Forest Inventory from Permanently Established Sampling Plots

U.S. Department of Agriculture (USDA) - Forest Service: Forest  Inventory and Analysis (FIA)
The USDA Forest Service's North Central Research Station, in association with the Minnesota Department of Natural Resources (MN DNR), has started Minnesota's sixth forest inventory.  The Forest Survey Program in the United States began in 1928 and is now called Forest Inventory and Analysis (FIA).  FIA is a national program of the USDA Forest Service that conducts and maintains comprehensive inventories of the forest resources in the United States.  FIA is a continuing endeavor mandated by Congress in the McSweeney-McNary Forest Research Act of 1928 and the Forest and Rangeland Renewable Resources Planning Act of 1974.  Its objective is to periodically determine the extent, condition, and volume of timber, growth, and depletions of the Nation's forest land.  The FIA Unit in St. Paul, Minnesota, is responsible for inventorying more than 80 million acres of forest land spread across 11 states in the North Central Region (Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, South Dakota, and Wisconsin). 

The 1990 Minnesota survey used a growth model-enhanced, two-phase sample design which consisted of a set of randomly located plots that were remeasured and a random set of new plots that were established and measured.  This sampling scheme and associated estimators is similar to sampling with partial replacement (SPR).  The 1990 Minnesota survey design included stratification for disturbance on the old sample and the use of a growth model to improve regression estimates made on old undisturbed forest plots.  The growth model used in the 1990 survey was the Lake States Stand and Tree Evaluation Modeling System (STEMS)

In phase 1, two sets of randomly selected points were located on current aerial photography.  The first is a set of new photo plots and the second is a set of relocated old ground plot locations from the 1977 inventory.  Aerial photos used in the Aspen-Birch and Northern Pine Units were 1:58,000-scale color infrared National High Altitude Photography program (NHAP) prints.  Photos used in the Central Hardwood and Prairie Units were 1:40,000-scale black and white National Aerial Photography Program (NAPP) prints.  The MN DNR provided 1:15,840-scale 35-mm color prints that were used in addition to the NHAP or NAPP photos to help detect disturbances in the 1977 ground plot locations.  The locations of the plots used in the 1977 inventory were transferred to these new photographs.  The photographs were assembled into township mosaics and a systematic grid of 121 one-acre photo plots (each plot representing approximately 190.4 acres) was overlaid on each township mosaic.  Each of these photo plots was examined by aerial photo interpretation specialists who stereoscopically classified the land use.  If trees were present, forest type and stand size / density class were recorded.  All of the 1977 ground plot locations were examined for disturbance (logging, fire, catastrophic mortality) with the aid of the 35-mm photographs.  After this examination, all the old "disturbed" sample locations and one-third of the old "undisturbed" forested plots were surveyed by foresters who verified the land use classification and gathered more detailed measurements. 

In the second phase, or ground location phase, field crews established a cluster of sample plots and collected detailed tree and landscape measurements.  On those plots classified as timberland, wooded pasture, or windbreak (at least 120 feet wide), a ground plot was established, remeasured, or modeled.  Old plots that were scheduled for remeasurement but could not be relocated were replaced with a new plot at the approximate location of the old one.  Each ground plot consisted of a 10-point cluster covering approximately 1 acre.  At each point, trees 5.0 inches or more in d.b.h. (diameter at breast height) were sampled on a 37.5 Basal Area Factor (BAF - cross-sectional area of a tree) variable-radius plot, and trees less than 5.0 inches d.b.h. were sampled on a 1/300-acre fixed-radius plot. 

In the past, inventory in the Midwest has consisted of 12 to 40 foresters establishing and/or remeasuring permanent research samples throughout a given state.  When a state was finished, they would then move on to the next state, cycling back to the original state in approximately 12 years. In 1998, a Blue Ribbon Panel of FIA users was assembled and recommended some very significant changes to the program.  The 1998 Farm Bill reflects their suggestions: 

· Survey 20% of the forest sample plots in all states every year; this will show 100% plot coverage over a 5-year period
· Inventory all forestlands, regardless of ownership or suitability for harvesting
· Prepare state reports every 5 years
· Incorporate analysis of forest health conditions in state reports
· Use common methods to collect and analyze the data across the entire nation 

The North Central Research Station has started to implement these directives in three states: Minnesota, Indiana, and Missouri.  Using aerial photographs, crews made up of two foresters will pinpoint approximately 1,400 forested locations per year, for the next five years.  At these locations, they will take detailed tree and landscape measurements.  This information will then be sent to the FIA office in St. Paul.  There, classifications of current forest area, timber volume, tree growth, tree mortality, forest health, ownership classifications, and current land uses will be developed.  In the 1990 survey, each plot consisted of ten variable radius prism points and ten 6.8 foot radius regeneration points.  In order to standardize the data collection procedures across the nation, FIA will be switching to a new plot design which contains four 24 foot fixed radius points on each plot visited.  The current survey will consist of remeasuring portions of the old prism points and the initial installation of the new fixed radius design.  In addition, FIA crews will be installing plots in "reserved" forest lands, as well as lands previously determined "unproductive." 

The entire state of Minnesota will be divided into approximately 9,100 hexagons, with each hexagon containing 5,936 acres of land.  There will be one plot located in each hexagon throughout the state.  Preliminary photo interpretation shows approximately 1,382 plots going to the field in 1999, the first year of the five-year cycle.  Inventory in Minnesota will be a cooperative effort between the Forest Service and the MN DNR.  The number of plots vary from 214 samples for large forested counties like St. Louis to fewer than 10 plots for smaller agricultural counties in the southwestern portion of the state.  The Forest Service will gather plot data in the southern two-thirds of the state and in St. Louis and Lake counties in the north.  The MN DNR will contract out the remaining counties in the north.  This inventory data will be gathered across all ownerships and land use types.  This statewide inventory will be a continuous process over the next five years.  The compilation, analysis, and report preparation is expected to be completed and available to the public within six months of the completion of the final field work.  This information is available from the North Central Research Station

E.C. Leatherberry,  J.S. Spencer, Jr., T.L. Schmidt, and M.R. Carroll. 1995.  An Analysis of Minnesota's Fifth Forest Resources Inventory, 1990.  Resource Bulletin NC-165.  St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station. 102 p. 

Sixth Minnesota forest inventory begins. 1998. Minnesota Forester, Minnesota Society of American Foresters. 17(2): 3-4.

Eastwide Forest Inventory Data Base User's Guide (.pdf).  This guide is for the database that contains data from field plots collected as part of State FIA inventories.

go to top