Source: WASHINGTON STATE UNIVERSITY submitted to
SOIL SURVEY AND CLASSIFICATION
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
0094342
Grant No.
(N/A)
Project No.
WNP00900
Proposal No.
(N/A)
Multistate No.
(N/A)
Program Code
(N/A)
Project Start Date
Jan 1, 2006
Project End Date
Dec 31, 2009
Grant Year
(N/A)
Project Director
Frazier, B. E.
Recipient Organization
WASHINGTON STATE UNIVERSITY
(N/A)
PULLMAN,WA 99164
Performing Department
CROP & SOIL SCIENCES
Non Technical Summary
The Agricultural Experiment Stations have been integral since the beginning of the soil survey program in 1899 as members of the National Cooperative Soil Survey (NCSS). The NCSS is a cooperative venture among state and federal agencies and native American nations that are responsible for resource management and Land Grant universities. Washington State has large areas of designated wilderness, as do other western states. In these inaccessible areas, new methods must be developed and evaluated to make soil resource inventories that rely more on remote sensing and environmental modeling and less on intensive ground traverses. We also must design soil interpretations that account for the unconventional uses and impacts that wilderness areas receive compared to non-wilderness and private lands. The purpose of this project proposal is to continue the university contribution to soil survey programs in the state of Washington through assistance to the soil survey effort and through basic research on the properties and genesis of soils in the state. In addition, a major new thrust of this proposal is to develop, test, and implement new and experimental methods of soil resource assessment in remote areas where conventional survey procedures are not feasible.In a state like Washington, the vital agricultural and forest industries and the non-agricultural users of soils are strongly dependent on a productive and sustainable soil resource. All constituent groups in the state potentially will benefit from the research and assistance offered in this proposal.
Animal Health Component
40%
Research Effort Categories
Basic
10%
Applied
40%
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1010199206020%
1010199206110%
1010199208020%
1017299206020%
1017299206110%
1017299208020%
Goals / Objectives
1. Contribute to the National Cooperative Soil Survey under the terms of the Memorandum of Understanding signed in 1953 between the WSU Agricultural Experiment Station and the USDA-NRCS, including technical field assistance to soil surveys and aid in establishing priorities and procedures for projects. 2. Conduct field and laboratory research on the nature and properties of the soils of the state and region. In the coming period, specifically, develop, test, and implement new procedures to predict soil distribution, reliant on remotely sensed data with limited ground transecting, for remote areas of the state. 3. Conduct field and laboratory research to predict soil distribution to upgrade detailed soil surveys. 4. Update the Washington State Soil Map in regard to legend and polygons. 5. Publish results and recommendations in the form of soil survey reports, maps, research papers, research center bulletins, and handbooks.
Project Methods
1. Contribute to the NCSS program: A. Serve as chair every other year at the annual Washington Interagency Soil Survey Work Planning Conference, which is to establish priorities for soil survey projects, plan survey operations, and design map units; B.Represent the University and State interests in the Western Regional Soil Survey Conference and on WRCC-93 (currently under revision) and conduct research accordingly; C. Attend soil survey field reviews, final reviews, correlation conferences, and regional and national soil survey conferences and serve on technical advisory committees to these conferences to make policy recommendations at the regional and national levels; D. The PI of this project will continue to serve on the State Technical Advisory Committee to NRCS for as long as appropriate at the discretion of the Dean of CAHNRS. 2. Conduct research: A. Develop, test, and implement new soil survey procedures, reliant on remotely sensed data, with limited ground transecting, for remote areas. An MOU is in place jointly among the U.S. Forest Service, the USDA-NRCS, and Washington State University. A work plan and research agreement is in place with the National Park Service. B. Describe and sample soils on the basis of well designed sequences, transects, or other experimental plans. Analyze physical, chemical, and biological properties to answer specific questions that arise during soil survey or other research; i.e., contribute to a better understanding of soil resources; C. Determine the geographic order of soils by relating soil properties to landscape features such as land form, drainage system, plant community, and other environmental attributes, especially those that are consistently discernible by remote sensing methods. Apply new geostatistical techniques to derive the geographic order of soils on remote landscapes. D. Incorporate traditional soil survey information into a GIS along with remotely sensed data and digital elevation data to provide an alternative information distribution system to traditional printed books. Make soil resource information for the state more readily available through services on the Internet. 3. Conduct field and laboratory research to predict soil distribution to upgrade detailed soil surveys. This research will be conducted at the Cunningham Agronomy Farm where 180 soil cores have been collected, described, and entered into a data base. Methods of modeling soil distribution will be tested using this extensive data base. 4. Updating the Washington State Soil Map in regard to legend and polygons will be accomplished by meeting with NRCS and searching the NRCS data base to update classification of predominant soil series and examine polygon boundaries in light of new detailed survey information. Models developed by projects conducted in the north Cascades will be used to predict new polygons for the state map in that area. 5. Publish results and recommendations in the form of digital soil surveys, maps, and research papers. Digital formats and online journals will be emphasized to reduce costs associated with publication in color.

Progress 01/01/06 to 12/31/09

Outputs
OUTPUTS: Wilderness areas in Washington State have historically been excluded from soil resource inventories due to the huge investment of time and resources required to map them. Yet, they are ideal places to study soil formation and distribution under natural conditions. Computer-based models are an efficient way to predict occurrence of soils. For several years we have worked on a modeling process known as the Remote Area Soil Proxy (RASP) model. In addition to development of the process, one of the research goals was to produce an improved soil map of Thunder Creek watershed (TCW) of North Cascades National Park based on new vegetation (climate proxy) data generated from 15m ASTER data and 10m DEM data. A project in 2004 conducted by C.A.D. Briggs had produced a soil map based on 30m LANDSAT derived vegetation data and 30m DEM data that was used for comparison. The revised soil map incorporating the new vegetation raster and using a 10 m DEM was created with the Model Builder functions contained within ArcGIS v. 9.3.1. A toolbox containing the model is included with the data sets. In addition to ArcGIS (ArcInfo level), the model requires the Spatial Analyst Extension and the TauDEM Extension. The latter is available from (http://hydrology.neng.usu.edu/taudem/). The model allows for easy substitution of any appropriate elevation, vegetation, or landform dataset as inputs. The RASP model was also implemented to compare both 30 m and 10 m DEMs with no other data being changed. The same decision tree nodes were used following the expert knowledge employed by Briggs only using higher resolution DEM input data. The resulting soil maps were evaluated with field observation of soil pits and applying soil taxonomy at multiple levels: orders, sub-orders, great groups and sub-groups. PARTICIPANTS: Dr. Bruce E. Frazier, PI Dr. Richard Rupp, Technical Assistant, GIS Phil Roberts, Graduate Student trained on the project Emily Meirik, Graduate Student trained on the project Collaborators: Toby Rodgers, USDA-NRCS Crystal Briggs, USDA-NRCS Dr. Jon L. Riedel, USDI-Park Service Dr. David J. Brown, Faculty Advisor, WSU Dr. Paul Gessler, Faculty Advisor, UI TARGET AUDIENCES: USDA-NRCS USDA-FS USDI-NPS PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
A process based on soil forming factors is defined that predicts soil occurrence for remote areas. Critical data layers are determined by field examination of soils in the area to establish dominant soil forming factors. A new soil map of TCW is presented based on vegetative cover derived from ASTER data, 10m DEM, and landform data. 75% of the new soil map pixels did not change. These unchanging pixels represent the large areas of conifer forest and deciduous shrubs.The largest changes occurred in soil map unit Typic Melanocryands-Vitrandic Dystrocrepts. Only 21% of its pixels were unchanged. 57% became Typic Vitricryands-Andic Haplocryods,13% went to Lithic Dystrocrepts-Typic Dystrocrepts. Typic Melanudands-Andic Dystrudepts had 40% agreement with the new map, 47% changed to Typic Udivitrands-Andic Haplorthods. Humic Dystrocryepts-Typic Dystrocryepts had 38% agreement, 33% changed to Typic Dystrocryepts-Vitrandic Dystrocryepts and 16% changed to Lithic Dystrocryepts-Typic Dystrocryepts. All of these changes reflect a reclassification of vegetation that produces dark surface horizons, e.g., meadow/deciduous shrubs, to classes of vegetation that do not, e.g., heather. Comparison of the vegetation maps shows that Forest Agreement was 74% with 16% going to Heather on the new map; Meadow/Shrub Agreement was 19%, 16% went to Forest, 42% to Heather, and 18% to Rock. Rock Agreement was 41%, with 33% going to Heather and 21% to Ice. Smaller pixels (15m) of the new vegetation map allowed smaller areas of Heather to be separated from its usual surroundings of Rock, Meadow and Forest. Further study of another objective to determine soils associated with defined landforms shows the following results: Observations of landforms were compared with soil properties at soil pit locations. Graphical representations of categorical soil variables show strong relationships with landscape stability and profile development. Older landforms support Spodosols and Andisols while younger, active surfaces support Entisols and Inceptisols. Backslopes and footslopes are stable except for small areas where mass movements have occurred. These trends are evident when comparing podsolization, tephra distribution, and presence of redoxomorphic features to landform classes. Podsolization is found on backslopes and footslopes, wetness occurs on toeslopes where runoff accumulates. Field descriptions of pedons on landforms were assigned series designations and correlated to higher taxa. Backslopes are Andisols (37%), Spodosols (39%), and Inceptisols(24%). Footslopes are occupied equally by the same three Orders. The majority of soils on toeslopes are Inceptisols (52%) followed by Spodosols (24%), Entisols (14%), and Andisols (10%). Soils on areas of mass movements are predominantly Inceptisols (71%) followed by Spodosols (14%) and Andisols (14%). As expected, finer levels of soil classification (sub-groups) show a decrease in accuracy when compared with broader taxa (orders). For every taxonomic level, the soil map produced from a 10 m DEM was 3 - 5% more correlated with field observations. Two students graduated with MS degrees in Soil Science from this project.

Publications

  • Frazier, B.E., T.M. Rodgers, C.A. Briggs, and R.A. Rupp. 2009. Remote area soil proxy modeling technique. Soil Survey Horizons, Vol. 50, No. 2:62-67.
  • Meirik, E.E., 2008. Mapping vegetation with remote sensing and analysis of soil-vegetation interactions in North Cascades National Park, WA. MS Project Report, Washington State University, 90p.
  • Roberts, P.H. 2009. Digital landform mapping and soil-landform relationships in the North Cascades National Park, Washington. MS thesis, Washington State University, 134p.


Progress 01/01/08 to 12/31/08

Outputs
OUTPUTS: Wilderness areas in Washington State have historically been excluded from soil resource inventories due to the huge investment of time and resources required to map them. Yet, they are ideal places to study soil formation and distribution under natural conditions and the data are important for planning decisions. Computer-based models are an efficient way to predict occurrence of soils. One of the research goals is to discern how landforms and soils are related in Thunder Creek watershed (TCW) of North Cascades National Park. Can landforms be predicted from DEM data or are soils predictable directly from DEM data During the 2008 field season, 15 clusters and a total of 226 GPS locations were recorded throughout the study area. These points complement the existing data set of 227 points, bringing the total number of observations in TCW to 453. Our digital database of field observations has been updated to include all new information. In 2008, approximately 180 soil samples were collected from genetic horizons of the described profiles. Current research is focused on inductive statistical classifications of DEM data to determine landforms and on the properties of the soils. Initial steps of landform mapping were to build classification models using field observations to train and evaluate the model predictions. This has been completed using the recursive partitioning and random forest algorithms in the R statistical software package. The recursive partitioning algorithm builds binary decision trees (BDT) to classify a response variable with multiple predicting variables. Random forest (RF) is a form of machine learning in which classification trees are iteratively fit to random sub-samples of the calibration data set. Modeled data are compared to National Park Service (NPS) geologists' maps of landforms within TCW using traditional methods of aerial photo interpretation and topographic maps validated through field observations. We have analyzed mineral phases present in eight soil samples from Thunder Creek Watershed. Samples were chosen based on reflectance in the visible and near-infrared regions of the electromagnetic, 350-2500nm, spectrum. (Soil reflectance was measured using an ASD Inc. Agrispec 2-8nm spectral resolution). The resulting data were used to obtain a digital Munsell color value, which was added to the geodatabase of existing field and laboratory data. Spectral Profiles were grouped into 5 nm bands and principal component analysis was used. Reflectance profiles were clustered using partitioning around medoids on the Mahalanobis distances of the first derivative of reflectance. All organic soil horizons (Oi, Oe) were excluded from the statistical selection methods. PARTICIPANTS: Phil Roberts, Emily Meirik, Bruce Frazier, David Brown, Ross Bricklemyer, Richard Rupp, Crystal Briggs, Toby Rodgers. TARGET AUDIENCES: Washington State Soil Survey (USDA-NRCS), National Park Service, USDA Forest Service. PROJECT MODIFICATIONS: Project extension granted to 12/31/09

Impacts
Field observations of landforms correlated well with the expert landform map (kappa = 0.59 and overall accuracy = 70 %). Of the two automated approaches, the random forest classification (kappa = 0.44 and overall accuracy = 59 %) performed better than the binary decision tree model (kappa = 0.37 and overall accuracy = 53 %). Resulting statistical models trained with field data have been applied to watershed mapping. An error matrix was produced comparing the NPS-expert classification to the automated classifications on a 10 m grid covering TCW. The overall accuracy is 58% for the entire watershed comparison using BDT and 61% using RF models. These maps indicate good agreement between NPS-Expert maps and two automated methods. These results are encouraging and show that automated landform mapping using remotely sensed terrain information is a viable method for analyzing geomorphic features in TCW. The Al and Fe concentrations suggest podsolization is the dominant pedogenic process. Sample 7 (Bs) has a Al + one-half Fe value of 2.59 and when compared to sample 8 (E) has over a 15-fold increase in Al + one-half Fe. These results indicate the soils contain spodic materials and classify as Spodosols.

Publications

  • No publications reported this period


Progress 01/01/07 to 12/31/07

Outputs
OUTPUTS: Wilderness areas in Washington State have historically been excluded from soil resource inventories due to the huge investment of time and resources required to map them. Yet, they are ideal places to study soil formation and distribution under natural conditions and the data are important for planning decisions. Computer-based models are an efficient way to predict occurrence of soils. One of the research goals is to discern how landforms and soils are related in Thunder Creek watershed (TCW) of the North Cascades National Park. Can landforms be predicted from DEM data or are soils predictable directly from DEM data? Among the options for improving the RASP (Remote Access Soil Proxy) model output is to increase the resolution of the elevation data in the Elevation Submodel. Ground observations have been made and data collected and located horizontally and vertically with GPS units. A rigid sampling scheme was followed that included detailed descriptions of soil, vegetation and geomorphology for each site visited. Potential sample locations were pre-selected based on areas of maximum variation in landform type and vegetative cover to assure a wide range of site characteristics and environmental combinations. One hundred potential locations were selected and georeferenced, each of which represents the center of 10-sample cluster. Ten clusters dispersed throughout the TCW were sampled during the summer of 2007. Soil pedons were described and recorded on a NRCS 232 form. Vegetation forms included detailed description of overstory and understory with relative abundance and percent cover of each species present. Geomorphology data included slope, aspect, slope complexity, landform type, and landform position. A total of 70 soil, 101 vegetation and 104 geomorphology descriptions were completed. Vertical accuracy of the four DEMs were analyzed by comparing GPS elevations to DEM values at 135 locations. While R squares were good for all DEMs, RMSE values for NED_10m were 4 times larger than NED_30m, SRTM 30m, or ASTER derived 30m. Further investigation of these relationships is warranted. PARTICIPANTS: Bruce Frazier, David Brown, Phil Roberts, Emily Meirik, Richard Rupp, Toby Rodgers, and Crystal Briggs TARGET AUDIENCES: USDA-NRCS, USDA-FS and USDI-NPS personnel and other users of the parks

Impacts
The impact of these findings to the modeling of soil distribution is that the more detailed NED_10m DEM may not be as useful as anticipated. The current sample set suggests that NED_30m is best to predict slope. Other terrain characteristics remain to be examined.

Publications

  • No publications reported this period


Progress 01/01/06 to 12/31/06

Outputs
Wilderness areas in Washington State have historically been excluded from soil resource inventories due to the huge investment of time and resources required to map them. Yet, they are ideal places to study soil formation and distribution under natural conditions and the data are important for planning decisions. Computer-based models are an efficient way to predict occurrence of soils. Models lessen the need for intensive field transecting compared to traditional soil surveys and offer a cost-efficient alternative to traditional cartographic techniques. We have used ArcGIS software to develop the Remote Area Soil Proxy (RASP) modeling process. It yields soil distribution models that are about fourth-order in detail with map units of complexes of soils classified at the Suborder level of Soil Taxonomy. Now we are investigating ways of improving on the model results. One of the research goals is to discern more information out of the area classified as valley wall landform in Thunder Creek watershed of North Cascades National Park. This is an immense formation and covers approximately 42% of the watershed (approximately 12,780 ha). Among the options for improving model output is to increase the resolution of the elevation data in the Elevation Submodel. The valley wall landform was isolated and DEM resolutions of 30m and 10m were compared using a suite of derived terrain attributes. Elevation, slope, aspect of slope, profile curvature, plan curvature, flow accumulation and compound topographic index (CTI) were calculated over the landform at both resolutions. Zonal Statistics were used to compare values over the different resolution sizes. When looking at the range of values for terrain attributes, the DEM with higher resolution generally produced a greater range in values. The CTI range increased from 2 - 14 to 0.4 - 17 while the mean decreased from 5.45 to 4.66. The ability to resolve smaller topographic features leads to an ability to characterize the landscape more precisely. This is especially true for profile curvature (rate of change of slope perpendicular to the contour), which predicts less concavity with higher resolution DEM. This may mean that areas of lithic soils are more abundant than previously mapped, since they tend to occur on convexities. Plan curvature remains the same. Relatively little difference is observable when slope, aspect and elevation are compared with DEM resolution.

Impacts
Impacts are largely unknown at this time since we have not completed the model or the research at this point. Overall impacts are expected to improve detail output of the RASP model which will improve our knowledge of soils in North Cascades National Park.

Publications

  • No publications reported this period


Progress 01/01/05 to 12/31/05

Outputs
We conducted one week of intensive field work to describe and sample soils in the Columbia Plateau and Cascade Range for a week-long field tour for the uncoming quadrennial meeting in July, 2006 in Philadephia, PA of the International Union of Soil Science. We collected bulk samples for physical and chemical characterization, for bulk density, and for soil microscopy and submitted them for analysis to the National Laboratory of USDA-NRCS. We held negotiations with the National Geospatial Development Center of USDA-NRCS to develop continued funding for FY 2006-2009 for graduate student education and continuing development of the RASP (Remote Area Soil Proxy) model

Impacts
The RASP model, developed at WSU between 1998 and 2004 is a technology to map soil distribution (make soil surveys) of remote areas such as wilderness areas and national parks. The USDA-NRCS, which is the program lead for all soil surveys in the U.S., has become very interested in the RASP model and incorporating it into a 'toolkit' for advanced soil survey methods.

Publications

  • Frazier, B.E., T.M. Rodgers, and R. Rupp. 2005. Modeling Soil Distribution in Wilderness Areas of Washington State. Proceeding ESRI International User Conference, San Diego, CA, paper 1862. http://gis.esri.com/library/userconf/proc05/papers/pap1862.pdf


Progress 01/01/04 to 12/31/04

Outputs
Wilderness areas in the Western United States have been excluded from traditional soil surveys due to issues of inaccessibility and a lack of urgency in acquiring this data. This report includes two studies where modeling has been applied to derive spatial knowledge of the soils in the Cascade Range of Washington State. Thunder Creek watershed (TCW) in North Cascades National Park (NOCA), and North Fork of the Skykomish River watershed (NFSky) along the western slope of the Cascades provide complex and unique environments in which to study soil formation. We sampled pedons on major combinations of landforms, parent materials, and vegetation types within these watersheds to determine the dominant pedogenic processes and the soil-forming factors that control them. Soils have developed in a mixture of parent materials derived from regional glacial and volcanic activity that have been continuously reworked and redistributed through disturbance. The soil moisture regime is udic throughout with minor areas of aquic regime in the valley bottoms and isolated depressions. The soil temperature regime is frigid up to approximately 1000 m and is cryic above that elevation. The vegetation is predominantly Pseudotsuga menziesii and Tsuga heterophylla forests at lower elevations and Abies amabilis above 1000 m to about 1800 m. From about 1800 m up to the elevation of permanent snowfields and glaciers, subalpine meadows consist of a mosaic of herbaceous cover and islands of Tsuga mertensiana and Abies lasiocarpa. Mixtures of Sorbus sitchensis and Alnus sinuata or herbaceous meadows cover active landscapes and avalanche chutes at higher elevations. Since glacial retreat, the watersheds have been blanketed by volcanic tephra originating from Mount Mazama, Glacier Peak, and Mount St. Helens. Based on field morphology and lab analysis we found that the dominant soils in TCW are andic subgroups of Spodosols and Andisols in approximately equal abundance, with lesser areas of Inceptisols and some Entisols. This information along with extractive chemistry allows us to conclude that true Spodosols are widespread in TCW and NFSky and have pedogenic albic and spodic horizons. No pseudo-albic horizons of white, unweathered tephra layers were found. The cold, wet, and dominantly conifer-covered landscape supports podzolization in tephra-rich materials on parts of landforms that have been stable for several thousand years. Andisols are forming in drier microclimates, on gravitationally active landscapes, and under herbaceous vegetation. Inceptisols dominate landforms controlled by water erosion, where tephra has not been preserved in the soil and on active avalanche chutes. Entisols were found only on active or recent floodplains. The distribution of each soil order corresponds to the distribution and preservation of volcanic tephra, vegetation type, and stability of the landform. Likewise, in the NFSky albic horizons forming within tephra layers have been stripped of their chemical properties required for andic classification through the process of podzolization.

Impacts
These studies provide critical documentation with field and lab data of the separation between Andisols and Spodosols in the Cascade Range of central Washington.

Publications

  • No publications reported this period


Progress 01/01/03 to 12/31/03

Outputs
Wilderness areas have historically been excluded from soil inventories due to the high cost and logistics of mapping inaccessible terrain. We are using GIS, remote sensing technology, and a focused effort to describe soils in the field to model and map the distribution of soils in the 29,000 ha Thunder Creek watershed in the North Cascades NP. GIS data layers such as current vegetation from processed remote sensing imagery, a 10 m DEM, and surficial geology serve as proxies for the soil-forming factors that control pedological processes. In the 2002 and 2003 field seasons, 120 pedons were observed and described to represent the major combinations of soil-forming factors. Thirty-five pedons were analyzed and classified: 17 of these were Spodosols (Andic Haplocryods and Andic Haplorthods), 12 were Andisols (Haplocryands and Hapludands), 4 Inceptisols, and 2 Entisols. The surficial geology layer was beneficial in predicting the soil properties controlled by parent material age and slope stability, whereas current vegetation helped distinguish between Spodosols and Andisols. Spodosols dominate on the majority of conifer-covered glacial-age materials; Andisols form on younger and more active erosional and depositional landforms. This is a demonstration project for future mapping of the entire 277,150 ha park. To solve a similar problem of having to map extensive areas, the USDA-FS has implemented a hierarchical framework for mapping landscapes. The landscape is stratified into progressively smaller units for ecosystem management purposes. The hierarchy is made up of various planning and analysis scales. In theory, the finer scale mapping units should nest within broader scale mapping units. An example of land unit mapping is an Order 3 soil survey, while a Landtype Association (LTA) map covers a broader area and is at a landscape-scale within the hierarchy. If the framework works correctly, a LTA map should be able to stratify soils on a broad scale, and could be used as a premap for a finer scale Order 3 soil survey. Using a remote sensing software package (Terrestrial Ecosystem Unit Inventory Geospatial Toolkit) provided by the Forest Service Remote Sensing Application Center, a LTA map was generated for a watershed in the Mount Baker-Snoqualmie NF, and soils were stratified to the subgroup level based on LTA map unit information. The area was then field checked to verify the projected soils. A LTA map can be produced cheaply and quickly using available digital forms of data, helping to speed up soil survey premapping procedures.

Impacts
These projects are developing methods to model soil distributions where access is limited by terrain, but adverse impacts to ecosystems must be avoided.

Publications

  • No publications reported this period


Progress 01/01/02 to 12/31/02

Outputs
Our program to predict soils in inaccessible wilderness areas is being expanded to also include adjoining park lands. We are working with the Park Service to create a model of soils in Thunder Creek basin of North Cascades National Park. Available data include a vegetation layer derived from Landsat, digital elevation models from USGS and Shuttle radar, land type association from the US Forest Service and glacial landforms from the Park Service. Graduate student Crystal Briggs spent ten weeks in this wilderness watershed analyzing landforms and describing soil pedons to charaterize the soil types associated with different landforms. Sixty soil pedons have been sampled to establish soil development relationships to landscape parameters. Laboratory characterization of the samples is underway and GIS models of landscape distribution of soils are being designed.

Impacts
Our ability to predict soils in inaccessible areas is critical to planning of recreational uses of those areas. This project will yield maps to facilitate planning and contribute to our knowledge of the area.

Publications

  • No publications reported this period


Progress 01/01/01 to 12/31/01

Outputs
Field work and model development continued for the remaining area of the Pasayten Wilderness area being mapped by GIS-based soil landscape analysis. Laboratory analysis is underway to identify airfall volcanic tephra layers that have contributed to soil development in the wilderness and to determine whether the pattern of occurrence of Andisols and Spodosols in these glaciated valleys is a result of climatic or vegetational control on Spodosol processes in tephra rich materials. Two manuscripts are in development from this research. A soil and landscape database has been developed for the Cunningham Research Farm. Thiessen polygons based on 184 soil cores are used to represent the highly variable distribution of 10 soils over 38 ha of the Cunningham Farm. Distribution of the soils is not well predicted from topographic variables. Mitigating factors of the prediction include agricultural erosion and presence of a pre-Holocene soil surface within 150 cm of the current surface.

Impacts
New methods of conducting soil surveys are needed in areas such as wilderness, where access is limited and field mapping of soils by traditional techniques is prohibitively expensive. This project is utilizing a GIS-based soil landscape partitioning into map units based on limited field work, to produce an order 4 soil survey for the Sawtooth and Pasayten Wilderness areas at low cost.

Publications

  • Rodgers, T.M., B.E. Frazier, A.J. Busacca, and P.E. Gessler. 2001. Towards the Implementation of Automated GIS Soil Mapping Techniques. [Abstract]. Western Society of Soil Science Annual Meetings, 18-20 June, 2001. Moscow, Idaho.
  • Wardell, M., B. Frazier, A. Busacca, and D. Huggins. 2001. Digital Terrain Analysis of the Cunningham Farm, Pullman, WA, USA. [Abstract]. Western Society of Soil Science Annual Meetings, 18-20 June, 2001. Moscow, Idaho.
  • Wardell, M., B.E. Frazier, and A.J. Busacca. 2001. Digital terrain analysis of a loessial soil landscape. Agron. Abst., s05-frazier182359-P. Am. Soc. Agron., Madison, WI
  • Wardell, M., B.E. Frazier, and A.J. Busacca. 2001. Digital terrain analysis of a loessial soil landscape. Agron. Abst., s05-frazier182359-P. Am. Soc. Agron., Madison, WI


Progress 01/01/00 to 12/31/00

Outputs
During the 2000 calendar year, more than 150,000 acres of the Pasayten Wilderness was mapped using a combination of field work and GIS-based soil-landscape modeling. Pedon descriptions were made for about 80 soils. Maps were developed for the Sawtooth Wilderness and Pasayten watershed portions of the Pasayten Wilderness using a decision-tree supervised classification in a GIS.

Impacts
The current focus is the project in the wilderness areas of the Okanogan Highlands of Northcentral Washington. In this project, methods are being developed and tested to map soils in rugged wilderness landscapes with remote sensing and GIS technology plus field reconnaissance and sampling. Maps of the soils of the wilderness areas will be published in GIS format with attribute data available. Methods developed will be applied to other wilderness areas and National Parks in the Pacific Northwest.

Publications

  • Rodgers, T. M. 2000. Modeling soils of the Sawtooth and Pasayten wilderness areas with a GIS. [M.S. Thesis]. Washington State University, Pullman. 110 p.
  • Rodgers, Toby M., A. J. Busacca, B. E. Frazier, and P. E. Gessler. 2000. GIS as a tool for extracting dynamic interactions between soil, vegetation, and topography in the North Cascade Range, Washington. [Abstract]. Geol. Soc. Am., Annual Meeting Abstracts with Programs, Reno, NV. Abstract No. 50074.


Progress 01/01/99 to 12/31/99

Outputs
A project to model soils of the Sawtooth and Pasayten Wilderness Areas with a GIS continued with soil sampling from June through August 1999 within the Sawtooth and Pasayten Wilderness. The goal of visiting the majority of the Pasayten Watershed (124,000 acres) was met. Soil pit descriptions were made for each of the 79 sites visited. Samples were analyzed by electron microprobe and X-ray diffraction for volcanic glass identification and mineralogical content. New soil models have been created and tested. New models include vegetative and landform covers to refine information on local climate and soil conditions within the wilderness areas. Separately, technical field assistance was rendered to the soil survey of Spokane County in 1999.

Impacts
(N/A)

Publications

  • Rodgers, T. M., A. J. Busacca, and P. A. McDaniel. 1999. Indicators of albic horizon development in Spodosols of the North Cascade Range, Washington. [abstract]. ASA-CSSA-SSSA 1999 Annual Meeting Abstracts p. 271.


Progress 01/01/98 to 12/31/98

Outputs
Objective 1: Contribute to NCSS: a) Attended and presented report at state annual interagency soil survey work planning conference. Attended progress field review for Spokane County Soil Survey. Objective 2: Conduct Research: a) Completed first year of three year project to create a soil resource inventory for the Sawtooth Wilderness Area in the North Cascades of Washington by novel methods. Three months of field reconnaissance were spent describing soils and categorizing soil-landform relationships in the wilderness. Representative soils were described and sampled in all sub-watersheds of the wilderness area. An initial GIS model of the distribution of soils on landscapes of the wilderness was made for presentation at the Soil Science Society of America (SSSA) annual meeting in Baltimore, MD. Preliminary comparisons of the computer-generated soil map to field observations show a strong correlation that can be improved upon with the use of a vegetative index of the wilderness area. Current work uses S-Plus statistical software to produce a regression tree representation of field observations for incorporation into the GIS model. In addition, statistical analysis will provide an improved sampling strategy for the upcoming 1999 summer field season in the Sawtooth and Pasayten wildernesses. Electron microprobe data of the composition of volcanic glass in soils will be gathered during the upcoming spring semester for use in soil characterization and presentation at the 1999 SSSA conference in Salt Lake City, UT. Objective 4: Publish Results: a) Completed and published color state general soil map at a scale of 1:750,000. b) Published keynote paper on the role of loess soils in human history for the international conference 'Dust Aerosols, Loess Soils, and Global Change' held in Seattle WA, USA from 11-14 October, 1998.

Impacts
(N/A)

Publications

  • Lumpkin, T.A., and A.J. Busacca. 1998. The critical role of loess soils in the food supply of ancient and modern societies. pp. 95-98 In A. J. Busacca (ed.) Dust Aerosols, Loess Soils, and Global Change. Washington State University College of Agriculture and Home Economics Miscellaneous Publication No. MISC0190, Pullman, WA.
  • Rodgers, T. M., B. E. Frazier, and A. J. Busacca. 1998. Utilizing a GIS for soil classification, Sawtooth Wilderness, Okanogan National Forest, Washington state. [Abstract]. Agronomy Abstracts, American Society of Agronomy, Annual Meeting, Baltimore, MD.
  • Boling, M., B.E. Frazier, and A.J. Busacca. 1998. General Soil Map, Washington. Department of Crop and Soil Sciences, Washington State University, Pullman. 1:750,000.


Progress 01/01/97 to 12/31/97

Outputs
Obj. 1: Continued final editorial preparation of the state general soil map. Obj. 2: a) Samples of surface soils from the Columbia Plateau were resuspended in the lab in a TEOM-based instrument to measure relative dustiness, that is, emission of <10 um particulate dust (PM-10). Samples from the eastern plateau emitted 0.9% of starting mass as PM-10 in a 1-hour experiment; samples from the southwestern plateau emitted less than 0.1% of starting mass as PM-10. Differences are related to particle size distribution of the loessial soils and degree of aggregation by clays and organic matter. b) Regrowth of beech (Fagus sylvatica) on formerly extensively used grassland in the Ticino of southern Switzerland affected the forms of aluminum (Al) and silica (Si), as they are expressed in the ratio of oxalate- to dithionite-extractable Al, and in the difference of oxalate- & dithionate-extractable Si. The former ratio & the latter difference are largest in the subsoil of the oldest succession stage. We concluded that rapid land-cover change may trigger soil chemical responses with a lag time of as short as 50 years that, in turn, affect the geochemical cycles of anionic nutrients (Elsenbeer, et al., 1997).

Impacts
(N/A)

Publications

  • MONTGOMERY, J.A., A.J. BUSACCA, B.E. FRAZIER, AND D.K. MCCOOL. 1997. Evaluating soil movement using 137Cs and the revised universal soil loss equation. Soil Science Society of America J. 61:571-579.
  • BUSACCA, A.J., et al. 1997. Dev. of a PM-10 "dustiness index" to assess wind erodibility of soils of the Columbia Plateau, PNW. [Abst.] Wind erosion: an intern'l symposium. Kansas State Univ.,
  • KENNEDY, A.C., A.J. BUSACCA, AND A.M. IBEKWE. 1997. Biomarkers for identification of displaced soil. [Abst.] Agronomy Abstracts, American Society of Agronomy, annual meeting, Anaheim CA, p. 204.
  • HARSH, J.B., A.J. BUSACCA, H. ELSENBEER, J.S. BOYLE, L. WAGONER, AND R.J. SOUTHARD. 1997. Clay mineralogy of reforested soils in the Ticino region, Switzerland. [Abst.] Agronomy Abstracts, ASA annual
  • ELSENBEER, H., R. SOUTHARD, M. BECK, J. HARSH, AND A. BUSACCA. 1997. Soil responses to land-use & land-cover change. Eos Trans. AGU


Progress 01/01/96 to 12/30/96

Outputs
Obj 1: Completed editorial corrections for 1:750,000 scale Washington State Generalized Soil Map and for a companion bulletin on soils of Washington; lack of funds have delayed publication, now expected in 1997. Obj 2: Continued analysis of surface soils from 146 locations on Columbia Plateau to measure properties related to wind erodibility and PM-10 dust. Content of volcanic glass has been assumed to cause poor aggregate formation and lead to high wind erodibility, but kriged maps of glass content in surface soils shows less than 7% glass in highly erodible soils of the southwestern Columbia Plateau, increasing to >16% glass in non-erodible soils of the northeastern plateau; thus, erodibility is not related in a simple way to single soil properties but rather to critical thresholds of several soil properties. Preliminary experiments with a laboratory soil resuspension system shows an order of magnitude difference in PM-10 yield after 15 minutes among soil samples from different parts of the plateau. Experiments will continue in 1997 to develop an empirical dustiness rating system for the soils of the plateau.

Impacts
(N/A)

Publications

  • MARKS, H. AND A.J. BUSACCA. 1996. Columbia Plateau surface soils and their susceptibility to wind erosion. Abst.
  • BUSACCA, A.J. 1996. The geologic context of the PM-10 issue in eastern Washington state. Abst.
  • MCCOOL, D.K., J.A. MONTGOMERY, A.J. BUSACCA, AND B.E. FRAZIER. 1996. Soil degradation by tillage movement. Abst.


Progress 01/01/95 to 12/30/95

Outputs
Obj 1: Continued editorial corrections for 1:750,000 scale Washington State Generalized Soil Map and for a companion bulletin on soils of Washington: lack of funds have delayed publication, now expected in 1996. Obj 2: Continued analysis of surface soils from 146 locations on Columbia Plateau to measure properties related to wind erodibility and PM-10 dust. Cation exchange capacity surface soils ranges from about 5 to 20 cmol kg-1 and appears to increase to the eastern part of the plateau where soils have higher clay and organic matter contents. 2) Exchangeable sodium percentage (ESP) of surface soils ranges from near zero to about 12%. Little geographic pattern is evident in the results. Possible reasons for the scattered pattern may be local topographic or subsurface horizon conditionsor field-scale fertilizer or water management history. 3) Research is in progress to design and test a laboratory soil resuspension system to estimate the dustiness (PM-10 yield) of field soil samples. Current challenges include controlling static charge buildup in the system, which diverts free PM-10 from reaching the low-volume PM-10 sampler.

Impacts
(N/A)

Publications


    Progress 01/01/94 to 12/30/94

    Outputs
    National Cooperative Soil Survey: 1) Continued editorial corrections for 1:750,000 scale Washington State Generalized Soil Map; publication expected in 1995. 2) Sampled surface soils at 146 locations on Columbia Plateau to measure properties related to wind erodibility and generation of PM-10 dust. Organic carbon in the samples ranged from 0.35% to 2.91% with a strong regional trend from the arid western Plateau to semi-arid eastern Plateau. 3) Classification of an AVHRR (Advanced Very High Resolution Radiometer) satellite image from September 1990 produced land use information for PM-10 emissions modeling. GIS regional data layers, including elevation and soils, also were compiled for PM-10 emissions models.

    Impacts
    (N/A)

    Publications


      Progress 01/01/93 to 12/30/93

      Outputs
      National Cooperative Soil Survey: 1) Continued editorial corrections for 1:750,000 scale Washington State Generalized Soil Map. Pre-production problems were encountered and several steps are bring redone. The map will be printed in early 1994. 2) The monograph Soils of Washington has passed technical editing for publication as a Washington State University Extension Bulletin to accompany generalized soil map. 3) WSU cooperated in three soil surveys. These were published in 1992 but were inadvertently omitted from last year's summary.

      Impacts
      (N/A)

      Publications


        Progress 01/01/92 to 12/30/92

        Outputs
        National Cooperative Soil Survey: 1) Continued editorial corrections for 1:750,000 scale Washington State Generalized Soil Map. Preproduction problems were encountered and several steps are bring redone. The map will be printed as soon as funds are available. 2) The monograph "Soils of Washington" has passed technical editing for publication as a Washington State University Extension Bulletin to accompany generalized soil map. Research: A research project in collaboration with the USDA-SCS is underway to assess multiple local source area for loess and complex polgenesis of soils in the Kittitas Valley area.

        Impacts
        (N/A)

        Publications


          Progress 01/01/91 to 12/30/91

          Outputs
          NATIONAL COOPERATIVE SOIL SURVEY: 1) CHAIRED ANNUAL INTER-AGENCY SOIL SURVEY WORK PLANNING CONFERENCE, SPOKANE, WASHINGTON. 2) FINALIZED EDITORIAL CORRECTIONS FOR 1:750,000 SCALE WASHINGTON STATE GENERALIZED SOIL MAP; COLOR SEPARATION PLATES ARE READY FOR PRINTER; MAP WILL BE PRINTED AS SOON AS FUNDS ARE AVAILABLE. 3) CURRENTLY READYING THE MONOGRAPH "SOILS OF WASHINGTON" FOR PUBLICATION AS WASHINGTON STATE UNIVERSITY EXTENSION BULLETIN TO ACCOMPANY GENERALIZED SOIL MAP. RESEARCH: RESEARCH IS UNDERWAY TO MEASURE THE IMPACT OF FOREST HARVEST PRACTICES ON BULK DENSITY AND INFILTRABILITY OF SOILS HIGH IN VOLCANIC ASH IN THE OKANOGAN HIGHLANDS OF NORTH-CENTRAL WASHINGTON.

          Impacts
          (N/A)

          Publications


            Progress 01/01/90 to 12/30/90

            Outputs
            A mockup of the State Soil Map with a detailed legend has been forwarded to the SCS State Office for final edit and transmittal to the SCS cartographic office for publication. A draft report to accompany the map has been written and is being reviewed. A 20 ha watershed in the eastern Palouse is being examined to determine the extent and quality of soil after many years of erosion due to agricultural practices. Eighteen soil cores were described, noting morphology, depth to paleosol argillic horizons, and albic horizons. These data combined with data from 36 hand augured holes are being incorporated into a relational data base to determine the spatial relationships between depth to argillic and albic horizons and lanscape position. The watershed soils are mapped 24% Palouse silt loam, 27% Naff silt loam, and 49% Palouse-Thatuna Complex. Palouse soils characteristically have no argillic horizon, Naff soils have one at 60 cm, and Thatuna soils have one at 120 cm. Results show 100% of the watershed has an argillic horizon and 86% of it is at 50 cm depth or less. Samples will also be analyzed to determine the extent of vitric subgroups in eastern Washington.

            Impacts
            (N/A)

            Publications


              Progress 01/01/89 to 12/30/89

              Outputs
              National Cooperative Soil Survey. 1) Attended and presented discussion of Andisols in the Blue Mountains of Washington at annual NCSS soil survey work planning conference, Pendleton, Oregon; 2) Reviewed and edited soil survey manuscripts for Snoqualmie Pass Area and Colville Indian Reservation Area, Washington. Research: 1) Completed laboratory experiments to devise acceptable ultrasonic dispersion method for andic soil materials (Hunter and Busacca, 1989); 2) Published results of regional Quaternary stratigraphic study of a sequence of Holocene- to early Quaternary-age alluvial terraces in the Sacramento Valley of California (Busacca et al., 1989) and results of analysis of elemental weathering trends in a chronosequence of soils on the terraces (Busacca and Singer, 1989). Si, K, Ti, and Zr become more abundant in silt fractions with increasing soil age and Na, Mg, Al, Ca, and Fe become less abundant due to differential weathering rates of primary minerals. Using Zr as a stable constituent in calculations, all other of the above elements decrease in silt fractions with increasing soil age. About 40% of the original Ti has been lost from silt fractions of the oldest soil, which is 1.6 Ma. Loss of Mg can be detected after 10 ka and all Mg-bearing silt-sized minerals have been weathered by 130 ka. Ca, Na, and Mg are virtually absent in silts in the upper 4 m of the 1.6 Ma soil, whereas more than one-half of original K is retained.

              Impacts
              (N/A)

              Publications


                Progress 01/01/88 to 12/30/88

                Outputs
                National Cooperative Soil Survey: 1) Busacca serving as president-elect of Washington Society of Professional Soil Scientists; 2) Prepared field guide and led field tour of Channeled Scabland and Palouse areas for 40 soil scientists from Washington, Oregon, and Idaho (Busacca et al., 1988); 3) Participated in field review for soil survey of Douglas County, Washington; 4) Attended and represented WRCC-30 at Western Regional Biannual Soil Survey Work Planning Conference. Research: 1) A state General Soil Map at a scale of 1:750,000 is ready for publication; funds for printing costs must still be obtained; 2) A study of three soils formed in Mazama ash (6800 ka) along a bioclimatic transect in the Blue Mountains of eastern Washington (Hunter, 1988) showed the presence of minerals with short range order; under TEM, it appears to be allophane at the low elevation site and allophane and imogolite at the middle and high elevation sites. Soils are Vitrandepts and Cryandepts. The soil at high elevation differs from the soil at low elevation: it had lower water pH (5.9 vs 6.8), higher NaF pH (11.0 vs 10.4), greater P retention (9.1 vs 6.3 g/kg) and more Al and Si extracted by Na oxalate reagent (22 vs 16 g Al/kg and 8.4 vs 6.2 g Si/kg). Differences in weathering and soil development are attributed to greater profile leaching at higher elevations; 3) A chronosquence of 6 alluvial soils that span 1.6 ma (Busacca, 1987) demonstrate systematic soil development trends.

                Impacts
                (N/A)

                Publications


                  Progress 01/01/87 to 12/30/87

                  Outputs
                  National Cooperative Soil Survey: Serving as Chair of WRCC-30, Western Region Coordinating Committee on Soil Survey; member of WRCC-50, Western region coordinating committee on Soil Climate; attended annual meeting of WRCC-50 and represented the WRCC-30 group interests at this meeting. Research: A state General Soil Map at a scale of 1:750,000 is entering the publication stage. Production estimates have been obtained, a cartographic contractor has eeen retained, and the Soil Conservation Service will provide partial financial support for the project. Completed a study of high elevation grasslands in northern Washington. Grass balds exist at elevations where forest would be expected because soils are shallow and stoney, water holding capacity is low, and because exposure to wind transfers snow off of these sites, increasing local aridity. Completed a soil-geomorphic study of the Yakima Firing Center in which thirty sites were studied for soil and geomorphic properties and grouped into four basic classes; ridgetops, north and south facing hillslopes, alluvial fans, and fan-piedmonts. In the terrain of the YFC, which consisted of anticlinal basalt ridges with a variable-thickness covering of loess, these groupings were found to be the most useful for the environmental impact assessment of the YFC.

                  Impacts
                  (N/A)

                  Publications


                    Progress 01/01/86 to 12/30/86

                    Outputs
                    National Cooperative Soil Survey: Organized and chaired the annual NCSS State Interagency Soil Survey Work Planning Conference; Attended & chaired the university component of the biennial NCSS Western Region Interagency Soil Survey Work Planning Conference. Research: A draft of a state general soils map at a scale of 1:750,000 has been revised to incorporate reviewers' comments. The map has about 120 map units that are based on associations of soil series with physiographic, climatic, and geologic groupings at higher levels. Funding for publication is being sought from state and federal agencies. We have sampled 5 pedons of soils similar to the Tolo series (Typic Vitrandepts), which have formed in a mantle of volcanic ash erupted from Mount Mazama 6800 years ago, in the Blue Mountains of S.E. Washington. The soils form a bioclimatic sequence. B horizons became more strongly developed and an E horizon of leaching is more strongly expressed with increasing precipitation and decreasing temperature along the transect.

                    Impacts
                    (N/A)

                    Publications


                      Progress 01/01/85 to 12/30/85

                      Outputs
                      National Cooperative Soil Survey: Organized the Annual NCSS Interagency Soil Survey Work Planning Conference; edited the manuscript of the soil survey report for Asotin County area; assisted in the USDA-SCS National Soil Survey Laboratory soil sampling for the Palouse area; prepared a map showing soil moisture and temperature regimes in Washington, for use in a compilation of soil climate in the western U.S.; distributed and tabulated a questionnaire to statewide soil scientists on the use of indicator plant species to determine soil moisture and temperature, and a second questionnaire tabulating sites in Washington where long-term soil moisture and temperature measurements are being made. Research: A draft of a state soil map at a scale of 1:750,000 has been completed and reviewers' comments will soon be incorporated as revisions and improvements to the map; A study of high elevation grass balds in northcentral Washington demonstrates that season drought on exposed south facing mountain crests is promoted by stony soils and transfer of snow off these sites and onto leeward positions. Tephra correlation and soil morphology shows that the grass balds have been stable for at least the past 2-300 years and that periodic fires have not caused major changes in these plant communities.

                      Impacts
                      (N/A)

                      Publications


                        Progress 01/01/84 to 12/30/84

                        Outputs
                        National Cooperative Soil Survey: (1) Chaired and presented a progress report at the Annual NCSS Interagency Work Planning Conference, (2) attended and presented a seminar at State All Soil Scientists' Conference, (3) edited soil survey manuscripts of Skagit, Pend Oreille, and Thurston County Areas, (4) assisted in the development of a National Soil Survey Laboratory soil sampling plan for the Palouse area, (5) reviewed and commented on proposed criteria for rating Farmlands of Statewide Importance in Washington, and (6) commented on definitions of Prime and Unique Farmlands in Washington. Research: (1) Work continues on the general soil map for Washington. A rationale for designing map units has been established. Map units have been 0-lineated and legend descriptions written for approximately 75% of the state at a scale of 1:750,000. Areas of high mountains that do not have modern soil surveys have not yet been generalized for addition to the map. (2) Laboratory characterization has been completed of surface charge properties of 3 ash-derived soils from the Colville Indian Reservation soil survey area. Analyses demonstrate significant differences in charge properties between soils. Significance of the Research: Publication of a generalized soil map that is based on Soil Taxonomy and on modern soil survey information will be especially useful to regional planners, high school and college teachers, and for scientists and policy makers at the local, state and federal level.

                        Impacts
                        (N/A)

                        Publications