Difference between revisions of "FAQ:How reliable are the growth projections?"
| (2 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
[[Newsdate::{{REVISIONYEAR}}-{{REVISIONMONTH}}-{{REVISIONDAY2}}]] [[language::en| ]] | [[Newsdate::{{REVISIONYEAR}}-{{REVISIONMONTH}}-{{REVISIONDAY2}}]] [[language::en| ]] | ||
| − | + | <br>[[:FAQ | Back to FAQ list]] | |
| − | ==== | + | ====How reliable are the growth projections in Heureka?==== |
[[FaqAnswer::Heureka’s growth functions are based on data for whole Sweden, for all forest types. Growth is a highly stochastic process and of course growth predictions will be uncertain. Still, for a specific even-aged stand you can expect that the predicted growth will be within 15 % of the actual outcome in 70 % of the cases independent of time horizon (Fahlvik et. al 2014), given there are no catastrophic events such as fire or storm. For larger forest areas, the total prediction error is significantly smaller. | [[FaqAnswer::Heureka’s growth functions are based on data for whole Sweden, for all forest types. Growth is a highly stochastic process and of course growth predictions will be uncertain. Still, for a specific even-aged stand you can expect that the predicted growth will be within 15 % of the actual outcome in 70 % of the cases independent of time horizon (Fahlvik et. al 2014), given there are no catastrophic events such as fire or storm. For larger forest areas, the total prediction error is significantly smaller. | ||
Latest revision as of 21:14, 3 June 2015
2015-06-03
Back to FAQ list
How reliable are the growth projections in Heureka?
Heureka’s growth functions are based on data for whole Sweden, for all forest types. Growth is a highly stochastic process and of course growth predictions will be uncertain. Still, for a specific even-aged stand you can expect that the predicted growth will be within 15 % of the actual outcome in 70 % of the cases independent of time horizon (Fahlvik et. al 2014), given there are no catastrophic events such as fire or storm. For larger forest areas, the total prediction error is significantly smaller.
The functions have also been tested on plots subject to uneven-aged management and for a specific stand you can expect that the predicted growth will be within 25 % of the actual outcome in 70 % of the cases. Uneven-aged management, as simulated in Heureka, relies on natural ingrowth of new trees instead of artifical regeneration. However, the number of ingrown trees varies considerably between stands and is difficult to project. The height growth for ingrown trees after selection felling is underestimated. The standing stock that consists of ingrown trees will increase with time and thus the projected forest state will be more and more uncertain.
In short, the uncertainties are largest in modelling of mortality, natural ingrowth of new trees, and establishment of young stands.