Survey of Energy Resources 2007
The existing large-scale use of forest resources for bioenergy detailed above implies that future expansion of biomass supplies will be from two very distinct streams: the residues associated with current agricultural commodity production and processing, and the planting of energy crops on available land. The latter is of course an option that is unavailable as a large-scale contribution in small highly-populated countries such as the UK and the Netherlands. But for countries such as Argentina, Brazil, Canada, the USA and Russia, as well as some in the eastern part of Europe, increased land utilisation is feasible.
Terrestrial biomass production can be determined from the characteristics of the different ecoregions and their associated biomes. Each ecoregion is defined by its climate, elevation and soils. Climate parameters include temperature, precipitation and solar insolation according to the pattern of the seasons and weather extremes. Satellite observations at several wavelengths of light can be used to prepare an estimate of the density of the green vegetation over the surface of the earth, by means of the normalized difference vegetation index (NDVI). The global estimates of net primary productivity (NPP - which is equivalent to the annual amount of photosynthetic carbon fixed in terrestrial biomass) were historically carried out by local sampling measurements in representative biomes that had been more or less mapped. Such extrapolation is now a thing of the past, as satellite sensor technology, mathematical models and geographic information systems (GIS) can now provide almost daily values of the NPP (Running, Nemani, et al., 2004) derived from the NDVI. The algorithm that provides the satellite-based value of NPP assumes a direct relationship between absorbed solar energy and the indices of vegetation, followed by biophysical restrictions such as temperature and water availability. The 2001 terrestrial value is estimated at 55.5 Pg carbon, or 489 g carbon/m2 on vegetated land (Running, Nemani, et al., 2004) equivalent to 1.665 ZJ of primary energy.
These methods can be extended to predict biomass productivity with different crops that have their responses to the biophysical parameters incorporated into calibrated crop growth models. Typically the biome characteristics are described as Agroecological Zones (AEZ) which following an environmental approach, provides a standardised framework for the characterisation of climate, soil and terrain conditions relevant to crop and forest species production, and uses environmental matching procedures to identify limitations of prevailing climate, soil and terrain for assumed management objectives. Such techniques have been used, for example to evaluate the potential for energy crops like miscanthus and willow in eastern Europe and Asia (Fischer, Prieler, et al., 2005), and incorporating land-use restrictions in the GIS treatment. Because these techniques often assume crops better suited to the biome than those currently occupying the land, and the use of best practices in crop management, the yields are usually greater than those generated by the current land use.
However, there are large-scale examples of how well the agricultural system can perform when all of the tools of modern biology are brought to bear. One of these is shown in the over 100-year record of corn yields in the contiguous 48 states of the USA in Fig. 9-7. The data are drawn from the US Department of Agriculture (USDA) and it can be seen that up until World War II the yield was quite low and the agriculture can only be described as extensive, using mainly animal power and low inputs of fertiliser. The last 50 years of the 20th century saw, in turn: the transformation to a mechanised agriculture; the extensive use of fertiliser; the development of F1 hybrids; genetically engineered resistance to pests and weed-control and now the application of genomics and metabonics to optimise the breeding and development of corn cultivars for different regions. The next steps will include the development of corn crops that are specific in their yield of fermentable sugars in the production of ethanol.
The application of genetic engineering to increase drought tolerance, to improve nutrient-use efficiencies also promises to increase the availability of suitable land for growing improved crops.
Energy crops in the form of tree crops and herbaceous energy crops such as switchgrass in the USA and miscanthus in Europe have been the subject of considerable research and development, which is leading towards deployment. In a study conducted by the National Laboratories in the USA, the USDA and the Forest Service (Perlack, Wright, et al., 2005) established that the physical near-term sustainable biomass potential was over 1 Gt of biomass (18 EJ of primary energy equivalent) that could theoretically replace the crude oil currently imported into the US. A GIS-based atlas of the biomass resources of the United States is available from NREL (Milbrandt, 2005).
Forestlands account for an estimated 33% of the US land area. The US Department of Energy (DOE) and the USDA estimate that 333 mt of biomass feedstock are available annually from forestlands. This includes: 47 mt from harvesting for fuelwood, 131 mt from wood processing and pulp and paper mills, 43 mt from urban wood residues, 58 mt from logging and site-clearing operations, and 55 mt from forest fire-hazard reduction efforts. In evaluating the feedstock to be generated from logging and site-clearing and fire-hazard thinning, it was assumed that all forestland not currently accessible by roadways were excluded; all environmentally sensitive areas were excluded; equipment recovery limitations were considered; and recoverable forest materials categorised as either conventional forest products or biomass for bioenergy and biobased products. Agricultural lands are estimated to account for approximately 46% of the entire US land base with 26% consisting of grassland pasture and range, and 20% consisting of cropland. The DOE and USDA estimate that biomass feedstock available from agricultural lands, while still meeting food, feed and export demands, could supply 910 mt of biomass feedstock annually. This includes the following: 390 mt from crop residues, 343 mt from perennial energy crops (both tree and herbaceous), 80 mt of grains for biofuels, and 96 mt from animal manure, process residue, and miscellaneous feedstocks on a dry basis.
Nearly all of the material identified as a near-term resource in the USDOE study is comprised of lignocellulosic material, thus explaining the high degree of research investment into the conversion of these materials into biofuels. A key challenge is in realising the sugar content for bioconversions to first-generation biofuels such as ethanol and eventually to second-generation fuels. Alternative pathways could include thermal conversion and the Fischer-Tropsch process into middle distillates.
For the near-term development, much depends on utilising the process residues generated today. In the section above on electricity much of the opportunity was identified with so-called black liquor in the pulping industry. This is a very large-scale resource, as is the bagasse generated in the production of cane sugar or ethanol, which in total is about 160 mt of dry material worldwide, but has the same advantage in that it is available at the mill site for no additional cost of harvesting and collection.