The potential use of papyrus (Cyperus papyrus L.) wetlands as a source of biomass energy for sub‐Saharan Africa

Four of five people in sub‐Saharan Africa rely on the traditional use of solid biomass, mainly fuelwood, for cooking. In some areas, the current rate of fuelwood consumption will exhaust biomass reserves within the next decade or two. A largely unrecognized source of biomass are tropical wetland ecosystems which have been shown to be some of the most productive ecosystems globally, exhibiting rates of net primary productivity comparable with high‐input, intensively managed agricultural systems. Papyrus (Cyperus papyrus L.) is an emergent sedge with C4 photosynthesis which is native to the wetlands, river valleys and lakes of central, eastern and southern Africa. The mean standing dry matter of culms and umbels measured at a number of locations throughout East Africa is 38.3 ± 21.6 tDM ha−1, and the aerial net primary productivity ranges between 25.9 and 136.4 tDM ha−1 yr−1. Papyrus vegetation can be harvested by hand and stacked on the rhizome mat for partial air‐drying, and it has been demonstrated that an annual harvesting regime has no negative impacts on long‐term productivity. The use of papyrus as a biofuel for cooking and heating depends on converting it to a suitably combustible form, such as compressed or carbonized briquettes with a calorific value approximately one‐third less than wood charcoal. While papyrus has significant potential as a biofuel, we argue that an integrated management and decision‐making framework for the sustainable utilization of papyrus wetlands is required, in which all ecosystem services including the provision of biomass energy need to be assessed. Sustainability of papyrus wetlands requires management which combines the strength of traditional communal governance and modern legislation to promote its utilization. In this way, local communities can benefit from the inherent advantages of tropical wetlands as very productive ecosystems.


Introduction
The primary source of energy for over 80% of the households of sub-Saharan tropical Africa is biomass (International Energy Agency, 2010), predominantly in the form of charcoal or wood (Cerutti et al., 2015), which is used mainly for cooking and heating (Murphy, 2001). Fuelwood remains the main source of domestic energy for the rural poor but charcoal is the major source of energy for the urban poor and even though the share of the energy source provided by biomass is declining, the numbers dependent on it continue to grow. An estimated 40% rise in the demand for bioenergy by 2040 will increase the pressure on the forest biomass stocks, with efforts to promote more sustainable wood production hindered by the operation of much of the fuelwood and charcoal supply chain outside the formal economy (International Energy Agency, 2014). The supply of fuelwood to produce charcoal or for burning directly is obtained from a wide range of sources ranging from scrubland and savannah to forest ecosystems. In many areas, the supply of this wood is not sustainable and its continued use has led to the depletion of natural ecosystems, contributing to the net global increase in atmospheric CO 2 concentrations and the loss of biodiversity (Wessels et al., 2013;Cerutti et al., 2015). This situation has long been recognized as the 'other energy crisis' (Eckholm, 1975), and it has recently been shown in some areas that at current levels of biomass use, reserves of fuelwood could be exhausted within the next decade or two (Wessels et al., 2013). It is therefore essential to identify new forms of biofuel that can be produced and supplied sustainably and would reduce the current dependence on unsustainable forms of energy. An obvious source is productive, nonwoody natural vegetation which could be harvested on an annual basis which then regenerates, but without the need for expensive inputs (Somerville et al., 2010). In North America, for example, Tilman et al. (2006) have highlighted the potential of low-input high-diversity grasslands as a sustainable biofuel option as they represent a carbon neutral or carbonnegative solution, where carbon dioxide assimilation and sequestration during the growth cycle exceeds release during combustion, and can be targeted at marginal agricultural land, reducing competition for land for food production (Valentine et al., 2012) and biodiversity loss through habitat degradation. Samson et al. (2005) have reviewed the potential of C 4 perennial grasses to supply a bioenergy industry based on densified plant material for stoves and boilers which has a 14 : 1 energy output : input ratio. This is dependent on the use of high-yielding grasses with minimal fossil fuel use during production and energy conversion. We demonstrate here that a largely unrecognized yet significant source of biomass supply are tropical wetland ecosystems which have been shown to be some of the most productive ecosystems globally, exhibiting rates of net primary productivity comparable with high-input, intensively managed agricultural systems (Piedade et al., 1991;Morison et al., 2000;Saunders et al., 2007).
Wetlands cover approximately 7% of Africa (Junk et al., 2013), and many of the permanently inundated wetlands are dominated by the emergent sedge papyrus (Cyperus papyrus L.) (Fig. 1a). These wetlands directly support the livelihoods of millions of rural people in sub-Saharan Africa through the provision of key ecosystem functions and services, such as the supply of water, food and building materials, a buffer function to regulate river discharge, carbon storage and maintenance of biodiversity. They also serve as the basis for a rich cultural tradition (Maclean et al., 2011;Gaudet, 2014). Moreover, they perform important ecosystem services on a larger spatial scale through biogeochemical cycling and their impact on the local and regional hydrological cycle and climate (Junk et al., 2013). However, they tend to be largely unused as a source of biofuel. The exact geographical extent of papyrus wetlands is not known, but in the East African region alone, papyrus wetlands are estimated to cover approximately 40 000 km 2 (Hughes & Hughes, 1992), although the area is almost certainly decreasing as a result of agricultural encroachment and economic development (Kansiime & Nalubega, 1999;Saunders et al., 2012). Currently, papyrus wetlands are under significant human pressure as large areas close to urban areas have been cleared for the cultivation of crops such as Colocasia esculenta (cocoyam), which has had a significant, negative impact on the ecology and ecosystem service provision of these wetlands (Saunders et al., 2012). As a result of this encroachment, it has been estimated that in some areas, such as the shores of Lake Victoria, the papyrus swamps could disappear by 2020 (Owino & Ryan, 2007).
As there is a significant risk that the valuable ecosystem services provided by papyrus wetlands will be degraded or even lost, it is important to draw attention to the importance of papyrus wetlands, to review the existing knowledge base and to present the most recent research findings. To enhance the use of such information in land management decisions and policy-making, this must be undertaken in an interdisciplinary context where the results from the natural sciences are linked to social science research (Peh et al., 2013). In this platform article, we suggest that in the search for new sources of biomass for biofuel production, wetlands, such as those dominated by papyrus, have been largely overlooked in recent times, possibly because of the highly variable estimates of documented productivity and regeneration potential (Osumba et al., 2010) but also because of concerns about the unsustainable utilization of these ecosystems that provide other valuable ecosystem services (Terer et al., 2012). The key question addressed here is whether papyrus vegetation can be used as a source of biofuel and can this be achieved without incurring negative impacts on the ecosystem. In other words, is it sustainable? We argue that while papyrus plant material has been shown to have significant potential as both a domestic fuel in the form of biomass briquettes (Jones, 1983;Morrison et al., 2013) and as a biofuel substrate (Charuchinda et al., 2011), sustainable harvesting regimes are required to maintain the biofuel supply chain while at the same time continuing to support the provision of multiple ecosystem services from these wetlands (Morrison et al., 2014).

Ecology, distribution and use of papyrus vegetation
Papyrus (Cyperus papyrus L.) is an emergent C 4 photosynthetic sedge which is native to the wetlands, river valleys and lakes of central, eastern and southern Africa and has been successfully introduced to Italy, United States and India (Haines & Lye, 1983;Hughes & Hughes, 1992;Terer et al., 2012). Papyrus tends to form floating monotypic stands consisting of plant culms, the main aboveground vegetative structure, which are topped by an umbel consisting numerous cylindrical rays and flattened, leaf-like, bracteoles which represent the main photosynthetic surface of the plant as true leaves are absent in mature plants (Fig. 1b). Culms emerge sympodially on a continual basis from the leading growth edge of the plant rhizome and extend up through the canopy reaching heights of up to 5 m where the umbel fully expands once the culm is completely extended. The height of the papyrus canopy ranges from 2.48 to 5.0 m at sites from Botswana to Sudan, and the number of culms per square metre ranges from 9.2 to 133 (Saunders et al., 2013). The life cycle of a culm and umbel ranges between 5 and 12 months depending on site (Jones, 1982;Osumba et al., 2010) after which the structure senesces and dies, resulting in the recycling of nutrients to the rhizomes and the formation of significant organic detrital deposits on the rhizome surface. Historically, papyrus was cultivated and used by the ancient Egyptians, Greeks and Romans in the production of papyri, a kind of paper made from strips of the culm pith laid and pressed together (Laws, 2011;Gaudet, 2014). However, papyrus is currently used only on a largely subsistence basis for making fences, roofing and matting (Kansiime & Nalubega, 1999;Terer et al., 2012).

Yield potential of papyrus
The optimization of biofuel production is dependent on utilizing the high primary productivity of vegetation while minimizing the energy inputs required during the production life cycle and maintaining the environmental and social acceptability of bioenergy products (Tilman et al., 2006;Elghali et al., 2007;Zhang et al., 2010). The highest rates of dry-matter production and the greatest efficiencies of nitrogen and water use tend to be found in species which possess the C 4 photosynthetic pathway . Some of the highest records of primary productivity in either managed or natural ecosystems have been recorded in tropical wetlands where the predominant species utilize the C 4 photosynthetic pathway (Jones, 1988). For example, in South America, the C 4 grass Echinochloa polystachya has an annual net primary production (NPP) of up to 3.97 kg C m À2 yr À1 (Piedade et al., 1991;Morison et al., 2000), and in Africa, the aerial NPP of C. papyrus ranges between 2.51 kg C m À2 yr À1 (Muthuri et al., 1989) and 3.09 kg C m À2 yr À1 . Carbon dioxide and water vapour flux measurements made over stands of both E. polystachya and papyrus have confirmed their very high net ecosystem production and have shown that these systems are potentially very large sinks for carbon. There is, however, a major difference between the growing cycle of E. polystachya and papyrus in that the former shows a very marked annual cycle of growth linked to seasonal fluctuations in water levels where the vegetation decomposes very rapidly and carbon is lost as the water levels fall. However, in the river valleys and lake fringing papyrus wetlands of East Africa, the hydrological status of the wetlands is less variable on a seasonal basis, and as a result, the canopy is maintained continuously and individual culms are present in multiple age classes and emerge continually from the plant rhizome .
The yield potential of papyrus wetlands is high because it has C 4 photosynthesis and it forms a continuously closed canopy which grows throughout the year (Jones, 2011). The standing dry matter of culms and umbels has been measured at a number of locations throughout East Africa, where estimates of harvestable biomass ranged between 11.6 and 86.9 t DM ha À1 across sites in the Democratic Republic of the Congo (DRC), Kenya, Rwanda, Tanzania and Uganda (Table 1). The values of standing biomass are relatively stable throughout the year as demonstrated by  and Osumba et al. (2010) who found no significant seasonal variation in monthly estimates of standing aerial biomass. In undisturbed papyrus wetlands, estimates of aboveground NPP show significant variability, ranging from 25.9 to 136.4 t DM ha À1 yr À1 across sites in the DRC, Kenya and Uganda (Table 2). At these rates of productivity, the canopy turnover time ranges from 705 days at its longest to 158 days at its shortest. The high rates of aboveground productivity in papyrus wetlands highlights the potential of this plant as a bioenergy source, as they represent some of the highest recorded rates of primary productivity in any natural ecosystem. In comparison, Piedade et al. (1991) found that E. polystachya had an NPP of 99 t DM ha À1 yr À1 and the highest annual dry-matter yield for the forage crop Pennisetum purpureum (elephant grass or Napier grass) was 85 t DM ha À1 yr À1 . In temperate climates, the average dry-matter yields from a large number of trials in N. America and Europe of the two perennial rhizomatous grasses with greatest potential for use as bioenergy crops are 22 t ha À1 yr À1 for Miscanthus 9 giganteus and 10 t ha À1 yr À1 for Panicum virgatum (Heaton et al., 2004). The effect of harvesting frequency on the sustainable yield of papyrus has been investigated in a small number of trials. Osumba et al. (2010) studied the effects of repeatedly harvesting papyrus over a six-month period on Lake Victoria and found that a monthly harvesting regime reduced biomass to nil after three consecutive months of harvesting. This dramatic effect of frequent harvests on growth was anticipated by  who pointed out that the removal of large quantities of nutrients as a result of papyrus harvesting may lead to reduced rates of production in subsequent regrowth periods. Terer et al. (2012) investigated the effects of repeatedly harvesting papyrus at 6 and 12 monthly intervals compared with unharvested controls over a period of 3 years in an undisturbed site at Lake Naivasha, Kenya. They showed that the six monthly harvesting regime for papyrus significantly reduced  aerial biomass production, culm density, canopy height and young shoot regeneration compared to the 12 monthly harvesting regime. These results illustrated the importance of allowing enough time between harvests for a full cycle of growth from young stems to senescence. However, there have not been any studies which have continued this process for several years, so that long-term sustainability is not assured at present.

Utilization of papyrus as a biofuel
Papyrus vegetation can be harvested by hand and stacked on the rhizome mat for partial air-drying before it is utilized (Jones, 1983(Jones, , 1984Morrison et al., 2013). Currently marketable goods derived from harvested papyrus include furniture, mats, baskets but no market for biofuel has been developed. The use of papyrus as a biofuel for cooking and heating depends on converting it to a suitably combustible form. Papyrus culms are not dense enough to be burned directly in stoves so they need to be densified or converted into charcoal. Jones (1983) described trials to compress the papyrus into briquettes which have a moisture content of 9-15%, a density of 1.2 g cm À3 , ash content of 5% and energy content of 20.5 kJ g À1 . The briquettes were manufactured in a single piston press as a long cylinder with a diameter of 6 cm. A pilot factory was established near Kigali, Rwanda, in the early 1980s but it did not proceed to commercial operation. More recent innovations in the small-scale production of charcoal briquettes for papyrus, developed by the 'Fuel from the Fields' project at Massachusetts Institute of Technology have been described by Morrison et al. (2013). Bundles of dried culms were carbonized using a methodology developed by MIT's D-Lab (2012), which utilizes a converted 200-l oil drum as a carbonizing kiln. A large hole in the top of the drum and a number of smaller holes in the base allow for airflow during the initial combustion of the biomass. The majority of the volatile organic compounds in the papyrus burn off during the initial stages of the combustion. Once the papyrus has reached carbonization temperature (~450°C), the kiln was sealed to create an anaerobic environment to produce charcoal. The cooled carbonized papyrus was crushed by hand and mixed with a 5-10% by weight binder of cassava (Manihot esculenta) flour in heated water to a porridge-like consistency and made into cuboid briquettes using a simple press made locally from scrap metal (Morrison et al., 2013). These briquettes have a calorific value of 20.25 kJ g À1 , which is around one-third less than locally available wood charcoal at 32.43 kJ g À1 (Morrison et al., 2013). One distinct advantage of the combustion of carbonized over the noncarbonized briquettes was lower emissions of concern to human health (Morrison et al., 2013). Combustion of biomass in the form of wood and charcoal is estimated to lead to the annual premature death of 400 000 individuals in sub-Saharan Africa, and it is suggested that a move from firewood to 100% charcoal would reduce air pollution by 90% (Bailis et al., 2005). The mean volatile matter of the papyrus briquettes (28.85%) was considerably less than that of firewood (79.87%) but more than wood charcoal at 10.96%. At a more industrialized scale, it has become increasingly clear in attempts to derive maximum benefits from the utilization of biomass that there are potentially a wide range of opportunities to utilize new methods to generate bioenergy while also producing new products of high value through the process of biorefining (Parajuli et al., 2015). Biorefining is fundamentally the sustainable processing of biomass to produce a spectrum of marketable products as well as energy. A form of biorefinery referred to as 'green' biorefining is probably the most suitable for the type of small-scale processing of biomass that is appropriate for sub-Saharan Africa (Bruins & Sanders, 2012). In this process, the green biomass is separated into fibre-rich press cake for combustion while the green juice can be processed to produce highquality protein for animal fodder or human consumption. As yet, papyrus has not been trialled as a feedstock for a biorefinery.
Economic assessment, environmental risk assessments and ecosystem service provision of papyrus wetlands There have been a small number of investigations into how the natural capital of papyrus wetlands can be harnessed for the benefits of local communities (Morrison et al., 2012;Johnston et al., 2013). Morrison et al. (2013) identified 27 subsets of human welfare benefits derived from papyrus wetlands associated with Lake Naivasha, Kenya, and Lake Victoria, which they organized into four groups, corresponding to the major categories of ecosystem services (provisioning, regulating, cultural and supporting) as defined by the Millennium Ecosystem Assessment (2005), with between six and eight subsets in each category. The most significant of the ecosystem services provided by papyrus wetlands are as a carbon sink (Mitsch et al., 2010;Saunders et al., 2012), in wastewater treatment , in flood control Ryken et al., 2015) and as a biodiversity reserve (Owino & Ryan, 2007). Principles of sustainable utilization of papyrus wetlands should mean that while benefits can be derived by utilizing papyrus for bioenergy, there would be maintenance of the large number of other human welfare benefits. At the same time, the external benefits of conservation of forest ecosystems, which currently provide the primary source of fuel wood in sub-Saharan Africa, need to be recognized.
Management and decision-making for wetlands needs an integrated approach in which all ecosystem services are identified, their importance assessed and objectives for their desired outputs are formulated (Rebelo et al., 2010;Morrison et al., 2012). Assessment of ecosystem services that inform local decision-making needs to be performed at a site scale which often involves technically demanding and expensive fieldwork (Fisher et al., 2011). Because the resources to achieve this are unlikely to be universally available, Peh et al. (2013) have recently described a Toolkit for Ecosystem Service Sitebased Assessment that guides local, nonspecialists through a selection of relatively accessible methods for identifying which ecosystem services may be important at a site and evaluating the benefits from them. However, this may be difficult to achieve in practice because of divergent perceptions among stakeholders regarding the priority issues in the management of papyrus wetlands that are subject to agricultural conversion. For example, Namaalwa et al. (2013) found that in an area in Eastern Uganda, resource users worry about water and land use conflicts, while local and national governments are more concerned about agricultural encroachment and biodiversity loss.
The sustainability of biofuels should be assessed based upon the net energy they contribute to society and the amount of greenhouse gases they emit compared to their energy equivalent in fossil fuel use. More importantly in the case of utilizing papyrus wetlands, the judgement should be based on the efficient use of resources such as land and water as these are competed for in the production of food, fibre and wood for ecosystem services. It is widely accepted that the metrics of this sustainability should be assessed by analytical tools developed to model impacts and assist with decision-making such as life cycle assessment (LCA). A key concern in constructing LCAs for biomass production in Africa is the realistic delineation of boundaries around the various inputs, outputs and processes that make up the theoretical life cycle of biofuel production and use. This is compounded by the fact that almost all LCAs so far have been undertaken in a European or North American context and not in lesser developed countries (Smith, 2010).

Conclusions
There is still a dearth of information concerning our understanding of the physiological, environmental, socio-economic and cultural functions of papyrus wetlands, which makes the development of strategic, interdisciplinary management and conservation measures difficult in the face of increasing human pressure for food, fuel and water. The potential utility of papyrus as an energy source will, however, depend on assurances that these ecosystems can produce a reliable supply of biomass which does not lead to the ultimate demise of these wetlands and that exploitation of this type does not compromise the other important ecosystem services provided by these wetlands. We therefore place strong emphasis on assessing the impact of repeated harvesting of biomass from the papyrus swamps on the range of ecosystem services provided by these wetlands and demonstrating how the use of papyrus biomass can provide a replacement for the current sources of woodbased biomass used by both the rural and urban populations of Central and East Africa.
Although we are advocating the sustainable utilization of papyrus vegetation, it is ironic that there is no consensus on what the term 'sustainable' actually means, particularly in lesser developed countries (Bosch et al., 2015). Currently, biomass assessments are a patchwork of voluntary standards and regulations and Bosch et al. (2015) argue that this leads to mistrust and protectionism, slow investment and slower growth. Consequently, a metric for evaluating biomass sustainability needs to be designed which includes social as well as environmental and economic factors. Management and decision-making for wetlands needs an integrated approach in which all ecosystem services are identified, their importance assessed and objectives for their desired outputs are formulated. This requires a decision support framework (DSF) to arrive at an optimum solution (Zsuffa et al., 2014). Components of the DSF should include calculation of biomass production and optimum harvesting strategies, estimates of nutrient removal and quantification of downstream water quality. There should also be a critical analysis of the social and economic factors that underpin the use of wetland resources (Rebelo et al., 2010). Life cycle assessments do not look at social impacts, so Bosch et al. (2015) have suggested a starting point could be the total factor productivity metric used to measure agricultural productivity.
Sustainability of papyrus wetlands requires management which combines the strength of traditional communal governance and modern legislation to promote its utilization . In this way, local communities can benefit from the inherent advantages of tropical wetlands as very productive ecosystems. This will become increasingly important where growing populations and improving standards of living demand increased use of available resources.