Skip

APPENDIX 2. 1997 NRI Data Gathering Protocols, Processes, and Procedures

Summary Report, 1997 National Resources Inventory, Revised December 2000

PROTOCOLS

For the 1997 NRI, cross-indexed data gathering instructions, training, and the survey instrument were developed to foster consistent data gathering standards, practices, and procedures. The protocols called for natural resources information to be gathered for survey (panel) years 1982, 1987, 1992, and 1997. Instructions and training articulated the requirement that trending information be gathered by the same protocols and for the same locations used in prior studies. The 1997 instrument included algorithms to evaluate trending data for consistency.

Data gathering protocols incorporated NRCS technical standards and procedures, records and maps in local USDA offices, and various federal publications and standards, primarily the following:

  • NRCS national and field office technical guide (FOTG) publications and standards relating to the universal soil loss equation (USLE) and wind erosion equation (WEQ),

  • NRCS-published or -correlated soil surveys,

  • NRCS information relating to provisions of the 1985 Food Security Act and subsequent farm bills,

  • U.S. Fish and Wildlife Service wetland maps,

  • U.S. Fish and Wildlife Service Cowardin wetlands classification system,

  • Society of American Foresters forest classification,

  • U.S. Geological Survey hydrologic and topographic maps, and

  • Bureau of the Census TIGER files and auxiliary information.

  • Data gatherers received the protocols as well as training and technical reference materials in printed and electronic forms. National staffs conducted multiple training sessions to support core work groups.

DATA GATHERING

NRCS national program staff in Washington (DC), Fort Collins (CO), Ames (IA), and Fort Worth (TX) guided and supported the data gathering activities. State or regional oversight authorities supported local data gathering staffs in matters relating to quality assurance. A single point of contact ('Help Desk') on the national staff responded to questions from data gatherers and coordinated technical responses provided by subject-matter experts from NRCS and the Iowa State University Statistical Laboratory.

Twenty-one Inventory Collection and Coordination Sites (ICCS) were established and assigned oversight and management authorities for data gathering. These sites were:

Ames, IA Amherst, MA Anchorage, AK
Auburn, AL Bismarck, ND Boise, ID
Bozeman, MT Davis, CA East Lansing, MI
Lakewood, CO Lexington, KY Little Rock, AR
Madison, WI Morgantown, WV Phoenix, AZ
Portland, OR Raleigh, NC Reno, NV
Salina, KS Spokane, WA Temple, TX

The ICCS's were the front-line management structures responsible for coordinating the day-to-day activities associated with the collection of data for the 1997 NRI. The ICCS leader trained subordinate staffs, provided technical support, and managed quality reviews during the operational phase of data gathering. Full-time, part-time, and temporary NRCS employees - and in a few places, volunteers - gathered 1997 NRI information. The organization of data gathering varied with regional land use and state staffing patterns. Geographic boundaries of ICCS organizations ranged from one state to all or portions of several states. Some collection sites assembled staffs at one central office, while others distributed staff among multiple office locations. The State Conservationists assigned various technical specialists to provide overall support and to work with the ICCS data gathering teams.

Data gatherers used photo-interpretation (PI) and other remote sensing (RS) methods and standards to gather information about the PSU's and sample points. For the most part, they employed analog PI techniques, although GIS technologies were evaluated. The agency contracted for the acquisition of aerial photography or obtained necessary imagery in cooperation with other USDA agencies and partners.

USDA field office records and local NRCS personnel provided information pertaining to historical cropping and management systems for calculating long-term erosion rates induced by wind or water, and to determine if the field at the sample point was enrolled in the Conservation Reserve Program. Visits to the sample sites occurred only when aerial photography was not available or was not suitable for reasons of age, quality, scale, or format, or for quality assurance purposes.

Data gatherers entered all sample data directly into hand-held computers called personal digital assistants (PDA's). All subsequent data quality checking and evaluations were similarly based on computer forms of survey information. The PDA's (Apple Computer's Newton MessagePad™ models 130, 2000, and 2100) were programmed to provide an intelligent survey questionnaire with historic information, procedural logic, and single or multi-variate checking for data completion and consistency. The PDA's uploaded and downloaded sample records via Internet protocols from a centralized database server at Iowa State University. The server controlled and monitored access by 'client' instruments and protected survey data from loss, unauthorized access, or accidental disclosure. A secure Web site allowed database access for purposes of survey management, review of progress, and data quality evaluation.

Quality assurance of NRI data was monitored at all organizational levels within the NRI program and was accomplished by several procedures and protocols. These included consistent training of data collectors in data collection processes, standardized formalized written data gathering instructions, documentation, and definitions of data elements; consistent national rules and methods for data collection, and a national help desk to resolve data collection issues.

Quality assurance procedures included the use of data validation software packages on the Personal Digital Assistants (PDA's). Hundreds of data collection rules comparing multiple data elements were run on the PDA's prior to submittal of data to the ISU Statistical Laboratory. The Statistical Laboratory performed additional data validation and consistency checks on all data received. After statistical estimation procedures were completed, tables were generated and sent to technical specialists at the ICCS's for further review and comment.

Quality assurance for the NRI process was guided by instructions, coordinated training, and ongoing technical support. Data quality checking procedures were embedded in every step, stage, and phase of the data collection process for the 1997 NRI.

Detailed soils information is essential for the analysis and interpretation of NRI data. Soils data were provided by the NRCS Soil Survey Program and were obtained from the NRCS Soil Interpretation Record database maintained at the Iowa State University (ISU) Statistical Laboratory. For the 1992 NRI extensive work was done to match individual State Soil Survey Databases (SSSD) with each point in the NRI. This process was designed to verify the accuracy and completeness of the NRI soils database. This work also provided accurate soils data for use in the 1997 NRI. Published soil surveys, advanced (pre-publication) soil mapping field sheets, state level databases, and ancillary lists of soils information maintained in field offices were used to provide critical soils data. Information on soil properties related to soil erosion and other soil-dependent interpretations (i.e., prime farmland) were linked to the NRI database.



< Back to Summary Report