| Description |
1 online resource (80 p.) : ill. |
| Series |
General technical report NRS ; 53 |
|
General technical report NRS ; 53.
|
| Note |
"October 2009." |
|
Title from Web page (viewed on June 3, 2011). |
| Summary |
The Forest Inventory and Analysis (FIA) program of the U.S. Forest Service collects vast amounts of field data to assess the condition and trends of the nation's forest resources. A quality assurance (QA) program is implemented to assure that data are collected accurately with consistent protocols. A random subset of field plots is chosen to receive an additional, independent measurement by another field crew. This 'blind check' approach allows for comparison of measurements between the two crews. The measurement differences are evaluated against measurement quality objectives (MQO), which specify a level of measurement precision for each attribute. In this report, differences in repeated measurements from blind check data are analyzed to assess the quality of forest health data nationwide. For the understory vegetation indicator, the ground cover variables attained the MQO or were slightly below the desired level. Variables related to canopy and species identification were substantially below the MQO. Tree crown attributes that effectively met the MQO include crown light exposure, crown dieback, and foliage transparency. The uncompacted crown ratio, vigor class, and crown density measurements did not attain the MQO standards. Quality assessment for the lichen indicator is evaluated using a second measurement by an expert and not an independent measurement by another field crew. Overall, the MQO was not attained, although compliance improved over time and western regions were close to achieving the standard. Most variables measured for the down woody material indicator were below MQO standards. Due to the sampling protocols for the ozone indicator, direct analyses of MQO attainment are not possible. Mean biosite index was not significantly different between crews, although there were some large discrepancies on individual plots. There was agreement on presence/absence of ozone damage for about 80 percent of the plots. For field measurements of forest soils, 10 of the 12 variables attained the MQO. Although MQOs are not specifically stated for laboratory analyses of soil properties, comparisons were performed to assess the variability of lab measurements. This information should be useful to data collection experts, as variables having poor measurement repeatability can be identified and examined for potential resolution. The results may also be of interest to analysts and researchers wanting to evaluate whether the repeatability of measurements is sufficient for their respective studies. |
| Note |
Description based on print version record. |
| Bibliography |
Includes bibliographical references (p. 79-80). |
| Subject |
Forest Inventory and Analysis Program (U.S.) -- Evaluation.
|
|
Forest health -- United States.
|
|
Forest surveys -- United States.
|
| Added Author |
Westfall, James A.
|
|
United States. Forest Service. Northern Research Station.
|
| Other Form: |
Print version: FIA national assessment of data quality for forest health indicators. Newtown Square, PA : U.S. Dept. of Agriculture, Forest Service, Northern Research Station, [2009] (DLC) 2009438992 (OCoLC)506577739 |
| Gpo Item No. |
0083-B-06 (online) |
| Sudoc No. |
A 13.88:GTR-NRS-53 |
|
