography. Due to an error in the Honey Lake area , meadows along or very near the Susan Rivers, the North Fork Feather River, and the Sacramento River downstream of the confluence with the Feather River used a separate, less accurate flow accumulation raster we developed from a California-wide 30m DEM.

Euclidean Distance : Using the Euclidean Distance (Spatial Analyst) tool in ArcGIS, the minimum distance to each meadow was calculated for NHD Springs/Seeps, NHD Streams/Rivers, CA lakes, and FS Topographic Transportation.

HGM Type : Assigned based on plot coordinates provided by Dave Weixelman (USFS) for 438 meadows.

Disclaimer : The State makes no claims, promises, or guarantees about the accuracy, completeness, reliability, or adequacy of these data and expressly disclaims liability for errors and omissions in these data. No warranty of any kind, implied, expressed, or statutory, including but not limited to the warranties of non-infringement of third party rights, title, merchantability, fitness for a particular purpose, and freedom from computer virus, is given with respect to these data.

This is an unofficial dataset compiled from the best available sources. The University of California, Davis makes no representations or warranties regarding the accuracy of data or maps. The user will not seek to hold the University of California, Davis liable under any circumstances for any damages with respect to any claim by the user or any third party on account of or arising from the use of data or maps. The user will cite the University of California, Davis as the original source of the data, but will clearly denote cases where the original data have been updated, modified, or in any way altered from the original condition. There are no restrictions on distribution of the data by users.

Sierra Nevada Multi-Source Meadow Polygons Compilation (v1.0) results are part of the Conservation Commons and are subject to a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License

Caveats The meadow polygons in this layer contain accuracy issues due to the variable quality of the input data. Individual input data layers were created using the following methods: heads-up digitization from aerial photographs (varying scales, photo years, etc.), extraction from coarse scale vegetation layers, Feature Analyst for ArcGIS, and field delineation using GPS devices (varying precision).