Dynamic economic modeling of forest markets is inherently uncertain and complex. Analysts make their best educated assessments of how market actors respond to changes in supply, demand, and price and use these assumptions to generate the forecasts in the model. These models can be used, for example, to project how forest inventories in a region might change over time in response to changes in demand for a particular wood product. However, varying the input assumptions can in many cases dramatically change model outputs. This variability leads to uncertainties in modeling results that are substantial and largely outweigh the benefits of estimating future directions in markets.
This tradeoff is evident in a new analysis published this week by researchers at Duke University. It concludes - I believe erroneously - that the use of wood pellets manufactured in the US Southeast for electricity generation in the European Union could represent a viable carbon reduction strategy. The study - based on a dynamic economic model - forecasts that more wood pellet demand from Europe is likely to result in increased removals from Southeastern forests over the next several decades. But it also results in an increase in the region's forested area and therefore gains in stored forest carbon - predominantly in new tree plantations.
There is substantial underlying uncertainty in this study, which likely compounds as the model projects further and further into the future. Specifically, the predicted results depend on a sequence of very tenuous "ifs". The presumed forest carbon changes occur if the price signals from increased demand are actually strong enough to generate additional planting; if there is land available to plant more trees; if planting trees on that land doesn't affect carbon associated with agricultural and other uses; if landowners have the resources and capacity to plant more trees. And so on.
Most importantly, the trends and dynamics predicted in the study are not borne out by historical data. According to the US Forest Service's Southern Forest Assessment, there was a dramatic, three-fold increase in demand for forest products in the US South between 1952 and the present - as reflected in forest product output. However, that significant historical increase in forest product demand coincided with a decrease in forested acreage in the region. This reality is contrary to view of the future captured in the Duke study.
NRDC believes that biophysical modeling is the better approach to assessing the carbon impacts of biomass use, rather than reliance on dynamic economic forecasting. Biophysical models simply track the carbon in forests as they are cut and regrow. The approach does not require many of the uncertain and conjectural assumptions and linkages required in dynamic economic forecasting.
Using a biophysical model developed by the Spatial Informatics Group, NRDC recently analyzed the carbon pollution from burning wood pellets from southeastern forests to produce electricity and compared those emissions with those from burning coal and natural gas. Our analysis, found here, shows that if the wood pellets are made of whole trees --even in relatively small proportions--they will emit carbon pollution comparable to or in excess of fossil fuels for more than five decades.