In this model, a contractor or broker (or other 3rd party) purchases the biomass and undertakes the necessary activity in feeding it into the downstream supply chain. Volumes/energy content must be determined on site at, or after, harvesting. Once again, established BEF can be utilised or the volumes can be estimated from pile dimensions and local guidelines. The advantage for the forest owner is one of a small (but somewhat uncertain) income on site and no responsibility for costs to contractors and uncertainties of supply. The benefit to the energy conversion plant is a longer term relationship with a single supplier representing many forest owners.

Standard models for determining energy content in the supply chain are given below:

6.8.3.1 At the plant

Ultimately the bioenergy plant determines and pays for the energy content of

biomass being delivered though its gates. This is done for chips as follows:

1. The (registered) truck driver notes the origin of the biomass on the incoming weighbridge bill, which includes a timestamp. The registration of trucks includes the details of the supplier for whom they are transporting.

2. The driver (or staff member) takes a bucket sample of the chips while they are being poured into the chip bunker — ensuring it is as representative as possible. Some plants are equipped with fully automated sampling devices that bore into the load while still on the truck.

3. The sample is placed in a drying oven together with the weighbridge bill

4. After 24 h at 105° C, the dry matter content of the load is calculated and remuneration to the supplier is affected.

When it comes to other forms of biomass (FT or residues), it is not possible to make accurate estimates of energy content before the material is chipped and dried. With well-developed delivery systems, e. g., from uniform plantation forestry, conversion figures including norms and standard deviations can be developed rapidly and improved on with time. Here, the supplier is paid a value derived from e. g., mean moisture content per specie and season.

Free water in excess of 65 % MC can be extracted by compressing the biomass with a press. In this way the MC can be reduced to about 20 %. To remove the remaining water, the biomass must either be air or oven dried. The drying time depends on the particle size and the composition of the biomass used. Typically wood chips reach a MC of 10-12 % after a few weeks of air drying in the Western Cape in South Africa and this process is faster for smaller particles (Sturos et al. 1983; Wondifraw 2010). Drying in a kiln or oven can remove the water entirely, depending on the time and temperature employed. Torrefaction (see Chap. 7) is an alternative to drying, resulting in products highly suited for thermochemical processing that can be stored for long periods of time.

Any moisture present in biomass reduces its calorific value. For combustion, gasification and pyrolysis the MC should therefore be as low as possible. Acceptable values for most reactor types are between 10 and 20 % (Schuck 2006; McKendry

2002) . If too much moisture is present, much of the energy contained in the biomass is used to evaporate the water. Furthermore, biomass with a high MC cannot generate certain compounds when pyrolysed, which also affects the acidity and composition of liquid produced (Guillen and Ibargoitia 1999).

On the other hand, moisture is desirable for biochemical conversion, such as fermentation and anaerobic digestion (see Chap. 7) and the biomass is kept as wet as possible, typically with MC values between 80 and 90 %.

Tree improvement programmes and advances in site management have increased the growth rates of commercial forestry tree species. This has also lead to earlier canopy closure in plantations. Under increased stand densities and shorter rotation lengths associated with intensive biomass production, canopy closure, with an associated Leaf Area Index (LAI) of 5 is now likely to occur after just 2 years on a favourable site. There are well established links between growth and water-use in trees, so an increase in the gradient of the water-use curve during the first few years of the rotation may be expected under faster growing trees. However, at canopy closure, competition for light, as opposed to water, may become the limiting factor to further increases in leaf area and hence water-use. Consequently, under intensive biomass production, the water-use curve is likely to peak and plateau earlier than before, resulting in an overall increase in water-use relative to tree age (Fig. 10.3).

10.2.1.1 Site and Species Choices

The location within the landscape of commercial forestry plantations; be they exist­ing pulpwood/saw-timber stands or proposed future intensive biomass production stands, undoubtedly has an impact on their water-use. This is most pronounced in the distinction between riparian and non-riparian sites. In a study quantifying the effect of changes in riparian zone vegetation on catchment water yield (streamflow),

Everson et al. (2007) showed significant responses in streamflow following clearing of Acacia mearnsii (Black Wattle) trees from riparian and upland areas in a small catchment in KwaZulu-Natal (Fig. 10.4).

During the 6 year period of the study (May 2000 to May 2006), increases in streamflow associated with the clearing of the A. mearnsii trees, which had initially been planted throughout the catchment including the stream channel, were monitored. Based on the areas cleared and the resultant streamflow changes observed, these results indicate that streamflow gains following clearing operations were 4.8 mm for every hectare of riparian area cleared, and 1.34 mm for every hectare of upslope area cleared (Table 10.1). A unit of land in the riparian area under A. mearnsii consequently represented the hydrological equivalent of 3.58 times the upslope area (4.8/1.34 = 3.58) when cleared.

While legislation currently prohibits the establishment of commercial forestry plantations in riparian areas (FIEC 1995), the above findings illustrate the
importance of focusing on riparian areas when clearing invasive exotics through activities such as the Working for Water programme (Turpie et al. 2008). They also help to quantify the water released by such activities, particularly when harvesting for biomass/bio-energy production. A further site-related hydrologicalconsideration is the utilization of more marginal forestry areas, particularly in terms of rainfall. Intensive biomass production in such areas is likely to have lower absolute water-use impacts (mm), but significantly greater relative water-use impacts (%), compared to optimum sites. Linked to this is the need to make distinction between impacts on total flows and impacts on low flows (e. g. driest 3 months of the year). Scott and Smith (1997) argued that low flows may be of greater relevance to decision makers than reductions in total flows, and several South African studies have focused on this aspect, most recently Jewitt et al. (2009). The significance of low flows is also attributable to the emphasis placed by the National Water Act (NWA 1998) on the human and ecological “reserve”, both of which are critical during periods of low flow.

As far as species selection is concerned, intensive biomass production is likely to favour coppicing Eucalyptus species due to their rapid growth, despite a relatively high ash content after combustion, However, pines will still be considered due to better pellet quality producing less ash, particularly where multiple-use of tree biomass is practiced (e. g. quality saw timber used for conventional sales, with off — cuts and branches used for biomass production). The implications of this in terms of water-use are that Eucalyptus species use more water than pines and wattle in turn. Allocation is made for these differences in the current SFRA water-use licensing system, however, changes from one species to another (e. g. Pine to Eucalyptus) will constitute an increase in water-use (greater streamflow reduction) and hence will be subject to species exchange adjustments to existing water-use licenses.

Given the scenarios discussed in this section, the evidence suggests that higher stand densities and faster growth rates (earlier canopy closure) will increase water — use while shorter rotations will reduce water-use. In general, however, a move from conventional pulpwood and saw-timber plantations to intensive biomass production plantations is likely to result in increases in water-use per unit area. If large-scale changes to this form of land-use are to be approved innovative solutions will be required to offset the increased water resource impacts. Options include accelerated clearing of high water-using invasive exotic trees from riparian areas, and possible replacement with low water-using indigenous tree species of high economic and ecological value (see Gush et al. 2011; Wise et al. 2011).

Sustainability of bioenergy production from wood in the tropics is mainly endan­gered by the ‘gold-rush fever’ phenomenon on the bioenergy market, which sometimes fosters developing projects in tropical countries, without a clear concept of sustainable resource supply. This applies to both traditional and commercial bioenergy production. Tropical countries face the challenge of designing holistic concepts for a sustainable implementation of bioenergy use that is adapted to local conditions. These concepts must embrace all three aspects of sustainability and carefully balance bioenergy production with all other socio-economic and ecologic demands (Fig. 1.3).

It entails that all levels of sustainability have to be met, starting with sustained economic feasibility, long-term beneficial impact on society and the avoidance of negative impacts on the environment. This challenge is best met with an integrated land-use management system, where different land-use forms and eco-system services such as food-production, fibre production, biodiversity conservation, water provision, job creation and biomass production are balanced (Fig. 1.4). A bias of land-use towards a singular objective of biomass production does not meet the sustainability criteria. Unfortunately, decision making support tools to balance land — use portfolios are rare and still have to be developed or adapted to tropical conditions (Furst et al. 2013; Seifert et al. in press).

Frequently, basic knowledge on the implementation of sustainable systems for biomass production and conversion to energy is also not readily available in many tropical and sub-tropical countries. The vast majority of the current literature on biomass production originates from countries in the temperate and boreal zone of the Northern Hemisphere and due to differences in climate zones; it may not be directly applicable to countries in the tropics and sub-tropics.

Fig. 1.3 The three spheres of sustainability

Statistical models are usually evaluated with reference to their accuracy, which is defined as the deviation of an estimated quantity from a true value. Accordingly, accuracy is a compound function consisting of the precision (repeatability of estimates or variation around a true value) and of the bias (directed deviation from a true value). An accepted measure of accuracy is the mean squared error (MSE) as used in Eqs. 3.9, 3.10, 3.11 and 3.12 (Hellman and Fowler 1999).

MSE (x) = var (x) + bias(x)2 (3.9)

Here MSE (x) is the mean squared error of a model (accuracy), calculated according to Eq. 3.10.

and,

xi = Model estimate of biomass

№ = True mean value

n = sample size

and Eqs. 3.11 and 3.12 apply

n

i=1

In addition, to model evaluation the error propagation of the full upscaling process should be addressed to provide error budgets for the biomass estimation. The fact that biomass modelling typically involves a combination of different
sampling and modelling steps complicates error budgeting. Detailed background information on error budgeting in biomass estimation can be found in Cunia (1987, 1990), Wharton and Cunia (1987), Yang and Cunia (1989), and van Laar and Akqa (2007). Cunia (1990, p. 169) points to three general sources of error: “There is first the sampling error: the same sampling procedure applied repeatedly to the same forest population leads generally to selection of different sample units and, thus, to different estimates. And then there is the measurement error when the same sample units (trees or plots) measured by different people lead to different recorded values and, thus, to different estimates. Finally, the third error component is that of the statistical model used in deriving estimates; same inventory data analyzed and interpreted by different statisticians may lead to different estimates”. Cunia (1990) adds the application error as a fourth source of error. This is based on the fact that biomass models are usually parameterised from data of a different population than the population where they are applied for estimation. However, the sampling and measurement error are usually assessed in combination (Cunia 1990). An assumption, which is often made in error budgeting of biomass models is that the models are unbiased, which is strictly speaking not always true but it reduces the focus to the variance as the only source of error (Cunia 1990). The following section will be based on this assumption as well. It is a frequently made mistake to exclude the upscaling from the sample to the tree from the error budget, which can only be done if the trees were fully harvested and their dry weight calculated without harvesting losses and sampling or ratio-modelling steps involved, which is rarely the case in most bioenergy studies. In all other cases error budgets for both upscaling steps (‘sample to tree’ and ‘tree to stand’) have to be determined.

Error propagation equations can be found for example in Ku (1966) or Bevington and Robinson (1992). Also van Laar and Akqa (2007, p. 264) stipulate functions for variance estimation for different forms of linear equations and point out the most frequently used ones in biomass upscaling to be the additive and the multiplicative combination (Eqs. 3.13 and 3.14).

Equation Variance

z = x + y s2 = s2 + + 2 sxy (3.13)

z = x • y sz = y2s| + x2s-2 + 2 • x • y • sw (3.14)

Here sxy is the covariance of x and y, and s| and s2 are the variances.

Equation 3.13 is used in biomass estimation if, for example, crown and stem biomass are added, while Eq. 3.14 is applied in allometric biomass models or for multiplying plot biomass estimates with plot areas (van Laar and Akqa 2007).

Equation 3.14 can also be rewritten as relative error (Chave et al. 2004), as indicated in Eq. 3.15. The last covariance term may be omitted if x and y are independent (van Laar and Akqa 2007, p. 264).

4 = 4 (1 ln(f) у 4 (1 ln(f) у MW/if (315)

z2 x2 1 ln (x) y2 1 ln (y) xy1 ln (y) 1 ln (x) ‘

The term 1 ln(/)/1 ln (x) is the partial derivate of lnf) with respect to ln (x) and is added to increase the accuracy with aid of a Taylor series (Chave et al. 2004). The error sagb2 for a typical allometric model that predicts aboveground tree biomass from diameter at breast height and tree height f(D, H) = aD«HB is calculated as indicated in Eq. 3.16; here noted without the partial derivatives to simplify of notation (vide Chave et al. 2004).

Finally, the errors of the different upscaling steps have to be combined. This results in a series of different terms that are combined according to variances determined by Formulae 3.12 and 3.13. It is essential that all the different upscaling steps involved are taken into account. The combination of errors for this example involves the upscaling from sample to the tree and the upscaling from tree to stand. The latter is basically also the standard procedure for error quantification in volume based forest inventories and is expressed by Eq. 3.16. The major error sources including their combination rules are illustrated in Fig. 3.7.

The final combination of the upscaling steps is attributed to the fact that biomass estimation should be viewed as a two-stage sampling process (Cunia 1990). Further upscaling steps might be added, such as from stands to strata according to the same combination rules under the inclusion of forest area information as can be received by remote sensing (Chap. 2). It should once again be emphasised here that it is critical to include the upscaling from the samples to the tree in this context. Ignoring this upscaling step, as frequently found in literature, is only warranted if the complete drymass of the tree was measured as indicated before. A good example to underpin this statement is the regression of the foliage dry mass from branch diameter as an essential part of the upscaling. The degree of determination of only R2 = 0.68 (Fig. 3.5) shows a considerable error potential and is only one of several error sources in the first upscaling step. Thus ignoring the first upscaling step would lead to a crude underestimation of the error.

Chipping is the most common method of comminuting biomass in preparation for combustion or other form of energy conversion. The two predominant chipping types are disc chippers and drum chippers (Fig. 6.2).

The working principle of the disc chipper is that 2-4 bevelled knives are fixed radially in a fast rotating disc. The knives, which can be adjusted for desired chip size (measured in the fibre direction) cut the biomass perpendicular or slightly offset to the feeding direction, and run up against an anvil to ensure the material is severed. A fan blade mounted on the rear of the disc creates a pneumatic force that blows the chips out of the spout and into a container or onto the ground. In larger chippers (>40 cm intake), the disc can have a diameter of over 120 cm and weigh more than 1,000 kg. Because the disc always cuts at a constant angle to the material, the disc chipper can produce very uniform chips. Disc chippers produce more ‘stickers’ than others (long slivers which cause stoppages in conveyor systems) as these are pulled into a parallel orientation to the knives. Various solutions have been found to reduce that problem which is more pronounced in small material (small trees, tops and branches).

The drum chipper consists of a number of knives mounted along the longitudinal axis of a steel cylinder, with a smaller diameter than the disc chipper. It is therefore more compact and can be built into smaller spaces (e. g., on chipper trucks). By nature of its design, the knives on a drum chipper cut into the material at different angles, depending on the size of the log or branches. This produces slightly more heterogeneous chips. The drum chipper can generally be built for larger diameter logs, or larger bunches of smaller material, as the disc chipper intake has to be limited to less than the radius of the disc. Provided enough power can be delivered to the drum, it is possible to build drum chippers with larger intake capacities. The length of the knives reduces the negative consequences of hitting dirt or a stone as this would represent a smaller proportion of the knife than the same damage on a shorter disc knife. The knives on both disc and drum chippers have to be maintained (sharpened or reversed) at least once a day, and even more frequently when working with material that has been contaminated with soil, sand or stones, or cutting carbonized bark.

Auger or conical screw chippers are robust and produce homogenous chips of good quality. However, they require much higher power drivers, due to the high forces required in severing the material that is fed in the same direction as the axis of rotation. Chip size is adjusted by exchanging the screw for one with a different pitch, while the whole screw needs to be sharpened in place, or exchanged for a newly sharpened one. An advantage of screw chippers is that large material (chunks — up to 150 mm) can be made for e. g., thermal-gasification. Irrespective of the working principle, chippers are deployed in many sizes and configurations. Some are built onto terrain going base machines, others on trucks or trailers for mobility, while others are located centrally at terminals or conversion plants.

The biochemical production of different biofuels occurs through the hydrolysis of polysaccharides contained in lignocellulosic biomass into fermentable sugars and organic acids. Hydrolytic catalysis of biomass is achieved using either acid — catalysed hydrolysis or enzyme-catalysed hydrolysis. While acid hydrolysis is applicable to certain instances of biofuel production, enzymatic hydrolysis is a more economically and technologically advantageous option given the specificity of enzymes, the mild conditions required for the process and the high product yields gained.

The complex structure of lignocellulose calls for a pretreatment step in order to render the substrate more amenable to hydrolysis. Knowledge of the composition and structure of the raw materials is imperative to determine the most suitable pretreatment, to choose which enzymes to use as catalysts and to select the microorganisms most suited for optimal biofuel production. In fact, identifying the most suitable pretreatments and conditions for the fractionation of different lignocellulosic materials into its main structural components is one of the most important goals of research and development (http://www. eng. auburn. edu/cafi/ index. htm). Furthermore, the pretreatment step is currently considered the second largest contributor to the cost of second generation biofuels after feedstock acquisi­tion (Stephen et al. 2010). The effectiveness of the pretreatment will determine the yield in each successive step and therefore also the final product yield.

The heterogeneity of lignocellulose feedstocks has driven the investigation of numerous pretreatments but these can be broadly categorized into physical, chemical and biochemical pretreatments or can be a combination of these different classes of pretreatment (Alvira et al. 2010; Agbor et al. 2011). Among the different pretreatment options, organosolv, dilute acid prehydrolysis, acid-catalysed steam explosion and the novel SPORL technique (Sulphite Pretreatment to Overcome Recalcitrance of Lignocellulose) are the most promising technologies to facilitate the commercialization of biorefining of woody materials (Wang et al. 2009).

However, other pretreatments could be of interest for forestry residues that are more easily digestible. Conventional as well as some novel pretreatments and their mode of action are briefly described next with reference to specific studies on Pinus, Eucalyptus and Acacia.

The conversion techniques described in Chap. 7 have different requirements with regards to the biomass feedstock. The energy yield can be optimised, if the biomass is chosen and prepared correctly, as the pre-treatment of the initial feedstock has a big impact on the product yield and quality. These pre-treatments can be mechanical and/or chemical, such as sieving, water washing, solvent or acid-leaching. Table 8.6 lists some of the requirements for a selection of conversion techniques and reactor types (extracted from van Loo and Koppejan 2008; Stephen et al. 2010):

The heightened awareness of environmental protection, and the possible impacts associated with products manufactured and consumed on it, has led to the devel­opment of methods to better comprehend and reduce those impacts (ISO 14040 1997). Life-cycle assessment (LCA) has been postulated as an important and comprehensive technique. In an LCA study, the whole system involved in the production, use and waste management of a product or a service is described. Intuitively, one can understand LCA as a structured and comprehensive technique to assess the potential environmental impacts and resources used throughout a product’s life-cycle, i. e. from raw material acquisition via production and use phases, to waste management (Baumann and Tillman 2004). LCA studies are done in various contexts (ISO 14040 1997), assisting in

• identifying opportunities for improving the environmental aspects of products at various points in their life-cycle;

• decision-making in industry, governmental or non-governmental organisations (e. g. strategic planning, priority setting, product or process design or redesign);

• the selection of relevant indicators of environmental performance, including measurement techniques; and

• marketing (e. g. environmental claim, eco-labelling or environmental product declaration.

The LCA method has its origins mainly in the packaging and waste management, as well as the oil crisis and the energy debate at the beginning of the 1970s. Pioneers in the field of LCA were industrialised countries such as the USA, the UK, Germany and Sweden. It is generally accepted the first LCA was a study on the consequences of packaging and manufacturing beverage containers by the Midwest Research Institute on behalf of Coca Cola (Baumann and Tillman 2004). Subsequent to that other studies were initiated in Europe in both the private (Tetra Pak) and public (German Federal Ministry of Education and Science) sectors. Although public interest waned due to the ending of the first energy crisis in the late 1970s, private businesses and certain industries (e. g. product design) remained interested in the LCA approach. An increased environmental awareness in the 1980s saw once again the need in focussing on a systematic approach for capturing the environmental

impacts of product or service systems. The 1990s were characterised by the harmonisation and standardisation of the LCA methodology (e. g. ISO standards 14040-14044). Today LCA represents a common environmental assessment tool which is predominantly applied in the primary and secondary production sectors. The importance and relevance of LCA can be identified by a steadily growing number of LCA studies. Various software suppliers, such as SimaPro, GaBi and Umberto, have developed user-friendly LCA interfaces.

An LCA is structured into of four phases, namely (1) goal and scope definition, (2) inventory analysis, where an inventory of the relevant inputs and outputs of a product, process or service system is compiled, (3) impact assessment, where the potential environmental impacts associated with those inputs and outputs are evaluated and (4) interpreting the results of the inventory analysis and impact assessment phases in relation to the objectives of the study (refer to Fig. 11.1). Each phase is elaborated on below.

Another remote sensing option, related to the assessment of biomass, is the combined use of different wavelengths of light to differentiate tree species in stands, i. e., enabling biomass-per-species maps. This is relevant because biomass functions for total biomass and biomass components are usually species-specific (Chap. 3). So knowing the species composition of stands would surely improve the prediction of total biomass. This species identification was only made possible with the devel­opment of spectral resolution increases from multispectral to hyperspectral sensors (see Fig. 2.7). Hyperspectral sensing, also called imaging spectroscopy, samples more regions of the electromagnetic spectrum than multispectral sensors, such that absorption from specific leaf pigments, canopy structure, or leaf water content can be estimated (Curran 1989; Wessman et al. 1989; Yoder and Pettigrew-Cosby 1995;

Kokaly and Clark 1999; Kokaly 2001). The electromagnetic canopy reflectance signature provides enough detailed data to discriminate between signatures of different species. It is therefore possible to apply species specific models to stratify a community into more representative components and calculating biomass by component using for example canopy structure profiles from LiDAR (Chambers et al. 2007). The reader is referred to seminal studies on this topic, including Gong et al. (1997), Martin et al. (1998), Fung et al. (1999), and van Aardt and Wynne (2001), with an extension to operational airborne data (van Aardt and Wynne 2007) and commercial plantations (van Aardt and Norris-Rogers 2008). These approaches also have been borne out in savannah regions by Cho et al. (2010). These studies report classification accuracies >90 % for deciduous species and as high as 85 % for coniferous species, while the commercial species, specifically Eucalyptus sp., have proven to be spectral separable with accuracies at approximately 90 %. All of these studies have approached the challenge by subsetting the hyperspectral data to those wavelengths necessary for separation of a specific set of species. In other words, an operational workflow should be preceded by a pilot study: (i) identify the species of interest, (ii) acquire hyperspectral data, (iii) determine which wavelengths are necessary to separate the species on a spectral basis, and (iv) assess the classification accuracy. And herein lies the caveat — hyperspectral data are by design “oversampled” data, i. e., we have more data than we need.

We therefore need to subset the data to the wavelengths required for a specific application in order to develop statistically robust models, or models with a reasonable number of independent or explanatory variables. An example may be warranted: Imagine that a pilot study shows that wavelengths at 452 nm (blue), 622 nm (red), 1,050 nm (near-infrared), 1,452 nm (shortwave-infrared), and 2,248 nm (shortwave-infrared) are essential to separating three coniferous species at an accuracy of 85 %. The logical approach, for an operational implementation, would be to approach a company that specializes in acquiring imagery via air­borne detector/s that can be “programmed” to these wavelengths, via the use of wavelength-specific filters. Such companies and sensors do exist, but they typically only operate silicon-based sensors, which are sensitive to the wavelength range of roughly 380-900 nm; the conundrum is obvious: in order to acquire the necessary wavelengths for the species classification application, one would have to build a relatively expensive sensor, unless the subset of wavelengths can be constrained to the silicon range. This latter solution often is viable, but may come at the cost of slight decreases in accuracy. However, this approach is actually not infeasible, given the lower cost and higher prevalence of silicon sensors; many forestry applications, e. g., species classification, nutrient mapping, moisture stress detection, etc., could be constrained to this wavelength range and thus executed on an operational basis. In fact, leaf-level studies have shown extension to airborne cases, thus adding to potential operational implementation.

Results from a study in tropical forests of Costa Rica indicate that there are spectral differences among species that permit classification at leaf to crown scales (Clark et al. 2005), further corroborated by van Aardt and Wynne (2007) in a mixed oak-pine forest in the Virginia piedmont, USA. However there are also temporal, spatial, and spectral variation within populations and even single individuals of forest tree species that will inevitably decrease classification accuracy and need to be assessed on a as-needed basis. A major challenge is to develop classification schemes that can maximize the spectral, spatial, and temporal information content of digital imagery while accommodating inherent variation within species.