DOFAW Timber Inventory
OVERVIEW OF THE DOFAW-HFCI TIMBER INVENTORY SYSTEM
Between 1997-2000 the Division of Forestry and Wildlife (DOFAW) and the Hawaii Forestry and Communities Initiative (HFCI) cooperatively conducted several timber inventory projects throughout Hawaii. This document outlines the principal timber inventory design components employed during those projects. It is intended to provide a reference for consultants or project coordinators who wish to conduct timber inventory for their own management planning or as a contract service.
The following survey methodology was designed primarily to function as a stand inventory for project areas that were several thousand acres in size. The intention was to provide data suitable for supporting long-term management and planning efforts for intensively managed State forestlands. This methodology was not intended to provide volume estimates with precision levels high enough to support timber sales or appraisals relating to timber value. In cases where specified stands are to be sold, timber cruising methods incorporating greater sampling intensity are recommended to provide highly accurate volume estimates.
This stand inventory methodology was designed to provide volume estimates in cubic feet for each forest stand or type, sorted by log size classes, or by species in one-inch diameter classes as required. The target survey precision for this design was mean volume estimates where standard error values would not exceed 20% of the mean.
The ability to design, oversee and implement a timber inventory project is a complex task. Those interested in pursuing such activity should have at minimum a working knowledge of the following skills: forest mensuration, forest management, contract negotiation and management, technical report writing, database management, statistical analyses, mapping (geographic information systems), and personnel management. If a potential project coordinator lacks some of these specific skills, he/she could investigate the possibility of sub-contracting the required services. This document is written assuming that the reader has a working knowledge of the skills listed above.
Inventory Design and Layout
Maps: Prior to initiation of field survey work, a map of each project area was developed using some or all of the following resources: boundary data from past surveys or maps, aerial photography, GPS, satellite imagery, and field reconnaissance. These types of data were incorporated into a project base map using GIS technology. Field survey crews would then modify the map boundaries based on observations recorded during their field surveys. Boundary updates included combining or partitioning existing map polygons, defining new polygons, or changing attributes related to polygons such as tree species or timber type.
Sample plots: The DOFAW-HFCI system utilized nested fixed radius plots. Main plots were either 0.20 or 0.10 acres in size. Sampling where plots contain 7-10 trees is most efficient, so the main plot size selected should reflect tree density in the project area. Sub plots were 0.05 acres in size, and were used to tally sapling regeneration of tree species on each plot. Variable plot sampling was not employed for two principal reasons. Most of the areas surveyed had quite dense groundcover and understory vegetation, making sighting of trees from plot center difficult and inefficient. In addition, plot centers and individual plot trees were monumented, creating the option for future re-measurements that could lead to growth and site productivity analyses.
Pre- vs. post-stratification: For project areas that cover several hundred or more acres, or areas comprised of timber stands or forest types that are typically similar in acreage, systematic sampling with post-stratification of plot data is possible. In this method, plots are sampled in a systematic pattern throughout the entire study area. The DOFAW-HFCI system utilized a grid overlay map with one point (i.e. potential plot location) every 5 acres. Sampling was conducted in either a 2x2 (one plot per 20 acres) or 3x3 (one plot per 45 acres) pattern, depending on the size of the project area.
After stand or type boundary updates were made, plot data were subsequently aggregated and analyzed by timber stand or type. At that time, if particular stands or types were represented by plot sample sizes that were too low (or sampling error was unacceptably high), additional plots were randomly selected and sampled in those areas. Approximately 30 plots per forest type was the target sampling intensity, irrespective of the acreage of that type. As a general rule, sampling additional plots beyond 30 does little to increase sampling precision of timber volume estimates.
For project areas that are relatively small in size, or areas comprised of heterogeneous mixes of timber stands or forest types, stratified random sampling is recommended. In this approach, initial mapping of stand or type boundaries should be given greater emphasis. Once the timber or type map was pre-stratified, typically 3-8 plots were randomly selected and sampled within a given stand or type, depending on its relative size and importance.
Data analyses and reporting: Field data were analyzed with Forestry Projection System v. 5.3b, software developed by Forest Biometrics (http://www.forestbiometrics.com/). Each timber inventory project was summarized in a written report, examples of which can be found in PDF format at the DOFAW web site (http://www.hawaii.gov/dlnr/dofaw/pubs/index.html).
The information above outlines some of the principal philosophies, methods and tools used to implement DOFAW-HFCI timber inventory. Any individual or organization that wishes to obtain additional details regarding these methods is encouraged to contact the DOFAW Survey Forester:
Sheri S. Mann
If several individuals or organizations express interest in additional information, DOFAW-HFCI will consider the possibility of offering a short training course.