Dear Netters, I have had several requests for information about Woodstock from readers of the forest and fmdss-l lists, and Sam Radcliffe <samr@gbcwke.sol.net> suggested that I make a general informational posting. So, here goes ... (My apologies to those of you on both lists for any duplication, or those who have already received this as a result of a personal request.) Woodstock was developed here at the University of New Brunswick in cooperation with REMSOFT, a firm specializing in resource management software (particularly fire weather and fire modeling software). Woodstock is a forest modeling system, not a growth and yield modeling system. It is similar in purpose to FORPLAN, FORMAN and other tools used for timber harvest scheduling. Actually, Woodstock is not just limited to timber harvest scheduling. It is currently being used: as a vegetation management tool in Banff National Park (British Columbia); to determine the effect of changing prices on timber supply (i.e., economic timber supply); and to model the impacts of climatic change. Woodstock actually uses some of the best ideas from a number of earlier tools and combines them into one system. The result is, I think, more powerful and more intuitive than other forest modeling systems. The basic premise behind Woodstock is that a modeler can view a problem in many different ways and may wish to use different analytical techniques. So instead of limiting the modeler to one particular analytical technique, it provides three techniques (with a fourth one being developed). The modeler can build a model of the successional patterns in a forest and the responses to interventions/treatments. Then with a few simple changes or additions, any of the three analytical techniques can be applied to the same forest model. This provides greater flexibility to address various issues than other modeling systems, without having to build new models from scratch every time. The first technique is simple inventory projection (what I call deterministic simulation). This technique is similar to New Brunswick's FORMAN. The second technique is stochastic simulation where the interventions (eg. harvesting by private woodlot owners, fire occurrence) or the resulting responses are random variables. The third technique is linear programming, similar to the US Forest Services's FORPLAN. The fourth technique is really just an extension of linear programming - mixed integer programming - which would allow all-or-none types of decisions (similar to coordinated allocation choices in FORPLAN). Woodstock itself is the strategic planning component of the modeling system, which means that it is stratum-based and long-term ( > 1 rotation). Separate components are used for tactical planning of harvest blocks in the medium term (30-40 years). These components, called Crystal and Block, were initially developed as Master's projects at UNB. Crystal is a program to aggregate stands into preliminary harvest blocks according to a strategic harvest schedule (what some people call the spatial disaggregation process). Block uses Monte Carlo Integer Programming to allocate the preliminary blocks to harvest periods in such a way that restrictions on clearcut size and adjacency (green-up) delay are met. These two components are now being developed into a single component called Stanley. Here are some references: Walters, K.R. 1994. Design and development of a generalized forest management modeling system: WOODSTOCK. pp.190-196 IN: Paredes, G.L. (ed). Proceedings of the International Symposium on Systems Analysis and Management Decisions in Forestry, March 9-12, 1993, Valdivia, Chile. Universidad Austral Chile. Jamnick, M.S. and K.R. Walters. 1991. Harvest blocking, adjacency constraints and timber harvest volumes. IN: Proceedings of the 1991 symposium on systems analysis in forest resources, March 3-6, Charleston, SC. Compiled by: M.A. Buford. USDA For. Serv. Gen. Tech. Rep. SE-74. pp. 255-261. Jamnick, M.S. and K.R. Walters. 1993. Spatial and temporal allocation of stratum-based harvest schedules. Can. J. For. Res. 23:402-413. Clements, S.E., P.L. Dallain and M.S. Jamnick. 1990. An operational, spatially constrained harvest scheduling model. Can. J. For. Res. 20: 1438-1447. Please feel free to contact me for more information or copies of the above papers. For information about acquiring a copy of Woodstock and the other programs, contact Ugo Feunekes at REMSOFT: phone: 506-450-4511 fax: 506-459-7290 email: remsoft@nbnet.nb.ca snail mail: REMSOFT Inc. 620 George Street, Suite 5 Fredericton, New Brunswick Canada E3B 1K3 -- Hubert Burger T6LV@unb.ca Forestry and Environmental Management University of New Brunswick, Fredericton, N.B., Canada
Mail converted by
MHonArc 1.1.0