Session 1: Annual Inventory

With many competing uses and large regional shifts in forestland use, the sustainability of the nations's forests is being questioned. The new annual forest inventory system that is being developed and implemented in the Eastern US addresses national, regional, and state questions regarding past, current, and projected changes in a variety of forest attributes. The annual data acquired with this system will provide the information needed to closely monitor and quantify the landscape dynamics of forests and will form the basis of state, regional and national forest sustainability assessments. As implem ented in the Eastern US, the system features several distinct components: (1) remote sensing; (2) an annual sample of measured field plots; (3) a database o f plot and tree information; (4) multiple estimation techniques including opti ons for updating plots measured in previous years; and (5) techniques for proj ecting current conditions into the future. The discussion focuses on the rati onale for developing an annual inventory system, details of common features of the system as implemented throughout the Eastern US, and the integration of i ntensive forest health measurements with traditional indicators of forest sust ainability.

^*Presented by Will McWilliams.

Annual Forest Inventories: The North Central Perspective

Ronald E. McRoberts

North Central Forest Experiment Station

USDA Forest Service

1992 Folwell St. Paul 55108 USA

Abstract

The USDA Forest Service's primary objective in developing procedures for annual forest inventories is to establish the capability of producing estimates of forested area, timber volume, related variables, and changes in these variables on an annual basis. The inventory system developed to accomplish this objective features several functions: (1) an annual sample of measured field plots; (2) remote sensing; (3) a database of plot and tree information; and (4) logistical procedures for supporting field crews. An optional function, (5) mechanisms for updating the status of plots measured in previous years, is also incorporated. The discussion focuses on specific features and options for each function and on options selected for implementation in the North Central region.

^*Presented by Ronald McRoberts.

SAFIS Design And Estimation Techniques

Gregory A. Reams

Southern Research Station

USDA Forest Service

PO Box 2680, 200 Weaver Blvd.Road Asheville 28802 USA

Abstract

To address the uncertainty of forest sustainability in the South, the Southern Forest Inventory and Analysis (FIA) program has initiated an annual survey system. Currently there are eight out of the thirteen southern states involved with annual forest surveys. The Southern Annual Forest Inventory System (SAFIS) sample design consists of five interpenetrating annual survey panels with no annual or multi-year overlap. In the transition period from periodic design to full implementation of the annual five-panel design, a number of analysis options are possible. Some options that have been considered include: (1) produce estimates based only on those plots measured each year; (2) average the new panel information with the previous periodic information using moving average models; (3) complete the first two or more annual panels (at least 40% percent of all FIA plots) before reporting current inventory information. These different options are being planned for because some states moved from to an annual survey immediately after completion of a periodic survey, while others moved from a periodic to annual survey with five or more years elapsed between surveys of any kind. The design and analysis concepts of SAFIS will be presented and discussed.

^*Presented by Gregory A. Reams.

Implementing Annual Forest Inventory Systems For The Mid-Atlantic States And New England

William H. Mcwilliams

Northeast Research Station

USDA Forest Service

634 University Place Swartmore 19081 USA

Abstract

A new state-level annual inventory system is being developed to monitor health and sustainability of forest ecosystems in the Mid-Atlantic States and New England. The new system is part of a national initiative to achieve consistency in sample design, compilation, and analytical procedures used by the USDA Forest Service, regional Forest Inventory and Analysis (FIA) units and the Forest Health Monitoring (FHM) program. The system uses a three-phase design that combines remote sensing, a revised FIA ground sample, and the more intensive FHM ground sample. Annual measurements of ground samples will occur on a 5-year cycle, or 20 percent of the sample locations. Annual mensuration data will be made available each year and a comprehensive analysis of results will be published at the end of each State’s inventory cycle.

Keywords: Forest Inventory and Analysis, Forest Health Monitoring, forest sustainability, mensuration, and remote sensing.

^*Presented by Will McWilliams.

Techniques For Updating Plot Information For Annual Forest Inventories

Ronald E. McRoberts

North Central Forest Experiment Station

USDA Forest Service

1992 Folwell St. Paul 55108 USA

Abstract

USDA Forest Service annual inventories feature measurement of a proportion of plots each year. Inventory estimates that satisfy precision requirements may be obtained by combining information from plots measured in the current year with information from plots measured in previous years. The estimation option that is expected to yield the most current estimates uses statistical techniques to update to the current year information for plots measured in previous years. Several simple updating techniques, including multiple imputation and model-based prediction, are evaluated with respect to bias and precision. The results indicate that several techniques are relatively easy to apply and are approximately equally unbaised and precise.

^*Presented by Ronald McRoberts.

Stand Signature Stability

H. Whiffen, C.J. Cieszewski, B. Borders, R. Lowe, W. Izlar

School of Forest Resources

University of Georgia

Athens 30602-2152 USA

M. Zupko

Georgia Forestry Association

500 Pinnacle Court, Suite 505

Norcross, GA 30071

B. Cooke

Southern Forest Experiment Station

Forest Inventory and Analysis

USDA Forest Service

201 Lincoln Green

Starkville, MS 39759

Abstract

The Landsat satellites collect data from the forested areas along the Alabama-Georgia border (Path 19, Row 37) every 16 days. These data represent the average electromagnetic energy reflected back to the sensor from the forested area. Satellite data are available at a finer temporal resolution (16 days vs. 5 years) and a finer spatial resolution (~100 ft between sample plots vs. ~200 ft (3 x 3 chain cruise)) compared to traditional data collection methods, but how reliable is the information?

We compared the information that comes from a traditional timber cruise with the information that comes from satellite data for basal area of timber stands and biomass volume. Specifically, we asked, "When can satellite data be trusted to accurately inform managers of the standing basal area and biomass volume on the timberland they manage?"

We started with the best case: unthinned loblolly pine stands between the ages of 14 and 29 years growing in old fields. Tree spacing varied between 525 TPA and 665 TPA. We cruised the timber in eight different sections at a fine scale (approximately 100-ft x 100-ft spacing). We compared the basal area and volume calculated from the cruise data with the signature responses found in both a leaf-on and leaf-off image and evaluated the remote sensing signature stability of similar timbered areas.

^*Presented by Helen Whiffen.

Last revised: April 7, 2000

Session 1: Annual Inventory

(Moderated by Chris J. Cieszewski)

Contents