PRISM BRIEFING BOOK AND QUESTIONNAIRE
A Description of the
PRISM Model for Spatially
Distributing Observed Precipitation
Presented by the
Natural Resources Conservation Service
Water and Climate Center
in partnership with
Oregon Climate Service
Oregon State University
Updated April 1998
Table of Contents
1.0 Purpose of this Briefing Book
This briefing book is designed to provide potential PRISM users with the latest description of the PRISM model for spatially distributing observed precipitation. Products produced by the PRISM model are completely compatible with a wide variety of GIS environments and can be the basis for a number of cartographic products.
2.0 Frequently Asked Questions About PRISM
The following questions provide a general overview of the PRISM project and are intended for general distribution.
2.1 What is PRISM?
The acronym PRISM stands for "Parameter-elevation Regressions on Independent Slopes Model." PRISM is an analytical model that distributes point measurements of monthly, seasonal and annual precipitation to a geographic grid. The current grid size used to represent basic PRISM derived products is 4 kilometers by 4 kilometers (km). However, through a PRISM compatible resampling algorithm, estimates of precipitation can be made to a resolution of approximately 1 kilometer by 1 kilometer.
PRISM-derived products are produced in a GIS-compatible latitude-longitude grid or a gridded map projection such as Albers Equal Area (AEA).
2.2 Who is performing the work?
Oregon State University, through the Oregon Climate Service (OCS), is performing the work under a cooperative agreement with the Natural Resources Conservation Service (NRCS) Water and Climate Center (WCC).
2.3 What Digital Elevation Model (DEM) is used to create the maps?
PRISM uses a digital elevation model (DEM) that contains information describing the earth's topography (slopes, aspects, elevations) and observed precipitation measurements to determine variations in precipitation as functions of elevations. Orographic precipitation and rain shadows are modeled in PRISM in a unique and accurate way.
The PRISM DEM is derived from a 3-arc second Defense Mapping Agency (DMA) digital terrain elevation data (DTED) obtained from the EROS Data Center. In general, DMA absolute horizontal accuracy specification is 130 meters circular with 90% probability, and an absolute vertical accuracy (feature to mean sea level) of +/- 30 meters linear error at 90% probability.
The EROS Data Center mosaiced the 3-arc second data into one large U.S. coverage. DEM verification was made by running basin delineation software. The data was then resampled at 15-arc seconds by sampling every 5th point. Finally, the 15-arc second data were resampled by Oregon Climate Service using a Gaussian filter to produce a 2.5-arc minute dataset. Each raster value represents the elevation over a given area and is more representative of areal elevation than box averages or point sampling.
2.4 What climate datasets are used to create the maps?
The precipitation data PRISM uses to distribute precipitation consists of three basic climate datasets:
1. The National Climatic Data Center 1961-1990 normals dataset (CLIM-81) observed by the National Weather Service Cooperative Climate Network;
2. The NRCS SNOTEL (SNOwpack TELemetry) network dataset;
3. Supplemental datasets submitted by the individual State Climatologists or Regional Climate Centers.
2.5 Who has reviewed the PRISM model?
The PRISM model has been evaluated and endorsed by the PRISM Evaluation Group (PEG) as a valid technique to distribute observed precipitation in the orographic West. PEG is composed of State and Regional Climatologists, representatives of national agencies, NRCS representatives and other potential PRISM users such as state and local governments. The PRISM model evaluation was performed in Idaho, Nevada, Oregon, and Utah. Each map was reviewed by PEG representatives as part of the evaluation process.
PRISM has been carefully scrutinized by the greater scientific community through more than 15 symposium presentations as well as publication in a referred journal (Journal of Applied Meteorology, 1994). Preliminary use of PRISM in the eastern United States has shown it to render very reasonable precipitation maps. A synoposis is available here.
2.6 What digital layers are available now?
Hard copy and digital precipitation for the 1961-1990 period is available in a variety of formats, scales, and map projections. Temperature data will be available by the end of 1997.
2.7 Can I run PRISM on my computer system?
At present there are no plans to distribute the software, since it undergoes frequent modification and requires significant technical support (which we are unable to provide).
PRISM digital output, however, is quite plentiful, and is easily viewed using a variety software such as GRASS, Arc/Info, Arc/View, WinGIF, and others in either a UNIX, DOS/Windows or Macintosh environment.
2.8 For further information about PRISM contact:
PRISM Project Plan:
Water and Climate Center
101 SW Main, Suite 1600
Portland, OR 97204-3225
George Taylor, State Climatologist
Oregon Climate Service
316 Strand Ag. Hall
Corvallis, OR 97331-2209
Oregon State University
USFS Forestry Sciences Lab
3200 SW Jefferson Way
Corvallis, OR 97331
Oregon Climate Service
326 Strand Ag. Hall
Corvallis, OR 97331-2209
3.0 Examples of PRISM Output
A wide variety of graphic products are included with this briefing book to give prospective users an opportunity to consider PRISM-derived products to satisfy their needs for spatially distributed precipitation. The scales shown progress from large to small to meet individual needs. The basic raster (grid cell) output can be manipulated by a wide variety of applications to meet either graphic or cartographic needs.
3.1 United States Map (Figure 1.)
This map was generated at a 2.5-minute (~4km) grid interval from the National Climatic Data Center (NCDC) CLIM-81 normals dataset, Natural Resources Service (NRCS) SNOTEL (SNOwpack TELemetry) network dataset and datasets contributed by State and Regional Climatologists.
The raster plot clearly captures the isoheytal features evident in previous national precipitation maps. These features include the Appalachians, Adirondacks, the meridional patterns of the Midwest, and the complex orographic patterns of the West.
3.2 Western US Map (Figure 2.)
This regional map was generated at a 2.5-minute (~4km) grid interval using the same datasets as the United States map. This product "knits" together each state's climate dataset into a single seamless climate layer. The map clearly shows the well-known precipitation patterns of the western US, including the wettest and driest areas in the continental US (the Olympic Peninsula and Mojave Desert, respectively).
3.3 Oregon Precipitation (Figure 3)
This state map was generated at a 2.5-minute (~4km) grid interval using the NCDC and NRCS datasets. This particular graphic provides general information about the precipitation pattern for Oregon.
3.4 Indiana Precipitation (Figure 4.)
This state map was generated at a 2.5-minute (~4km) grid interval using the NCDC and NRCS datasets. This graphic represents the actual output from the PRISM model, which is gridded (raster) precipitation values from each grid cell. County boundaries have been overlaid in this example. Other digitized attributes (cities, rivers, roads, etc.) may be added to enhance readability.
4.0 The PRISM Evaluation Group (PEG)
PEG was formed to determine if the PRISM technique was a valid method to distribute precipitation in the orographic western US. The group was composed of State and Regional Climatologists, a National Climatic Data Center representative, a National Weather Service representative, Natural Resource Conservation Service engineers, hydrologists, meteorologist and GIS experts.
The group met three times formally and twice informally to evaluate the PRISM technique, determine user needs and evaluate existing PRISM products. After evaluating Idaho, Montana, Nevada, Oregon, and Utah, the group has endorsed PRISM as a valid technique to distribute observed precipitation in the orographic West.
5.0 Professional Publication Describing PRISM
A paper describing the PRISM technique was peer reviewed and published in the February 1994 issue of JAM. The article is titled "A Statistical-Topographic Model for Mapping Climatological Precipitation over Mountainous Terrain". A synopsis is available for review here.
The modeling concepts described in this paper have evolved significantly since publication in response to the PEG process. However, the basic theory has not changed and has improved to provide a more representative spatial display of average precipitation.
6.0 Metadata Information Describing Digital Elevation Model (DEM)
Since the underlying philosophy of the PRISM technique is based on topography, it was very important that a clear DEM definition be included as part of the PRISM metadata. The United State Geological Survey recently published a Spatial Data Transfer Standard (SDTS) as a FIPS 173 document. This standard for spatial data has been adopted for the PRISM project.
7.0 State Climatologist and Regional Climate Center Roles and Responsibilities
Since the primary goal of the PRISM project is to produce updated annual and monthly precipitation maps (1961-1990) for each state, it is very important that the PRISM Project members understand your needs and level of interest in this project. The State Climatologists and Regional Climate Centers play very important roles in the entire PRISM process.
With this in mind and based on the project experiences in the West, the following roles and responsibilities are seen as important for the State Climatologist and Regional Climate Centers.
7.1 State Climatologist Roles and Responsibilities
a. Play a significant role in the review process, including assessment of the technical content of the map.
b. Supply appropriate supplemental datasets.
c. Coordinate external review among participating agencies within the state.
d. Coordinate funding within the state.
e. Understand and support the technology used to create the maps and GIS layers.
7.2 Regional Climate Center Roles and Responsibilities
a. Provide technical review of PRISM from a regional perspective.
b. Coordinate dataset contributions from State Climatologists within their region.
c. Coordinate multi-state and watershed boundary funding within their region.
d. Cost-share funding of State Climatologists participation in PRISM coordination meetings as appropriate and available.
e. Mediate technical content of map is conflicts arise.
f. Understand and support the technology used to create the maps and GIS layers.
The purpose of the PRISM project was to develop an analytical model for distributing point measurements of observed monthly and annual precipitation to regularly spaced grid cells at regional and continental scales. The main focus was developing a conceptual framework for mapping orographic precipitation in complex terrain. The result was an objective precipitation distribution model called PRISM. PRISM brings a unique combination of climatological and statistical concepts to the mapping of orographic precipitation.
The PRISM-derived raster values for annual average precipitation have be mapped into an Albers equal area projection and are congruent with various NRCS datasets such as STATSGO. A wide variety of GRASS GIS functions are now fully supported when using the PRISM-derived 2km grid of average annual precipitation.