Swiss National Forest Inventory,
Methods and models of the second assessment.
Brassel, P.; Lischke. (eds.) 2001
Birmensdorf, Swiss Federal Research Institute WSL, 336 pages.
The goal of the National Forest Inventory (NFI) is to record the current state and recent development of the Swiss forest in a representative and reproducible manner, using various data sources. To this end, in the second inventory (1993–1995) a combination of methods was used. Sampling followed a double sampling design: In the first phase aerial photos on a 0.5 x 0.5 km grid were used to estimate strata sizes, to identify forest plots and stocks outside the forest and to provide reference points for the field survey. In the second phase terrestrial sample plots on a 1.4 x 1.4 km grid were surveyed to record a number of variables having to do with the individual trees and stands, young growth and damage by game, as well as features of the surrounding areas.
The work and costs involved in the different steps of the terrestrial survey were recorded and evaluated. Ongoing training of the survey teams and control surveys ensured that the data was of a high quality. Further information was obtained by interviewing the local forest services, from external data sources and models describing the site conditions, and from specially designed studies of forest transportation systems and the effects of game browsing on tree growth.
The data were stored in a relational database and evaluated using statistical software developed specifically for this purpose. Static models were used for the evaluation of the following complex forest characteristics: the volume of standing and cut timber, tree growth, the work and cost involved in timber felling and extraction, the sustainability of forest regeneration, the protection provided by the forest against avalanches and rockfall, its recreational value, and the biotope values of the stands and forest border areas.
Furthermore, a dynamic model was developed to yield prognoses of the future development of each single tree depending on management scenarios. The models were supplemented by studies of error and uncertainty propagation to ensure good quality output variables. The raw and derived variables were comprehensively documented.
Keywords: multipurpose forest inventory, double sampling, terrestrial survey, aerial photography, growing stock, increment, forest growth model, sustainable regeneration, game browsing, forest functions