Current issue: 58(4)

Under compilation: 58(5)

Scopus CiteScore 2023: 3.5
Scopus ranking of open access forestry journals: 17^th
PlanS compliant

Articles containing the keyword 'linear mixed model'

Category : Research article

article id 10196, category Research article

Karri Uotila, Jari Miina, Timo Saksa, Ron Store, Kauko Kärkkäinen, Mika Härkönen. (2020). Low cost prediction of time consumption for pre-commercial thinning in Finland. Silva Fennica vol. 54 no. 1 article id 10196. https://doi.org/10.14214/sf.10196

Keywords: forest vegetation management; early cleaning; release treatment; linear mixed model; topographic wetness index; work productivity

Highlights: Time consumption (TC) in pre-commercial thinning (PCT) can be predicted by variables describing site and stands conditions and previous silvicultural management; Applying variables available in forest resources data the field-assessment of worksite difficulty factors is not needed; The TC model could facilitate the predictions of the labour costs of PCT in forest information systems.

Abstract | Full text in HTML | Full text in PDF | Author Info

The time consumption (TC) of pre-commercial thinning (PCT) varies greatly among sites, stands and forest workers. The TC in PCT is usually estimated by field-assessed work difficulty factors. In this study, a linear mixed model for the TC in PCT was prepared by utilizing forest resources data (FRD). The modelling data included 11 848 and validation data included 3035 worksites with TC information recorded by forest workers within the period of 2008–2018. The worksites represented a range of site and stand conditions across a broad geographical area in Finland. Site and stand characteristics and previous management logically explained the TC in PCT. The more fertile the site, the more working time was needed in PCT. On sites of medium fertility, TC in the initial PCT increased with stand age by 0.5 h ha^–1 yr^–1. Site wetness increased the TC. PCT in summer was more time consuming than in spring. Small areas were more time consuming to PCT per hectare than larger ones. The between-forest worker variation involved in the TC was as high as 35% of the variation unexplained by the TC model. The coefficient of determination in validation data was 19.3%, RMSE 4.75 h ha^–1 and bias –1.6%. The TC model based on FRD was slightly less precise than the one based on field-assessed work difficulty factors (removal quantity and type and terrain difficulty): RMSE 4.9 h ha^–1 vs. 4.1 h ha^–1 (52% vs. 43%). The TC model could be connected to forest information systems where it would facilitate the predictions of the labour costs of PCT without field-assessing work difficulty factors.

Uotila, Natural Resources Institute Finland (Luke), Natural resources, Latokartanonkaari 9, FI-00790 Helsinki, Finland E-mail: karri.uotila@luke.fi
Miina, Natural Resources Institute Finland (Luke), Natural resources, Yliopistokatu 6 B, FI-80100 Joensuu, Finland E-mail: jari.miina@luke.fi
Saksa, Natural Resources Institute Finland (Luke), Natural resources, Survontie 9, FI-40500 Jyväskylä, Finland E-mail: timo.saksa@luke.fi
Store, Natural Resources Institute Finland (Luke), Bioeconomy and environment, Teknologiakatu 7, FI-67100 Kokkola, Finland E-mail: ron.store@luke.fi
Kärkkäinen, E-mail: kauko.karkkainen@gmail.com
Härkönen, Tornator Oyj, Pielisentie 2–6, FI-81700 Lieksa, Finland E-mail: mika.harkonen@tornator.fi

article id 1005, category Research article

Marjut Turtiainen, Jari Miina, Kauko Salo, Juha-Pekka Hotanen. (2013). Empirical prediction models for the coverage and yields of cowberry in Finland. Silva Fennica vol. 47 no. 3 article id 1005. https://doi.org/10.14214/sf.1005

Keywords: generalized linear mixed model; abundance; berry yield; Vaccinium vitis-idaea L.

Highlights: The site fertility significantly affected the abundance of cowberry on mineral soils, spruce mires and pine mires; The stand basal area and dominant tree species were among the most important forest structural predictors in the model for the coverage; In the cowberry yield model developed for mineral soil sites, the stand basal area and coverage of cowberry plants were statistically significant predictors.

Abstract | Full text in HTML | Full text in PDF | Author Info

Empirical models for the coverage and berry yield of cowberry (Vaccinium vitis-idaea L.) were developed using generalized linear mixed models (GLMMs). The percentage coverage of cowberry was predicted as a function of site and stand characteristics using data from the Finnish National Forest Inventory (NFI) in 1995. The average annual yield, including the between-year variation in the yield, was predicted as a function of percentage coverage and stand characteristics using permanent experimental plots (MASI) established in different areas of Finland and measured in 2001-2012. The model for cowberry yields (Model 2) was developed for mineral soil forests. The model for the coverage (Model 1) was constructed so that it considers both mineral soil sites and also many other sites where cowberry occurs in the field layer. According to Model 1, the site fertility significantly affected the abundance of cowberry on mineral soils, spruce mires and pine mires. The stand basal area and dominant tree species were among the most important forest structural predictors in Model 1. The site fertility was not a significant predictor in the cowberry yield model. Instead, the stand basal area and coverage of cowberry plants were found to be statistically significant predictors in Model 2. The estimated models were used to predict the cowberry coverage, average annual yield and its 95 % confidence interval along with stand development. The models of this study can be used for multi-objective forest planning purposes.

Turtiainen, University of Eastern Finland, School of Forest Sciences, P.O. Box 111, FI-80101 Joensuu, Finland E-mail: marjut.turtiainen@uef.fi
Miina, Finnish Forest Research Institute, Joensuu Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: jari.miina@metla.fi
Salo, Finnish Forest Research Institute, Joensuu Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: kauko.salo@metla.fi
Hotanen, Finnish Forest Research Institute, Joensuu Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: juha-pekka.hotanen@metla.fi

article id 181, category Research article

Jari Miina, Juha-Pekka Hotanen, Kauko Salo. (2009). Modelling the abundance and temporal variation in the production of bilberry (Vaccinium myrtillus L.) in Finnish mineral soil forests. Silva Fennica vol. 43 no. 4 article id 181. https://doi.org/10.14214/sf.181

Keywords: vegetation; generalized linear mixed model; heath forest

Abstract | View details | Full text in PDF | Author Info

Miina, Finnish Forest Research Institute, Joensuu Research Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: jari.miina@metla.fi
Hotanen, Finnish Forest Research Institute, Joensuu Research Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: jph@nn.fi
Salo, Finnish Forest Research Institute, Joensuu Research Unit, P.O. Box 68, FI-80101 Joensuu, Finland E-mail: ks@nn.fi

article id 504, category Research article

Sylvain Jutras, Hannu Hökkä, Virpi Alenius, Hannu Salminen. (2003). Modeling mortality of individual trees in drained peatland sites in Finland. Silva Fennica vol. 37 no. 2 article id 504. https://doi.org/10.14214/sf.504

Keywords: Pinus sylvestris; Betula pubescens; simulation; peatlands; mortality; generalized linear mixed models; multilevel models

Abstract | View details | Full text in PDF | Author Info

Multilevel logistic regression models were constructed to predict the 5-year mortality of Scots pine (Pinus sylvestris L.) and pubescent birch (Betula pubescens Ehrh.) growing in drained peatland stands in northern and central Finland. Data concerning tree mortality were obtained from two successive measurements of the National Forest Inventory-based permanent sample plot data base covering pure and mixed stands of Scots pine and pubescent birch. In the modeling data, Scots pine showed an average observed mortality of 2.73% compared to 2.98% for pubescent birch. In the model construction, stepwise logistic regression and multilevel models methods were applied, the latter making it possible to address the hierarchical data, thus obtaining unbiased estimates for model parameters. For both species, mortality was explained by tree size, competitive position, stand density, species admixture, and site quality. The expected need for ditch network maintenance or re-paludification did not influence mortality. The multilevel models showed the lowest bias in the modeling data. The models were further validated against independent test data and by embedding them in a stand simulator. In 100-year simulations with different initial stand conditions, the models resulted in a 72% and 66% higher total mortality rate for the stem numbers of pine and birch, respectively, compared to previously used mortality models. The developed models are expected to improve the accuracy of stand forecasts in drained peatland sites.

Jutras, Département des sciences du bois et de la forêt, Université Laval, Ste-Foy, Québec, G1K 7P4, Canada E-mail: sj@nn.ca
Hökkä, Finnish Forest Research Institute, Rovaniemi Research Station, P.O. Box 16, FIN 96301 Rovaniemi, Finland E-mail: hannu.hokka@metla.fi
Alenius, Finnish Forest Research Institute, Rovaniemi Research Station, P.O. Box 16, FIN 96301 Rovaniemi, Finland E-mail: va@nn.fi
Salminen, Finnish Forest Research Institute, Rovaniemi Research Station, P.O. Box 16, FIN 96301 Rovaniemi, Finland E-mail: hs@nn.fi

Category : Research note

article id 10679, category Research note

Jari Miina, Mikko Kurttila. (2022). A model for the sap yield of birches tapped by citizen scientists. Silva Fennica vol. 56 no. 2 article id 10679. https://doi.org/10.14214/sf.10679

Keywords: Betula spp.; non-timber forest product; linear mixed model; crowdsourcing

Highlights: Tree diameter and mean stand height positively affected the sap yield of birches; The sap yield varied between trees, stands, and years; The sap yield model can be utilised in profitability analyses for sap tapping.

Abstract | Full text in HTML | Full text in PDF | Author Info

The sap yield of birches (Betula pendula Roth and B. pubescens Ehrh.) was modelled as a function of tree diameter (girth) at breast height, as well as site and stand characteristics measured and reported by citizen scientists representing mainly non-industrial private forest owners in the South Savo, North Karelia, and Northern Ostrobothnia regions in Finland. Birches (tree species not recorded) growing on both mineral and peatland sites were tapped during the springs of 2019 and 2020. Citizen scientists were mainly voluntary forest owners who received the instructions and equipment (spouts, drop lines and buckets) for collecting sap from three birches of different diameters in the same birch stand. Citizen scientists were instructed to measure and report the sap yield and girth of the trees, as well as stand characteristics from the forest resource data, if available. Based on the linear mixed model fitted to the data, the sap yield increased with the increasing tree diameter and mean stand height, and varied between years, stands, and trees; between-region variation was not significant. In a birch stand, the simulated total sap yield ha^–1 was depended on the average tree size and the stem number ha^–1 and was at its highest just before the first commercial thinning and again before the second thinning. The sap model can be used to predict the necessary sap yield in profitability analyses for sap tapping.

Miina, Natural Resources Institute Finland (Luke), Yliopistokatu 6 B, FI-80100 Joensuu, Finland https://orcid.org/0000-0002-8639-4383 E-mail: jari.miina@luke.fi
Kurttila, Natural Resources Institute Finland (Luke), Yliopistokatu 6 B, FI-80100 Joensuu, Finland https://orcid.org/0000-0001-5290-4771 E-mail: mikko.kurttila@luke.fi

article id 1573, category Research note

Marjut Turtiainen, Jari Miina, Kauko Salo, Juha-Pekka Hotanen. (2016). Modelling the coverage and annual variation in bilberry yield in Finland. Silva Fennica vol. 50 no. 4 article id 1573. https://doi.org/10.14214/sf.1573

Keywords: abundance; berry yield; generalised linear mixed model; Vaccinium myrtillus L.

Highlights: The highest bilberry coverage was found in mesic heath forests and fell forests; On peatlands the coverage was, on average, lower than on mineral soil sites; The approach introduced in this study to calculating annual berry yield indices is a promising way for estimating total annual bilberry yields over a given period of time.

Abstract | Full text in HTML | Full text in PDF | Author Info

The coverage of bilberry (Vaccinium myrtillus L.) was modelled as a function of site and stand characteristics using the permanent sample plots of the National Forest Inventory (NFI) (Model 1). The sample sites consisted of mineral soil forests as well as fells and peatland sites. Annual variation in the bilberry yield (Model 2) was analysed based on measurements over 2001–2014 in the permanent sample plots (so-called MASI plots) in various areas of Finland. We derived annual bilberry yield indices from the year effects of Model 2 and investigated whether these indices could be used to estimate annual variation in bilberry crops in Finland. The highest bilberry coverage was found in mesic heath forests and fell forests. On peatlands the coverage was, on average, lower than on mineral soil sites; the peatland sites with most bilberry coverage were meso-oligotrophic and oligotrophic spruce mires and oligotrophic pine mires. Our bilberry yield indices showed similar variation to those derived from the mean annual berry yields reported and calculated earlier using the MASI plots; the correlation between the indices was 0.795. This approach to calculating annual berry yield indices is a promising way for estimating total annual bilberry yields over a given period of time. Models 1 and 2 can be used in conjunction with the Miina et al.’s (2009) bilberry yield model when bilberry coverage, average annual yield and annual variation in the yield are to be predicted in forest planning.

Turtiainen, University of Eastern Finland, School of Forest Sciences, P.O. Box 111, FI-80101 Joensuu, Finland E-mail: marjut.turtiainen@uef.fi
Miina, Natural Resources Institute Finland (Luke), Management and Production of Renewable Resources, Box 68, FI-80101 Joensuu, Finland E-mail: jari.miina@luke.fi
Salo, Natural Resources Institute Finland (Luke), Bio-based Business and Industry, Box 68, FI-80101 Joensuu, Finland E-mail: kauko.salo@luke.fi
Hotanen, Natural Resources Institute Finland (Luke), Management and Production of Renewable Resources, Box 68, FI-80101 Joensuu, Finland E-mail: juha-pekka.hotanen@luke.fi

Click this link to register to Silva Fennica.

If you are a registered user, log in to save your selected articles for later access.

Contents alert

Your selected articles

Follow @SilvaFennica