ArcGIS REST Services Directory
JSON

Layer: MORA_3m_polys (ID:1)

View In:   Map Viewer

Name: MORA_3m_polys

Display Field: Name

Type: Feature Layer

Geometry Type: esriGeometryPolygon

Description: The map is based on a vegetation classification developed during the project and was created using an inductive modeling approach. Data used to construct the classification were collected between 2005 and 2015, and included plots from Olympic National Park and North Cascades National Park Complex. These plots were used to develop and refine the association-level National Vegetation Classification (NVC). The associations were combined into map classes based roughly on the NVC alliance-level classification, but updated to allow improved map detail and accuracy. Model training data relied only on plots from MORA, collected during the same years. Independent field accuracy assessment data were collected in 2011, supplemented in 2014 and 2019, and applied to the final map generated later. The map development process was organized around the random forests machine learning algorithm. The modeling used 1,900 plots representing 124 vegetation associations and 37 map classes. Imagery from the National Agriculture Imagery Program and the Sentinel-2 and Landsat 8 satellites, airborne lidar bare earth and canopy height data, elevation data from the U.S. Geological Survey 3D Elevation Program, and climate normals from the PRISM Climate Group were used to develop a variety of predictor metrics. The predictors and the map class calls at each plot were input to a process in which each map class was modeled against every other map class in a factorial random forests scheme. We used the plot-level modeling outcomes and species composition data to adjust the crosswalk between association and map class so that floristic consistency and model accuracy were jointly optimized across all classes. The map was produced by predicting the factorial models and selecting the overall best-performing class at each 3-meter pixel.The final vegetation map, including a buffer surrounding the park, contains 33 natural vegetated classes, five mostly unvegetated natural classes, and four classes representing burned areas or anthropogenic disturbance. Coniferous forests and woodlands cover about three-fifths of the park. Upper montane forest codominated by silver fir (Abies amabilis), mountain hemlock (Tsugamertensiana) and/or Alaska-cedar (Callitropsis nootkatensis) is the most abundant forest zone by far. Lower montane forest dominated by silver fir and western hemlock (Tsuga heterophylla), and subalpine forest and woodland dominated by subalpine fir (Abies lasiocarpa) and mountain hemlock are about equally abundant. Lowland forest dominated by Douglas-fir (Pseudotsuga menziesii) and western hemlock is more limited, covering less than ten percent of the park. Each of the forest zones are found throughout the park in appropriate habitat, but subalpine types are most abundant in the northeastern park quadrant and lowland forests are associated primarily with the lower Ohanapecosh/Cowlitz, Carbon and Nisqually river valleys. Broadleaf and mixed forests occupy less than two percent of the park, mainly near major rivers, and often in an early successional state following disturbance by flooding. Shrublands cover nine percent, mostly as high-elevation mountain-heather, post-fire successional shrublands and tall shrubs in avalanche tracks. Herbaceous vegetation occupies just over five percent, mainly in lush subalpine and sparse alpine meadows. Sparsely vegetated and entirely bare rock, especially colluvial deposits, cover thirteen percent of the park, and exposed snow and ice occupy eight percent. Lake and river surfaces round out most of the remaining two percent. The accuracy assessment (AA) was based on 761 independent field-collected plots representing all the vegetated classes, as well as alluvial, colluvial and bedrock barrens, which also often host vegetation communities. They were gathered from an inference area covering 6.9% of the park. The overall map accuracy based on this sample was 86.9%. After correcting for map class prevalence in the inference area, the overall accuracy was 83.3%. Six of the 35 classes evaluated in the AA failed to meet the 80% NPS standard for user’s accuracy; seven fell short of the standard for producer’s accuracy. The AA discussion in the report contains a review of all classes failing to meet either standard, considers possible remedies for each, and provides recommendations to NPS for possible modifications of the map in response to the issues identified.Most NPS VMI maps have been produced by photo-interpretation (PI). We used model-based methods instead, because of the large size of the NCCN parks and the visual similarity of many of the key plant communities. Machine learning methods were used to extrapolate from a large set of classified field plots to the full extent of the park. The mapped vegetation units were 3-meter pixels rather than polygons, because we found that pixel-based modeling was the only reliable method for boundary detection between visually similar map classes. We invented and developed a variety of innovative image processing and modeling techniques to achieve finer spatial resolution and greater accuracy than is typical of model-based vegetation maps.

Copyright Text: North Coast Cascades Network (NCCN) of the National Park Service (NPS). Institute for Natural Resources (INR), Portland State University.

Min. Scale: 0

Max. Scale: 0

Default Visibility: true

Max Record Count: 1000

Supported query Formats: JSON, geoJSON, PBF

Use Standardized Queries: True

Extent:

Drawing Info:

HasZ: false

HasM: false

Has Attachments: false

Has Geometry Properties: true

HTML Popup Type: esriServerHTMLPopupTypeAsHTMLText

Object ID Field: OBJECTID

Unique ID Field:

Global ID Field: GlobalID

Type ID Field:

Fields:
Templates:

Is Data Versioned: false

Has Contingent Values: false

Supports Rollback On Failure Parameter: true

Last Edit Date: 8/1/2023 3:56:55 AM

Schema Last Edit Date: 8/1/2023 3:56:55 AM

Data Last Edit Date: 6/1/2022 9:34:01 PM

Supported Operations:   Query   Query Top Features   Query Analytic   Query Bins   Generate Renderer   Validate SQL   Get Estimates   ConvertFormat