Uncertainty assessment of large-scale forest growth predictions based on a transition-matrix model in Catalonia

Abstract

Abstract& Key message When predicting forest growth at a regional or national level, uncertainty arises from the sampling and the prediction model. Using a transition-matrix model, we made predictions for the whole Catalonian forest over an 11-year interval. It turned out that the sampling was the major source of uncertainty and accounted for at least 60 % of the total uncertainty. & Context With the development of new policies to mitigate global warming and to protect biodiversity, there is a growing interest in large-scale forest growth models. Their predictions are affected by many sources of uncertainty such as the sampling error, errors in the estimates of the model parameters, and residual errors. Quantifying the total uncertainty of those predictions helps to evaluate the risk of making a wrong decision. & Aims In this paper, we quantified the contribution of the sampling error and the model-related errors to the total uncertainty of predictions from a large-scale growth model in Catalonia. & Methods The model was based on a transition-matrix approach and predicted tree frequencies by species group and 5-cm diameter class over an 11-year time step. Using Monte Carlo techniques, we propagated the sampling error and the model-related errors to quantify their contribution to the total uncertainty. & Results The sampling variance accounted for at least 60 % of the total variance in smaller diameter classes, with this percentage increasing up to 90 % in larger diameter classes. & Conclusion Among the few possible options to reduce sampling uncertainty, we suggest improving the variance-covariance estimator of the predictions in order to better account for the multivariate framework and the changing plot size.

Estimation and Uncertainty of the Mixing Effects on Scots Pine—European Beech Productivity from National Forest Inventories Data

“…Moreover, the use of models always implies a certain degree of uncertainty in the predictions, stemming from different sources of errors such as the model parameter estimates or independent variable sampling and measurement errors [43]. The importance of these errors in predicting growth has been studied in forest science since the mid-1980s, first at plot or stand level and more recently at regional or national level [44], the general consensus being that sampling errors are the most important component of the uncertainty of predictions [45].…”

Improving local forest growth prediction by terrain-derived attributes, airborne γ-ray, and leaf area index

“…In most cases, models fitted for larger areas tend to be locally or even regionally biased, and there is always a need for localization (Räty & Kangas 2007;Räty & Kangas 2010;Sironen 2009;Fortin et al 2016). The uncertainty arising from applying growth models at a national level can reach up to 60% (Fortin et al 2016).…”

“…In most cases, models fitted for larger areas tend to be locally or even regionally biased, and there is always a need for localization (Räty & Kangas 2007;Räty & Kangas 2010;Sironen 2009;Fortin et al 2016). The uncertainty arising from applying growth models at a national level can reach up to 60% (Fortin et al 2016). Thus, it is generally challenging to explain the whole range of variations in a sample-based national model fitted for a large area, even for a species-specific measure, and as a result, there is a genuine need for localization (Sterba et al 2002;Burkhart & Tomé 2012).…”

Editorial: “Forest Inventories at the European level”

“…(iii) The used models and tools allowing a quantitative assessment of forest resources at different scales (McRoberts and Westfall 2015;McRoberts et al 2015;Mantau et al 2016) and projections for investigating the different management strategy options for wood production and/or mitigation of the occurring climatic changes Fortin et al 2016).…”

“…Two contributions also focus on uncertainty in the estimates of present (McRoberts and Westfall 2015) and future (Fortin et al 2016) forest resource attributes, pointing out to a growing need for precision in the assessment of these resources. As original efforts to date, Barreiro et al (2016) and Bosela et al (2016) further offers comprehensive reviews across European countries of (1) existing models for wood resource projection and (2) approaches to stem quality assortments and directions for their harmonization.…”