Finnish Forest Research Institute  Metla

Press release 23.11.2006

New models for predicting regeneration establishment

Models have been developed that predict the success of forest regeneration and show differences between the established seedling stands. Mounding was shown to be the most effective site preparation method on Norway spruce plantations. However, the results show a good deal of variation that cannot be predicted on the basis of the explanatory variables relating to the regeneration method or site fertility.

The Finnish Forest Research Institute (Metla) has worked together with the Forestry Centres and Forest Management Associations to develop a method to improve forest regeneration results. The project was financially supported by the Ministry of Agriculture and Forestry. Since 2000, regeneration survey data has been collected from private 3-5 year-old regeneration plantations in Southern Finland. The collected material is also applicable for modelling regeneration results. Metla has created models that represent the expected regeneration result and its variation with different regeneration methods and on various sites.

Mounding is an effective method to improve regeneration results on Norway spruce plantations. Compared with harrowing and patching, mounding increased the number of planted spruce seedlings on non-paludified sites by 7 % and on paludified sites by 31 %. The number of spruce seedlings on regeneration plots with no preparation treatment was 21 % less than that on prepared areas.

In addition to the mean result depending on the regeneration method, stochastic between-geographic area, between-stand and within-stand variation must also be taken into consideration. According to model simulations, 3-year old, harrowed regeneration Norway spruce stands of Myrtillus type (MT) produced an average of 1400 planted seedlings per hectare, but only 27 % of the plantations produced 1600 planted seedlings or more.

The models describing the regeneration establishment can be linked to the calculation software used in forest counselling provided to forest owners and to support decision making. The calculation program will then predict the expected result for the regeneration stand and the variation, as well as the expected sapling stand development and management need for the studied site.

The estimated regeneration models can also be linked to a stand simulator to predict the effects of different forest management alternatives on stand growth and development during the next 10-20 years. Forest planning programs should be able to propose the best possible regeneration alternative for the site and to predict the regeneration result and the expenses caused by the method.

Publication: Jari Miina & Timo Saksa. 2006. Predicting regeneration establishment in Norway spruce plantations using a multivariate multilevel model. New Forests 32(3): 265-283.

Additional information:

Press releases in English are available at:   Metlan RSS-newsfeed
Finnish Forest Research Institute - Metla/Communications
Unioninkatu 40 A, FI-00170 HELSINKI, FINLAND. Phone +358 10 211 2000    E-mail: info @