Biodiversity of areas under annual crops, perennial grass crops or plantations may be different from that of the nature that they replace(d). To the extent that transport biofuel production leads to losses of natural habitats, there is apparently a consistent negative effect on species diversity (Fahrig 2003). The effect appears to be stronger at higher trophic levels (Dobson et al. 2006).

In practice, expansion of biofuel production is often associated with cutting forests, and this will probably also hold for future expansion (Gurgel et al. 2007). In such cases, land with annual biofuel crops functions in a number of other ways dif­ferently from a forest. Vegetation influences surface roughness, which in turn may alter wind, turbulence and moisture convergence, and forest and cropland may be significantly different in this respect (Notaro et al. 2006). Change of forest into crop­land in semi-arid climates may increase dust emissions, which in turn may change radiative forcing (Betts 2007). The albedo of arable land, on which biofuel crops are grown, is often different from the albedo of land under natural vegetation, and actual differences may also be crop dependent (Gustafsson et al. 2004; Schneider et al. 2004; McPherson 2007). The albedo is a measure of the reflection of solar ra­diation by the earth’s surface (including vegetation), which in turn is a determinant of net radiation. Biofuel crops tend to be shorter and have less foliage than forests, and so the surface albedo of arable land with biofuel crops tends to be higher than the albedo of a forest. The release of water to the atmosphere by evapotranspira — tion linked to biofuel crops may also be different from natural vegetation or other crops (Gustafsson et al. 2004; McPherson 2007). The difference in evapotranspira — tion may have impacts on soils and the atmosphere. An example of the former is the large-scale salinization of soils in Australia, following the replacement of native woody vegetation by crops. The change in evapotranspiration caused groundwater tables to rise, and the rising water mobilized salt (Folke et al. 2004).

As to the atmospheric effect of changed albedo, it may be noted that, besides net radiation, evapotranspiration influences local climate, including temperature (Schneider et al. 2004; Notaro et al. 2006; McPherson 2007), and there may also be an impact on precipitation (Liu et al. 2006). Moderate and local deforestation may lead locally to enhanced rainfall (Malhi et al. 2008). Also, in tropical regions, replacement of forest by annual biofuel crops may cause warming, mainly due to a decrease in evaporation and cloud cover (Betts et al. 2007). In temperate regions, the overall effect of turning forest into cropland may lead to regional cooling that may be partly offset by increased aerosol concentrations (Betts 2007).

When changes in vegetation are widespread, there may be an impact on mesoscale climate (McPherson 2007; Liu et al. 2006). For instance, a simulation study assum­ing a major reduction in forested area and a major increase in annual cropping in Amazonia found a substantial reduction in precipitation over Amazonia and a posi­tive radiation forcing at the top of the atmosphere linked to increased airborne dust (Betts 2007). Another model study suggests that removal of 30-40% of the Ama­zonian rainforest may push much of Amazonia into a permanently drier climate (Malhi et al. 2008). Such a climate change may lead to additional damage to the current Amazonian rainforest (Nepstad et al. 2008).

Mesoscale changes in climate may have knock-on effects. For instance, mesoscale changes in Amazonia may in turn have effects on precipitation in the Northern Hemisphere and across the rest of South America (Malhi et al. 2008). A simula­tion regarding deforestation in Indonesia has suggested that the surrounding ocean surfaces may be warmed and that this may have a widespread impact on atmospheric circulation (Delire et al. 2001).

It has furthermore been found that cultivation of land for arable crops reduces the uptake rate of CH4. In an experiment in Rothamsted, it appeared that extended (150-year) cultivation of arable land decreased CH4 uptake and oxidation by 85%, if compared to that in the soil under woodland (Powlson et al. 1997). Also, the ‘leakiness’ of nutrients and, more in general, soluble minerals is much increased in agricultural land, if compared with soils under native forest cover (Williams and Melack 1997). Moreover, forests are better in the capture of nutrients (such as P and N) from air than annual crops (Lawrence et al. 2007). So the function of annual crops on arable land is much different from that of forests.

But how about the functioning of tree plantations, which would seem rather sim­ilar to forests? Differences in biodiversity between oil palm plantations and the forests which these plantations replaced have been analyzed by Danielsen et al. (2008). They found that total species richness of vertebrates, invertebrates and flora on oil palm plantations was impoverished. Similarly, Lindenmayer and Hobbs (2004) found reduced faunal diversity in Australian eucalypt plantations if com­pared with native forests. Barlow et al. (2007) studied differences between native forest and eucalypt plantations in Brazil for 15 taxonomic groups and found overall diversity reduced in plantations, with major differences in biodiversity change be­tween the taxa. They pointed out, however, that faunal diversity can be improved by integrating elements of original biota into plantations, by modifications to plantation management (e. g. regarding harvesting and thinning) and by having extensive areas of remnant native vegetation adjacent to plantations.

The functioning of tree plantations has been found to be different from the func­tioning of primary forests. This has consequences for ecosystem services. Com­parison of tropical tree plantations with secondary tropical forests showed, for in­stance, that in secondary forests, root densities and nutrient concentration in roots were higher and root penetration was deeper in forests than in plantations (Lugo 1992). This allows secondary forests to better recapture nutrients which become available by mineralization and could otherwise be lost to water and the atmosphere. In the highly productive plantations, which are important for achieving a large fu­ture biofuel supply (e. g. Hoogwijk et al. 2003), there is furthermore an intensive use of herbicides and other pesticides (Tuskan 1998; Robison and Raffa 1998) which will cause increased leakiness regarding nutrients and will limit nitrogen fix­ation.

Primary forests are more efficient than plantations in protecting watersheds, in reducing peak flows, soil erosion and nutrient emissions, in maintaining good water quality, in stabilizing local climate and in generating OH radicals that are impor­tant cleansing agents in the atmosphere (Hartshorn 1995; Perry 1998; Daily 2000; Monson and Holland 2001; Brauman et al. 2007; Wallace 2007). The risk of pests tends to be increased in plantations if compared with native biota, but such risk may be reduced by integrating patches of native vegetation into plantations (Lin — denmayer and Hobbs 2004). Li et al. (2005) found that secondary forests may stock more long-term soil organic carbon than plantations in the wet tropics. This is rel­evant to the sustainability of plantations (cf. Chap. 3) and to the net emission of greenhouse gases from plantation-derived biofuels (cf. Chap. 4). Still, differences in non-monetary ecosystem services between primary forests and arable land tend to be larger than those between forests and plantations (Brauman et al. 2007).

The effect of habitat loss on non-monetary ecosystem services also has world­wide aspects. These are closely linked to the biogeochemical cycles of relatively mobile elements, of which the carbon cycle is a good example. There are sev­eral links between natural biomass and worldwide atmospheric concentrations of CO2. Reduced C sequestration in ecosystems, which is often associated with ex­panded biofuel production (see Chap. 4), increases the atmospheric CO2 concentra­tion (Houghton 2007). It is furthermore likely that natural terrestrial biomass may give rise to a better negative feedback by fixing increasing amounts of CO2 when the atmospheric concentration thereof increases than most agro-ecosystems (Cao and Woodward 1998; Luo 2007). And, overall, natural systems are better at seques­trating carbon than agricultural systems (Vitousek et al. 1997). Moreover, when land use is changed back from agriculture to nature, the functioning of secondary forests in biogeochemical cycles such as the carbon cycle may for a long time be substan­tially different from primary forests (Grau et al. 2003).

Losses of natural terrestrial biomass are contributing significantly to the increase in the atmospheric CO2 concentration (IPCC 2001; Houghton 2007). This may also have monetary effects. An increase in atmospheric CO2 concentration will (ceteris paribus) on average increase costs of production and consumption in the world econ­omy (IPCC 2001; Stern 2007).

All in all, to the extent that transport biofuel production leads to habitat loss for living nature, this affects functions of nature that can be considered beneficial to mankind such as its contribution to a benign biological, chemical and physical environment and socio-cultural fulfilment (Vitousek et al. 1997; Wallace 2007).