1997 National Resources Inventory, Revised December 2000
Revised Inventory Results
This bulletin presents revised summary results from the 1997 National Resources Inventory (NRI). The NRI is conducted by the U.S. Department of Agriculture's Natural Resources Conservation Service, in cooperation with the Iowa State University Statistical Laboratory.
This bulletin has been reissued because in March 2000 an error was discovered in the 1997 NRI results originally issued in December 1999. The error was in the statistical software used by the Statistical Laboratory to calculate estimates for the inventory. Re-release of the inventory results was delayed until December 2000 in order to conduct a complete quality assurance review of all phases of the inventory, to institute a number of additional quality checks, and to include wetlands data in the final release.
The most noticeable difference between the data released in December 1999 and the corrected estimates issued in December 2000 is in the estimated increase in developed lands. The corrected data show an increase of 11.2 million acres during the period 1992 to 1997; this figure is 30% lower than the December 1999 estimate. The corrected estimate of total developed land for 1997 is 98.3 million acres, which is 7.1 million acres lower than the December 1999 estimate. This means the revised figures show an additional 7.1 million acres of rural land for 1997 and, subsequently, more acres of cropland, rangeland, and forest land than were reported in December 1999.
The computer programming error occurred in a portion of the statistical estimation procedure that assigns weights to sample points. These weights are based on sampling (selection) probabilities, estimates from previous NRIs, and known land base attributes from the Census Bureau and other sources. Because of the programming error, incorrect weights were assigned to some sample points. Sample points that changed into developed land were affected in a systematic manner (not randomly), so that the corrections lowered all estimates of 1997 developed land.
This bulletin includes state and national level estimates for changes in broad land cover/use, cropland use by irrigated and nonirrigated acres, broad land cover/use by land capability class and subclass, prime farmland, erosion and erodibility, wildlife habitat diversity, and wetlands and deepwater habitats. These basic summary statistics are presented via the Internet and in hard copy to provide base-line natural resource information to a variety of groups and individuals interested in obtaining insight into the condition of our Nation's nonfederal rural lands. Subsequent sections of this bulletin discuss the broader suite of information available from the 1997 NRI and methods for obtaining access to other results.
The NRI is a scientifically based, longitudinal panel survey of the Nation's soil, water, and related resources, designed to assess conditions and trends every five years. The 1997 NRI provides results that are nationally consistent for all nonfederal lands for four points in time — 1982, 1987, 1992, and 1997.
The Natural Resources Conservation Service, formerly the Soil Conservation Service, was established in response to the Dust Bowl catastrophe of the mid-1930's. Hugh Hammond Bennett, the agency founder and first administrator, convinced the U. S. Congress that soil erosion was a national menace and that a permanent agency within the Department of Agriculture was needed to call landowners' attention to their land stewardship opportunities and responsibilities. The results of the 1934 National Erosion Reconnaissance Survey, which was the first formal study of erosion conducted in the United States, were instrumental in the passage of the Soil Conservation Act of 1935. Through the Act, the Soil Conservation Service was established, and a nationwide partnership of federal agencies, local conservation districts, and communities was developed to provide assistance to the rural and urban sectors in the conservation of natural resources. Today, more than 60 years later, NRCS champions the vitality of the land as USDA's lead conservation agency. No other federal agency speaks for the health of America's private land.
Throughout its history, NRCS has conducted periodic inventories of the Nation's natural resources. The 1945 Soil and Water Conservation Needs Inventory (CNI), a reconnaissance study, was the foundation for the 1958 and 1967 CNI's, the agency's first efforts to collect data nationally for scientifically selected field sites. The 1975 Potential Cropland Study examined the conversion of the Nation's best farmland to urban development. National Resources Inventories were conducted in 1977, 1982, 1987, 1992, and 1997. Several less intensive, special-issue inventories have been performed during the 1990's to investigate topical matters of concern and to supplement recent NRIs.
In addition to these recurrent NRI inventories, NRCS also collects large quantities of field level natural resources data in support of conservation planning activities and the Soil Survey Program. Thousands of NRCS technical specialists, including soil scientists, soil conservationists, range conservationists, foresters, wildlife biologists, and agronomists, collect data at the field and farm level in order to provide conservation assistance to farmers and ranchers in the development of conservation systems uniquely tailored to the land and their individual way of doing business. Assistance is also provided to rural and urban communities to help reduce erosion, conserve and protect water resources, and solve other resource-related problems. The information that NRCS collects about natural resources in the United States is critical for sustaining agriculture, promoting the conservation and stewardship ethic, and preserving the health of the Nation's natural resources and environment.
Legislation also has mandated that NRCS collect natural resources data. The Rural Development Act of 1972 was a key statute in authorizing resource inventory activities within NRCS. It directs the Secretary of Agriculture to implement a land inventory and monitoring program and to issue a report on the conditions and trends of soil, water, and related resources at intervals not exceeding 5 years. The Soil and Water Resources Conservation Act of 1977 and other supporting legislation augmented the statutory mandate for periodic assessment of the Nation's natural resources. To fulfill this requirement, the NRI was developed to provide critical information regarding natural resources and to supplement the NRCS Soil Survey Program.
The objectives of NRCS resource inventories have expanded over time, as the focus of agricultural policy has moved toward a balance between short-term production goals, long-term capabilities, and environmental quality. Statistical techniques, data collection protocols, and data handling and dissemination technologies have evolved as inventory goals have become broader and more sophisticated.
The primary objective of the 1997 NRI was to provide natural resource managers, policy makers, and the public with scientifically valid, timely, and relevant information on natural resources and the environment. This information can provide the scientific basis for effective public policies, sound agricultural and natural resource legislation, sensible state and national conservation programs, and targeted USDA financial and technical assistance in addressing natural resource concerns. NRI data are designed to be part of the core components of the agency's strategic planning and accountability efforts, and to help assess consequences of existing legislative mandates, such as the 1996 Farm Bill.
To accomplish these objectives in a cost-effective manner, it was necessary to conduct the 1997 NRI in much the same manner as the 1992 NRI. Careful consideration was given to assure that 1997 NRI data elements were consistent with definitions, categories, and concepts from previous inventories. The same sample used for the 1992 NRI was used for 1997 data collection. This enables analysis of trends extending over 15 years (1982, 1987, 1992, and 1997).
NRI data are collected at scientifically selected sample sites. The sample constitutes a two-stage stratified area sample of the entire country. Samples are located in all counties and parishes of the 50 states and in Puerto Rico, the Virgin Islands, the District of Columbia, and selected portions of the Pacific Basin. The first-stage sampling unit, or primary sampling unit (PSU), is an area/segment of land; the second-stage sampling units are points located within the PSUs. Detailed NRI data are collected for the specific sample points, but some items are also collected for the entire PSU/segment. Some data, such as total surface area, federally owned land, and area in large water bodies, are collected on a census basis external to the sample survey. The NRI database accounts for and represents the total area of the United States, but very little information is given for points on federal lands.
Data for the 1997 NRI were collected for about 300,000 PSUs and 800,000 sample points, using photo-interpretation and other remote sensing methods and standards. Data gatherers utilized a variety of ancillary materials; extensive use was made of USDA field office records, information provided by local NRCS field personnel, soil survey and wetland inventory maps and reports, and tables and technical guides developed by local field office staffs. The NRI is unique because of established linkages to the NRCS Soil Survey Program. NRCS expertise in identifying soil occurrences and patterns, and utilizing this knowledge to provide technical assistance in the development of conservation plans for landowners is a primary agency function. The NRI data gathering process relies heavily upon information contained in the Soil Survey Database. Information about specific properties and characteristics of the soil and surrounding landscapes is utilized to develop NRI data elements and interpretations. NRI data is a primary source of information for evaluating the success of NRCS soil conservation programs in reducing soil erosion.
Inventory procedures were developed to ensure that data reflect 1997 growing season conditions, that inventory results are nationally consistent, and that data recorded for the years 1982, 1987, and 1992 are consistent with the 1997 determination. Intricate quality assurance procedures were developed to make sure that year-to-year differences reflect actual changes in resource conditions, rather than differences in the perspectives of two different data collection specialists or changes in technologies and protocols.
Data gathering for the 1997 NRI occurred from July 1997 through October 1998. This time frame took into account that some aerial photography needed to be flown during a time period that highlighted late growing season conditions. Consequently, delivery of imagery to some data collection sites did not occur until later in the data collection cycle.
Field visits were not required for the 1997 NRI unless available imagery and ancillary materials were not suitable for making determinations for one or more data elements. Field visits were also made for training purposes and other facets of the quality assurance process. All NRI sample sites were visited on-site for the 1982 NRI. Subsequent on-site visits of selected PSUs also occurred in 1987, 1991, 1992, or 1995.
The computer-assisted survey information collection methods developed for the 1997 NRI provided substantial efficiencies in data gathering and data processing and were important facets of the quality assurance process. The system featured direct entry of data into hand-held computers called personal digital assistants (PDAs), modern information technology, a national database server at the Iowa State University Statistical Laboratory, and elaborate data checking protocols that featured review and edit of data recorded during previous inventories.
Standards and protocols for the NRI were developed nationally by NRCS, in collaboration with the Statistical Laboratory. Oversight and management of data-gathering activities were assigned to 21 units established during 1996 and 1997. These units, called Inventory Collection and Coordination Sites (ICCS's) were established according to regional land use patterns and according to state allocations of resources. Geographic boundaries of ICCS organizations ranged from one state to all or portions of several states. Some ICCS's distributed data collection staff among multiple office locations, while others assembled staff at one central location.
Inventory methodology is evolving as part of an ongoing effort to better assess soil conservation, natural resource health, and other agri-environmental issues. The NRI has been conducted as a longitudinal survey designed to assess condition and trends of nonfederal lands every 5 years. Current initiatives include transitioning to a continuous resource inventory process, developing a multi-agency integrated inventory approach, incorporating a wider variety of assessment tools for resource health, and further development of geospatial analysis and modeling capabilities to support policy analysis and program implementation.
Utilization and Interpretation of NRI Data
Uses of the Data
The NRI database contains millions of pieces of information. It can serve as the foundation for inspection and analysis of the condition of our Nation's natural resources. It indicates
how our Nation's nonfederal lands are being used;
the condition of our natural resources;
how land use patterns have changed over time.
The NRI database has been constructed in a manner that facilitates the inspection and analysis of these data. Sophisticated statistical procedures developed collaboratively with Iowa State University have been used to provide a database that scientifically incorporates a broad array of data into a format that is easy to use and manipulate. Appendix 1 presents an overview of statistical considerations and provides references that contain further details.
The 1997 NRI database contains data for four points in time (1982, 1987, 1992, and 1997) that are comparable and consistent, and that reflect true trends. Reliable and accurate temporal analysis is available from this data set. Analytical capabilities are greatly enhanced because NRCS's extensive soil interpretations database is an integral and easy-to-use part of the NRI database.
The NRI is conducted to obtain scientifically valid, timely, and relevant data on natural resources and environmental conditions, with the specific goal of supporting agricultural and environmental policy development and program implementation. Historically, NRI information has been used to formulate effective public policies, to fashion agricultural and natural resources legislation, to develop state and national conservation programs, to allocate USDA financial and technical assistance in addressing natural resource concerns, and to enhance the public's understanding of natural resources and environmental issues. Information derived from the NRI is used by natural resource managers; policy makers and analysts; consultants; the media; other federal agencies; state governments; universities; environmental, commodity, and farm groups; and the public.
Interpretation of the Data
Statistics derived from the NRI database are estimates and not absolutes. This means that there is some amount of uncertainty in any result obtained using NRI data. Statistical reliability guidelines are discussed in Appendix 1.
The NRI database contains linkages to other databases, in particular to the agency's extensive soil interpretations database. Linkages can be made to other databases by using other themes, such as cover/use, forest cover type, and spatial features. Analysis of NRI data in conjunction with other data sources is encouraged; nonetheless, differences in definitions, concepts, and data collection protocols should be carefully examined. It is worth repeating that the NRI includes very little data for federal lands.
The 1997 NRI database has been designed for use in detecting significant changes in resource conditions relative to the years 1982, 1987, 1992, and 1997. All comparisons for two points in time should be made using the new 1997 NRI database. Comparisons made using data published for the 1982, 1987, or 1992 NRI may produce erroneous results, because of changes in statistical estimation protocols, and because all data collected prior to 1997 were simultaneously reviewed (edited) as 1997 NRI data were collected. Note, for example, that federal land area for 1992 has been adjusted from 408 to 402 million acres, and that the estimate of 1992 nonfederal rangeland has changed from 399 to 407 million acres.
The NRI provides not only overall estimates of change in resource conditions but also the dynamics of the changes. For example, it is typically more informative to examine gross losses and gains in cropland (rather than just the net change from one year to another) and further to determine why cropland was lost (e.g., to urban development), how much had been prime farmland, and "where" these losses are occurring. If new cropland is gained, the question is whether this will cause additional conservation and environmental concerns because the new cropland is more erodible, or the soils are less productive and require higher levels of fertilization, or the land is sited in some other sensitive location.
The erosion data cannot be used to compute the actual erosion occurring during a particular year. Erosion rates are estimated average annual (or expected) rates based upon the cropping practices, management practices, and inherent resource conditions that occur at each NRI sample site. Climatic factors used in the erosion prediction equations (models) are based upon long-term average conditions and not upon one year's actual events. Note also that NRI estimates of sheet and rill erosion are based upon the standard Universal Soil Loss Equation (USLE) and not the revised USLE (RUSLE), and that erosion estimates are made only for cropland, CRP land, and pastureland.
The NRI category of "developed land" varies from that used by some other data collection entities. For the NRI, the intent is to identify which lands have been permanently removed from the rural land base. Therefore, the developed land category includes: (a) large tracts of urban and built-up land; (b) small tracts of built-up land, less than 10 acres in size; and (c) land outside of these built-up areas that is in roads, railroads, and associated rights-of-way.
The 1997 NRI shows only minor changes in land under Conservation Reserve Program (CRP) contracts for the time period 1992 to 1997, even though most original CRP contracts expired in the mid-1990's and there were extensive sign-ups during that period. This is because the 1997 NRI reflects conditions as of the 1997 growing season, and most actual on-the-ground changes in CRP land did not occur until later in 1997 or until the 1998 growing season.
For the NRI, land is considered irrigated if irrigation occurs during the year of inventory, or for two or more of the last four years. Other entities typically consider land to be irrigated only if irrigation water is applied for the year of interest.
NRI classification of wetlands is slightly different than that used by the Fish & Wildlife Service (FWS) in their statistically based Wetlands Status and Trends study. The NRI and the FWS inventory have different legislative mandates; sampling methodology, inventory protocols, data handling, and analysis routines have evolved independently over the past two decades. Recent collaborative efforts have resulted in enhanced classifications for both programs, but wetlands data collected by the two agencies are currently neither comparable nor interchangeable. The NRI multi-resource approach is quite beneficial to USDA analysts and others who examine conservation and agri-environmental issues. Results from the FWS study are quite beneficial to the Department of Interior and others.
The NRI has been designed to facilitate geospatial analysis. This not only enhances the analysis process, but also makes it possible to use a map to present analytical findings. Maps produced from NRI data depict only patterns or trends within an area and do not provide an estimate of conditions for any specific location on the map.
This report presents selected NRI summary data at the national level. Further information regarding the NRI and additional data summaries can be obtained from the national NRI site.
Additional data summaries from the NRI will be released periodically as more comprehensive analyses are performed. Of particular interest are detailed compilations of data at the state level, which can be accessed via this Internet site. Active links to individual state Internet sites are available for obtaining specific state-level NRI information.
Explanation of the Tables
On the following pages selected national summary data are displayed in 19 tables. Definitions of terms are the same as for the 1992 NRI (see Appendix 3 — Glossary of Selected Terms). National totals include results for the 48 contiguous states, Hawaii, and the Caribbean area. Results for Alaska and the Pacific Basin islands of Guam, Rota, Tinian, and Saipan will be released at a later date.
The category "other rural land," which occurs in many of the tables, includes farmsteads and other farm structures, field windbreaks, barren land such as salt flats or exposed rock, and marshland.
The figures used in the tables are estimates, not absolutes. They are based on data collected at sample sites, not data taken from a complete census. Therefore, sampling variation is present but generally small for state and national totals. However, sampling variation may be significant when using these totals to calculate 5- and 10-year changes. Small changes may not be statistically significant.
Table 1 presents NRI findings on surface area, federal land, nonfederal rural land, developed land, and water area. Since 1982, federal land increased by 2.7 million acres, nonfederal rural land decreased by 28.9 million acres, and developed land increased by 25.0 million acres. Developed land totaled 98.3 million acres in 1997 (fig. 1).
Tables 2 and 3 present estimates of acreage of land cover/use for six components of nonfederal rural land (cropland, CRP land, pastureland, rangeland, forest land, and other rural land). Cropland is classified as irrigated, nonirrigated, cultivated, or noncultivated acreage. Cropland acreage nationally decreased by 44.0 million acres between 1982 and 1997. Rangeland decreased by 10.8 million acres and pastureland decreased by 12.0 million acres. Table 3 further reveals a shift in irrigated agriculture from west to east across the country.
Table 4 presents acres of land cover/use by land capability class and subclass. The land capability classification system was originally developed by the Soil Conservation Service and provides a quick, uniform, and useful way to evaluate the potential of land for crop production. Each capability class has several subclasses to identify specific limitations on use:
e = erosion risk,
w = wetness,
s = shallowness or root zone problems, and
c = climatic limitations.
Class I soils have few limitations that restrict their use. Class II soils have moderate limitations that reduce the choice of plants or that require careful management. Land identified as Class IIe, for example, would be suitable for growing crops if adequate measures were installed to reduce or prevent soil erosion.
Class III soils have severe limitations that reduce the choice of plants, require special conservation practices, or both. Class IV soils have very severe limitations that reduce the choice of plants, require very careful management, or both.
Class V soils are not likely to erode but have other limitations, impractical to remove, that limit their use largely to pasture or range, woodland or wildlife. Class VI soils have severe limitations that make them generally unsuitable for cultivation and limit their use largely to pasture, range, woodland, or wildlife. Class VII soils have very severe limitations that make them generally unsuitable for cultivation and limit their use largely to pasture, range, woodland, or wildlife. Class VIII soils and miscellaneous land types have limitations that preclude their use for commercial crop production and restrict their use for recreation, wildlife, water supply, or esthetic purposes.
Tables 5, 6, 7, and 8 provide an overview of land use changes from 1982-97, 1982-87, 1987-92, and 1992-97. These tables show all land conversions, whereas previous tables presented net land use change. For example, table 5 shows that a total of 70.7 million acres of 1982 cropland was converted to other uses by 1997, which was offset by 26.7 million acres converted to cropland from non-cropland uses since 1982. The net change was therefore a reduction of 44.0 million acres of cropland, as shown in table 2. Table 5 further shows that, of the 70.7 million acres of cropland converted to other uses, 30.4 million acres went to CRP, 19.3 million acres went to pastureland, 3.7 million acres went to rangeland, 5.6 million acres went to forest land, 3.2 million acres went to other rural land, 7.1 million acres went to developed land, and 1.5 million acres went to water areas and federal land. Of the 26.7 million acres converted to cropland from other uses, 15.3 million acres came from pastureland, 7.0 million acres came from rangeland, 2.0 million acres came from forest land, 1.4 million acres came from other rural land, 0.2 million acres came from developed land, and 0.8 million acres came from water areas and federal land.
Table 9 presents the distribution of prime farmland by land cover/use. Prime farmland is land that has the best combination of physical and chemical characteristics for producing food, feed, forage, fiber, and oilseed crops and is also available for these uses. There were 331.9 million acres of prime farmland in 1997, which was down 10.0 million acres from 1982. Most (64%) of the prime farmland is in cropland, but large amounts are in pastureland (35.5 million acres) and forest land (48.7 million acres).
Tables 10 and 11 present estimates from the NRI for soil erosion rates. Table 10 shows rates of sheet and rill erosion, which is erosion caused by water; table 11 presents estimates of wind erosion. Average erosion rates for 1997 are substantially lower than erosion rates for 1982. The average rate of sheet and rill erosion fell from 4.0 tons per acre per year in 1982 to 2.8 tons per acre per year in 1997. The average rate of wind erosion on the same land base fell from 3.3 tons per acre per year in 1982 to 2.2 tons per acre per year in 1997. The combined wind and water erosion reduction translates to a savings of more than 1.2 billion tons of soil per year on cropland (fig. 2).
Tables 12 and 13 present acres of cropland (cultivated and noncultivated), pastureland, and CRP land with erosion rates greater than T, the soil loss tolerance or rate at which soil productivity is maintained.
Table 14 shows acreage according to six classes of erodibility index scores. The erodibility index (EI) provides a numerical expression of the potential for a soil to erode, considering the physical and chemical properties of the soil and climatic conditions where it is located. The higher the index, the greater the investment needed to maintain the sustainability of the soil resource base if intensively cropped. EI scores above 8 are equated to highly erodible land.
Table 15 presents statistical information dealing with wildlife habitat composition and configuration. Median diameter of habitat patch size is an indicator of habitat diversity. For the 1997 NRI, general cover data were collected along X-shaped transects (the length of each diagonal line of the transect was 1,000 feet). Patches of cover were classified to one of nine general cover types (see Appendix 3, glossary). Entries in Table 15 denoted as "> 1,000" indicate that at least 50% of the transects were classified as having a 1,000 foot length of the same cover type.
Tables 16, 17, 18, and 19 present estimates of status and changes in wetland and deepwater habitat acreage; classification of these systems is according to the Cowardin et al. (1979) classification system. Table 16 shows an estimate of 111.2 million acres of Palustrine and Estuarine wetlands on water areas and nonfederal land for 1997, as well as an additional 47.8 million acres of other aquatic habitats. Table 17 presents the distribution of Palustrine and Estuarine wetlands by land cover/use; most (59%) of these acres are on forest land. Table 18 shows 506.0 thousand acres of Palustrine and Estuarine wetlands were converted to uplands between 1992 and 1997, and 343.2 acres of uplands were converted to Palustrine and Estuarine wetlands during this time period (fig. 3). These conversion estimates show an average annual gross loss of 101.2 thousand acres, an average annual gross gain of 68.6 thousand acres, and an average annual net loss of 32.6 thousand acres per year. These figures do not take into account any losses or gains occurring on federal land. Table 19 presents more particulars regarding the losses and gains. Development accounts for 49% of the losses; agriculture accounts for 26 % of the losses (fig. 4). Natural variations in climatic cycles and hydrology are responsible for the majority of wetland losses captured in the "miscellaneous causes" category. Estimates dealing with losses and gains are presented nationally and by region, but not by state; almost all state numbers are smaller (in absolute value) than their margins of error and, therefore, not significantly different (statistically) than zero. Estimated margins of error (or 95% confidence intervals) are presented in table 19 to assist the reader in determining which regional estimates are statistically significant. Margins or error and statistical reliability are discussed in Appendix 1.
To convert acres to hectares, multiply the number of acres by 0.405.
To convert tons to metric tons, multiply the number of tons by 0.907.
To convert tons/acre to metric tons/hectare, multiply the number of tons/acre by 2.24.
< Back to Summary Report Contents