This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
 Author's personal copy

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2

Available online at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

Exploring the potential of reed as a bioenergy

crop in the Netherlands

Tom Kuhlman a,*, Vasco Diogo b, Eric Koomen b

a
LEI, P.O. Box 29703, 2502LS The Hague, The Netherlands
b
VU University, De Boelelaan 1105, 1081HV Amsterdam, The Netherlands

article info abstract

Article history: Second-generation biofuels that produce biomass for combustion or ethanol production do
Received 23 January 2012 not yet appear to be a viable alternative to agriculture as they are low-value products. This
Received in revised form may change, however, when energy prices increase and their production is combined with
15 June 2012 the provision of other services. The current analysis explores the potential for the
Accepted 18 June 2012 production of an often overlooked biomass feedstock that can be combined with water and
Available online 15 July 2012 nature management objectives: reed. This crop has the additional advantage that it can be
grown under conditions that are unfavourable to most other crops. An economics-based
Keywords: land-use modelling approach is applied to simulate the local competition between reed
Reed and grassland used for dairy farming under four different future scenarios in the
Land use simulation Netherlands. Based on a location-specific assessment of potential costs and benefits of
Cost-benefit analysis these crops under scenario-based conditions this analysis shows that the cultivation of
Landscape services reed for bioenergy, in combination with providing additional land-use functions, while not
Water management viable option under current economic and political conditions, may become competitive
The Netherlands within the next twenty years if any of the following developments occur: energy prices
increase substantially; water tables rise in the low-lying western parts of the country due
to climate change; a policy is implemented that increases bioenergy prices; or a policy is
implemented that stimulates water buffering and the preservation of peat soils.
ª 2012 Elsevier Ltd. All rights reserved.

1. Introduction: demand and supply of
bioenergy
Demand for bioenergy in Europe will expand considerably
in the years to come. The EU Directive on Renewable Energy
of 2009 requires that 20% of all energy consumed in the EU
and 10% of all transport fuel should be renewable by 2020
[1]. Although the production and use of renewable energy
are rising rapidly, there is a long way to go: in 2008 the
share was 10.3% [2]. The Directive does not prescribe
how much of this renewable energy should be bio-based,
but the 10% target for transport has to mean either
electric vehicles (to the extent that the electricity
comes from renewable sources) or biofuels; in 2008
renewable fuel made up only 2.5% of fuels used in road
transport. Furthermore, 17% of renewable electricity is at
present based on biomass; and for heating that contribu-
tion is 94% e firewood being the principal source of
domestic heating other than fossil fuels or electricity
(calculations based on Eurostat data).
The most important bioenergy feedstocks are [3,4]:

Starch crops such as maize or sugarbeet, serving as a basis
for making ethanol.

* Corresponding author. Tel.: þ31 70 3358232; fax: þ31 703615624.

E-mail addresses: tom.kuhlman@wur.nl (T. Kuhlman), vascodiogo3@gmail.com (V. Diogo), e.koomen@vu.nl (E. Koomen).
0961-9534/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biombioe.2012.06.024
 Author's personal copy
42 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2

Oilseed crops such as rapeseed or sunflower, for making
biodiesel.

Residues: from agriculture (crop residues, manure), from
forestry (sawdust, pellets, black liquor) and from urban
areas (municipal solid waste, biogas from landfills).

Biomass: while the first two are agricultural crops in which
only the seed or the tuber is used for energy generation, here
it is the entire plant. Energy is obtained either through direct
combustion or as a feedstock for second-generation bio-
fuels. Biomass crops are usually either grasses e which can
produce prodigious amounts of biomass e or fast-growing
trees such as willow or poplar.
Various concerns have made conventional biofuels
(starch-based ethanol and oilseed-based biodiesel) seem less
desirable in recent years. Firstly, the feedstocks are also
sources of food, leading to higher food prices and, perhaps,
lower food security in poor countries. While this concern may
be somewhat overblown [5] there is undoubtedly some effect,
and the reluctance to turn food resources into fuel is under-
standable. Secondly, expanded production of feedstocks such
as sugarcane in Brazil, soybeans in Brazil and Argentina, and
palm oil in Malaysia and Indonesia leads to accelerated
deforestation e some of which at least is at the expense of
tropical rainforest [6]. Not only will this lead to a decline in
biodiversity, but also the greenhouse gas emissions resulting
from such land-use change may well exceed any possible gain
from replacing fossil fuels by renewable energy [7]. Thirdly,
the production of these biofuels itself requires large amounts
of energy: fuel for agricultural machines and transport of
crops, manufacturing of nitrogen fertilizer, and processing
into ethanol or diesel [8,9]. Finally, the application of fertilizer
leads to emissions of nitrous oxide (N2O), a highly potent
greenhouse gas. This effect plus the aforementioned one
together nullify the reduction of greenhouse gas emissions of
most types of biofuels, except for Brazilian sugarcane [10].
This has led to increased interest in the other two types
of feedstocks mentioned above: waste flows and biomass.
The Netherlands has excellent potential for the utilization
of waste flows: being a densely populated country with
a high-density livestock sector, it has lots of them. The
most important energy bearers in this sector are the
combustion of municipal solid waste for electricity gener-
ation, and biogas from animal manure, sewage or organic
household waste. All of these are used in the Netherlands,
although the production of biogas is far below potential: in
2009, the country produced 7.7 PJ energy from biogas, of
which only 38% came from agriculture. In this paper,
however, we focus on the potential for biomass. The
Netherlands does produce energy from biomass at present:
a total of 19 PJ net in 2009, all of it used for combustion. The
country also produces liquid biofuels (12 PJ in 2009), mostly
from imported feedstocks.
This paper explores the potential for the production of
biomass feedstocks, notably reed, in the Netherlands. Such
potential would at first sight be small: Dutch agriculture
specializes in crops and livestock products with a high value
per hectare, whereas biomass is a low-value product suitable
for marginal lands. However, some agricultural areas could
become marginal in the future. Moreover, the production of
reed could be part of a multifunctional land-use system, and
such a system might become an attractive alternative to
present land use under certain conditions. These conditions
include climate change, energy prices, changes in markets for
agricultural products, as well as policies promoting the
conservation of landscapes and of the environment. They are
explored here in the form of four scenarios representing
different combinations of such conditions. The projections are
made for the year 2030.
To a great extent, the potential for bioenergy production
depends on the availability of land for energy crop
production and competition with other uses, such as food
production and biodiversity conservation. Global and
regional assessments of bioenergy potential have been
performed through non-spatial analysis [11,12]. In these
assessments, the demand for food and feed was fulfilled
and biofuel crops were allocated in the remaining agricul-
tural surplus land. These assumptions were meant to
ensure that both food competition and deforestation within
the country were avoided.
Despite providing valuable information and insights, these
studies used statistical averages for physical parameters such
as crop yields and water availability. However, those can vary
considerably and depend strongly on the spatial location.
Therefore, a spatially explicit method was preferred as that
would allow calculating the potential more accurately by
taking into account the landscape heterogeneity, as well as
identifying explicitly the locations with a higher potential
yield.
This study aims to combine the economic rationale
common to plant siting models with the attention for
spatial detail more generally found in land allocation
models. We, therefore, selected the economics-based Land
Use Scanner [13] model to simulate the local competition
between reed and grassland used for dairy farming under
different future scenarios. This model allows these crops to
compete for land directly according to their economic
performance, which is determined by local biophysical
properties (e.g. soil type and water table) and more general
economic conditions (e.g. production costs and crop
prices).
In the following two sections we first describe the
specific characteristics of reed as part of a multifunctional
land-use system and then go on to discuss costs and
benefits of reed cultivation for bioenergy, both in
commercial terms and from a social point of view. This
analysis indicates potential favourable conditions for the
cultivation of reed. In a subsequent section we present
a spatial exploration that highlights where reed cultivation
is likely to take place when favourable commercial and
social conditions apply. This simulated spatial distribution
of reed areas is helpful when their influence on the land-
scape is to be assessed. Such assessment would most likely
consider the degree of dominance of reed in the landscape,
i.e. the proportion of land dedicated to it and its distribu-
tion. A land-use model is applied to simulate this spatial
distribution for each of the four scenarios. The paper
concludes that reed, while not currently a viable land-use
option, may become so under several scenarios, in partic-
ular as an alternative to dairying on peat soils.
 Author's personal copy
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
2. Methodology
As indicated above, the methodology for this study consists of
two parts, namely the estimation of costs and benefits and the
spatial exploration. For the costs and benefits we have relied
heavily on a previous study which examined the economy of
reed cultivation in fenland (peat) areas [14]. Although this
study looks at the problem from the point of view of the
individual land user, it assumes that the land user will receive
compensation for generating benefits such as water purifica-
tion and carbon sequestration. This makes it, in effect, a social
cost-benefit analysis. These figures are then supplemented by
additional information from other studies. Also, their anal-
ysis, which concerns the current feasibility of reed cultivation,
is extended to scenarios for the future.
In this study, both biofuel potential and the spatial distri-
bution of biofuel crops will be determined with a spatially
explicit land-use model that is capable of simulating the local
competition between different crops based on their relative
profitability. Land-use models can integrate both socio-
economic and biophysical driving forces and allow the
exploration of potential future developments in the land-use
system, as well as alternative (what-if type of) scenarios
resulting from policy decisions or developments on
society [15].
Land-use models have been applied in recent studies to
explore the potential of bioenergy production at various
scales: global [16], continental (e.g., [17,18]), national [19] and
regional (e.g. [20e22]). By taking into account land-use
dynamics and spatial variation in location-specific parame-
ters such as crop yield, water availability and existing infra-
structures, such models allow calculating this potential more
accurately and identifying locations where production is likely
to take place.
In general, land-use model applications focused on biofuel
production have two main objectives: biofuel production
plant siting (e.g. [21,22]) or land allocation for biofuel feedstock
crops (e.g. [19,20]), while some studies try both (e.g. [17]).
Production plant siting models usually apply techno-
economic optimization algorithms, taking into account
parameters such biomass yield, feedstock supply and price,
transportation infrastructures and costs conversion tech-
nology and efficiency and (regional) biofuel demand. Although
these models provide insight into the economic potential of
using biomass resources, they normally put the emphasis on
the economic drivers, while dealing with biophysical drivers
for crop production at fairly coarse resolutions. As such, they
somewhat neglect the importance of spatial variation of
biophysical parameters on the biofuel production potential.
Land allocation models spend a considerable effort on
determining the biophysical suitability for crop production at
high resolution. Allocation procedures can vary [19]: explored
the potential for sustainable biofuel production in
Mozambique by first ensuring nature conservation and land
allocation for food, feed and other materials production
according to the expected demand and location-specific yields
[20]; assessed the competitiveness of bioenergy crops in
a Northern region of the Netherlands by comparing the net
present value of biofuel production with current agricultural
43
land uses [17]; explored the spatial distribution of biofuel
crops and refineries in Europe, but since the model did not
have a separate class for biofuel crops, this was performed
through a multi-criteria analysis. Despite of mimicking spatial
allocation processes reasonably well, this type of ad hoc
approaches are less successful at explaining the human
behaviour that causes these processes, even though they
include explanatory variables capturing economic effects [23].
3. Reed as part of a multifunctional land
use setting
Several grasses and trees suitable as feedstocks for bioenergy
can be grown in the Netherlands. Among the trees, willow is
a highly suitable crop for wetlands [24]. As for grasses, Mis-
canthus has been studied for this purpose [20]. Another grass
which is indigenous to the Netherlands (as well as in many
parts of the world) is the common reed (Phragmites australis). It
is not normally cultivated, but for centuries it has been har-
vested in the wild as a raw material for handicrafts; today it is
still used for thatching. It is a prolific producer of biomass,
both above and below ground; and since it is an inhabitant of
wetlands, it holds promise for the alternative use of fens.
Moreover, it is capable of tolerating a fairly high degree of
salinity.
There are three ways of converting reed into energy:

direct combustion, in order to generate electricity or in
a district heating system (possibly with electricity as a by-
product); this appears to be the most profitable conversion
path at present, requiring a fairly large-scale of operation;

production of cellulosic ethanol, by first converting the
cellulose into sugars, then fermenting and distilling as with
first-generation ethanol, also requiring large-scale facilities.
This is not yet economic at present, but may become
a suitable method in future. It also requires a large-scale
form of production;

co-digestion with animal excreta (or sewage sludge) into
biogas. One advantage of this method is that unlike the
other two options it can be done on a small scale, operated
at the level of a large farm or a group of smaller farms.
As a form of land use, reed can have several other uses in
addition to being a bioenergy feedstock. Since these uses are
simultaneous, we can speak of multifunctional land use. The
first of these is inextricably tied to the production of bioenergy
itself: the impact on greenhouse gas emissions. Because the
carbon dioxide released to the atmosphere during the
combustion process has first been absorbed from the atmo-
sphere during the growth process, the net increase of carbon
dioxide from burning fossil fuels is avoided; from this effect
has to be deducted the fuel used during cultivation, harvesting
and transport. Since no fertilizer is added, this deduction is
much lower than with conventional biofuels; also, there is no
additional emission of nitrous oxide, which, as we saw above,
greatly diminishes any positive effect that those biofuels may
have. Furthermore, biomass produces more energy per
hectare, so its beneficial effect on greenhouse gases is that
much larger.
 Author's personal copy
44 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
There is an additional effect if reed is grown on peat
soils. Peat is a large store of carbon dioxide, and its oxida-
tion under current land use in the Netherlands leads to
increased CO2 and N2O emissions. Changing from pasture
to reed would stop that process, because the groundwater
table would be higher under this type of land use, which
will reduce the process of peat degradation; although the
resulting wetland will emit methane, a more potent
greenhouse gas than carbon dioxide, the net effect is still
positive in the sense of lower emissions [25]. In fact,
rewetting peat will in itself lead to a large reduction of
greenhouse gas emissions; producing a biofuel from the
rewetted land will reduce greenhouse gases even more, as
well as providing an alternative source of income to
dairying [26]. Fig. 1 shows the location and extent of peat
soils in the Netherlands. The total extent is some
270,000 ha, or about 8% of the land area of the country.
Apart from a source of bioenergy, reed can also be used in
a constructed wetland for the purification of surface water.
This is because reed has the capability of absorbing many
pollutants such as nitrates and phosphates. It is already used
for this purpose on an experimental scale [27]. It has been
estimated that 140,000 ha of such filters will be needed to
satisfy the requirements of the European Water Framework
Directive [28].
Another function of a reed-growing area can be water
buffering. 26% of the Netherlands lies below sea level, and
a further 40% is at risk from flooding [29]. In the past, it used to
be thought that this risk could be averted indefinitely by
simply building higher dikes and more powerful pumps. That
way of thinking has changed fundamentally in recent times:
the notion of ‘integrated water management’ enjoins
providing space for water courses where excess water can be
temporarily stored [30]. Areas are being earmarked for this
purpose, and such earmarking means a restriction on land
use: any use which would lead to major damage when the
land is flooded is prohibited. Since reed suffers relatively little
from flooding, it is a most convenient land use for water
buffering. Nor is getting rid of excess water the only reason for
such buffering: it can also be necessary to store water to cope
with dry periods in summer e a need that is likely to become
more acute with climate change. The frequency and length of
flooding depend on the purpose of buffering: from less than
once per century for perhaps a few days in the case of calamity
prevention, to several months per year for seasonal storage in
drought prevention.
Since reed is a traditional part of the Dutch landscape, it
can provide landscape functions. Providing it does not
dominate the scenery, it can be an attractive addition to it.
In order to maximize this benefit, it may be useful to
cultivate reed in combination with willow in short-rotation
coppicing. As we saw above, willow too belongs in the
wetland landscape and it is a good producer of biomass.
Reed and willow also provide shelter for numerous birds
and other animals. Although disturbance during harvesting
will somewhat limit this benefit, it can still be considerable
[31]. Moreover, in addition to any nature benefits in the
reed-growing area itself, benefits may also arise in adjacent
natural areas, especially wetlands: improvement of water
quality and quantity [32].
4. Costs and benefits of reed cultivation
We shall now attempt to put figures on the benefits mentioned
above. The first step is to assess how much energy reed can
generate. Yield figures obtained from available literature are
highly variable, as Table 1 shows. We have opted for
a cautious estimate of 15 t ha1 a1, which should be easily
achievable when reed is cultivated but without adding any
fertilizer. This is also the yield used in an economic analysis
[14] based on a review of other literature. With nutrient-rich
water fed to the reed in purification schemes, higher yields
(25 t ha1 a1) would be possible; but as the water becomes
cleaner (due to implementation of the European Water
Framework Directive) yields would decline. On the other
hand, yields in agriculture have a tendency to increase due to
technological advances, and it can reasonably expected that,
as experience is gained, this will also happen in reed cultiva-
tion. An increase in yield of 1% per year is assumed, leading to
a projected yield of 18.3 t ha1 a1 by 2030.
The amount of energy to be generated from this biomass
depends on the conversion path chosen. The relevant figures
are given in Table 2. At first sight, combustion seems to be by
far the most productive way to make energy out of biomass.
That conclusion would be rash, however: combustion is
directly useful only when heat is the intended product, i.e. in
heating systems. If combustion is used to generate electricity,
the energy yield will be only a small proportion of the amount
listed in the table. Ethanol and biogas can be used to generate
electricity as well, but because of their greater portability (as
compared to the raw biomass) they are much more widely
applicable. In other words, they command a better price. As
the last column in the table shows, on that basis ethanol
would seem to have the best yield. Nor are the relatively low
values for biogas to be scoffed at: it has the considerable
advantages that it is produced in combination with waste
flows, that it can be very well combined with farming and on
a small scale (0.5 MW).
The cost of production are estimated at 1285 V ha1 a1 for
the combustion path, including harvesting and transport to
the combustion facility (put at 10 km) [14]. For producing
ethanol, the cost of processing the reed has to be added. POET,
a large ethanol producer in the U.S., reported in late 2009 that
it was able to produce cellulosic ethanol at 0.48 V L1 (con-
verted at the then prevailing exchange rate of $1.29 per euro).
With a capacity of 2295 L ha1 (15 tonnes of biomass times
153 L per tonne), this comes to 1102 V ha1. For co-digestion,
the cost per hectare is estimated at V 1900.
We may conclude that all three methods of making bio-
energy from reed lead to a loss, and that the lowest loss is
achieved by the direct combustion method. Cellulosic ethanol
may become more profitable once a breakthrough is achieved
in the production of enzymes for breaking down the cellulose;
for now, it is possible but expensive. That the production of
biomass as such is not profitable does not mean it is not
feasible; it does mean that additional benefits are required for
reed cultivation to be worthwhile.
The first of these additional benefits, as described in the
previous section, is the impact on greenhouse gas emissions.
If we assume that the combustion of reed substitutes for that
 Author's personal copy

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2 45

Fig. 1 e Distribution of peat soils in the Netherlands.

of natural gas (whether in heating or in generating electricity),
each Gigajoule of energy from reed biomass saves 62 kg of
carbon dioxide. This means a saving of 16.4 t ha1 of carbon
dioxide, after accounting for the energy spent in producing
and transporting the biomass. If grown on peat soil, there is an
additional saving from preventing the oxidation of peat. This
saving is estimated at 1.7 t ha1, net of increased methane
emissions. At a price of about 15 V t1 at the carbon exchange
market, this represents a benefit of 245 V ha1, and 271 V on
peat soil. Wichtmann et al. give an even better figure for this
 Author's personal copy

46 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2

Table 1 e Yield figures for reed. Table 3 e Economics of multifunctional reed cultivation.
Source Reported yield of Location Production Item Value in 2010 Value in 2030
above e ground conditions
dry matter (V ha1 a1)
(T ha1 a1) Production cost of reed 1285 1285
(combustion path, including
[33] 7.5e13.0 Swamps in Europe Natural
harvesting, storage and
[34] 8.6 the Netherlands Natural
transport)
[26] 12 Temperate zone in Not stated
Value of energy production 475 772
general (conser-
Net reduction of greenhouse gas 271 542
vative estimate)
emissions (on peat soil)
[14] 15 the Netherlands Cultivated
Water buffering (where 400 400
without added
applicable)
fertilizer
Water purification 400 300
[35] 18.6 New Jersey, Natural
Landscape and nature services e e
brackish
Net benefit 261 729
tidal marshes
[36] 22e44 China Cultivated on
mine tailings,
fertilized with
sewage sludge potatoes as the main crop) in the analysis, but we found
and compost pasture to be the strongest competitor as it yields a higher net
[37] 40e63 Unspecified Cultivated with value added per hectare [41].
fertilizer Grassland yields an overall benefit of about 800 V ha1 a1
[14]. This is considerably higher than multifunctional reed-
growing. If economic conditions change over the coming
aspect of reed production: 15 t ha1 [26], which would raise the twenty years as expected in the reference scenario, reed will
climate-mitigating benefit of reed production on peat to become more competitive, but may still not outcompete dairy
490 V ha1 a1. production e unless the value of landscape and nature
The benefit of water purification is estimated at services comes to the rescue. However, other scenarios may
300e500 V ha1, and that of water buffering at V 250e550 [30]. lead to different results as we shall see below.
V 400 were assumed for both benefits. Nature benefits were
also estimate at 900 V ha1, but this is based on current
subsidies for nature management, which are dependent on 5. Spatial exploration
fickle government policies. Therefore, this component will not
be taken into account. Table 3 presents an overview, together In this study, the Land Use Scanner is used for simulating
with the figures as we project them for 2030, with an increased land-use change and allocating land for biofuel crop produc-
yield of reed, higher energy prices (medium scenario [40]) and tion. This model is based on economic and discrete choice
higher values of emission rights; the benefit of water purifi- theory and allocates cells to competing land-use classes,
cation has been reduced for 2030, as it is expected that water based on local suitability (net socio-economic benefits derived
will be cleaner anyway. The figures for 2030 have been from that location), the amount of available land and regional
computed using the assumptions pertaining to the reference demand constraints per land-use type [13]. Model character-
scenario, which is described in the section on scenarios. istics and typical applications have been described elsewhere
The net benefit to be obtained from multifunctional reed (see, for example, [42,43]). Cells have four types of character-
cultivation must be set against current land use. We have istics influencing their suitability for any particular use,
taken this to be pasture for dairy production, which is the namely present land use; physical characteristics such as soil,
main land use on peat soils. If we extend the analysis to other groundwater level and relief; proximity to urban areas, infra-
soils, an alternative competitor would be arable land. In structure and other relevant places; and zoning policies. On
another publication we have included arable land (with the basis of these characteristics, the model constructs

Table 2 e Energy yields per hectare for reed, per conversion path.
Path Fuel quantity per Energy yield per unit Energy yield per Energy yield Energy value Production
tonne of dry matter of fuel quantity [39] tonne of dry matter in GJ ha1 (V ha1, energy cost (V ha1)
prices of 2010)

Combustion 1T 17,700 MJ T1 17,700 MJ 265.5 475 1285

Ethanol 153 L3a [14] 21.1 MJ L1 3228 MJ 48.0 1151 2387
Biogas 153 m3b [38] 23 kJ L1 3519 MJ 52.8 483 1900

a 340 L per tonne of cellulose, with cellulose making up 45% of dry matter.
b Co-digesting with animal excreta with 25% reed and 75% manure. The figures represent the productivity of the entire digestate.
 Author's personal copy
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
suitability maps for any particular land use. These suitability
maps are scenario-dependent. Finally, the model is fed with
claims on space emanating from each land-use class; these
claims, too, are scenario-dependent. The model then opti-
mizes, for each scenario, future land use. The model thus
mimics a land market, in which land-use classes fulfil the role
of buyers. Their bid prices for a particular cell are determined
by the benefits they can extract from it, and those in turn
depend on the suitability maps.
Different land-use classifications can be used in the model:
simple ones determining the allocation of land between urban
uses, agriculture and nature, or more complex ones in which
various crops compete for agricultural land.
For the present study, a new land-use class was created:
reed cultivation combined with water buffering, which will be
assumed to directly compete for land with arable farming and
pastures according to their economic performance. The suit-
ability map for reed depends primarily on the second of the
four types of characteristics mentioned above, i.e. biophysical
suitability: existing land use hardly applies; as for proximity to
cities and infrastructure, this is of limited concern as we
assume that the reed would be processed within the project;
and there are at present no relevant zoning policies on reed.
The economic performance of each crop will be deter-
mined through a comparative assessment of net present
values. The net present value (NPV) is a standard method to
appraise long-term projects, by measuring discounted time
series of expected cash flows. The NPV that can be expected by
producing a certain crop depends on the specific costs to
produce that crop and on the benefits that can be obtained
from selling it:
X
n
Bc;i;y  Cc;i;y
NPVc;i ¼
y¼0
ð1 þ rÞy
where NPV is the net present value derived from land-use i in
cell c and Bc,i,y and Cc,i,y are the benefits and costs respectively
of land-use i in cell c in year y; r is the discount rate and n is the
lifetime of the project. This method has already proven to be
suitable for spatially explicit assessments on the economic
performance of biofuel crops and agricultural uses [20].
Benefits and costs will be derived from the cost-benefit anal-
ysis. A lifetime of 20 years and a discount rate of 3.5% have
been assumed, which are common for public investment
projects in the Netherlands.
The economic performance of each crop at any particular
location depends on the benefits that can be obtained from it,
which in turn largely depend on the local crop yields. Yields
can vary widely across the country according to different soil
qualities and can also vary among crops according to their
specific requirements. In order to depict yield spatial vari-
ability, yield reduction maps are used, expressing the fraction
of the maximum attainable yield that can be met when local
biophysical conditions are not optimal for that crop.
The HELP (Her-Evaluatie Landinrichtings Project) system
has been implemented in order to produce crop-specific yield
reduction maps. In this method, soil and groundwater maps
are used in combination with tables relating soil and water
conditions to suitability for particular crops. Then, physical
yields are determined by a combination of soil characteristics
(e.g. clayesandepeat ratios, rooting depth and stoniness) and
47
water tables in summer and winter [44]. The total yield
reduction (Dtot) relative to the maximum potential yield is
determined by the yield reduction caused by drought (Ddr) and
the yield reduction caused by water surplus (Dwa) assuming no
irrigation:
 
100  Dwa
Dtot ¼ Dwa þ  Ddr
100
A yield reduction map of reed was not available in the HELP
system and had to be constructed based on a literature review
of biophysical conditions influencing potential yields. Yields
of reed under different soil and groundwater conditions were
estimated on the basis of literature [45e49]. Although reed
appears to grow best on light soils, the groundwater table is
the most important factor. Furthermore, a map of suitability
for water buffering was built on the basis of altitude and
proximity to main drainage channels. Each of these two land
uses can now offer a bid price for any particular cell, and the
suitability for the multifunctional land-use class is equal to
the sum of the two bid prices. The total bid price also incor-
porates the value of reed cultivation for water purification and
for greenhouse gas mitigation (with a slightly higher value on
peat soil because of the additional CO2 benefit there).
5.1. Scenarios
Four different scenarios have been defined for examining the
potential of multifunctional reed cultivation as described
above:

A reference scenario, called ‘global cooperation’, inspired by
the storylines for the B1 scenario introduced by IPCC [50]
and further developed for Europe in the Eururalis project
[6]. It combines a global orientation with a preference for
social, environmental and more broadly defined economic
values. The scenario features high economic growth,
gradual removal of tariff barriers and agricultural subsidies,
and public funding for maintaining natural and cultural
heritage [18]. For the Netherlands this scenario has been
further specified by a Dutch forecast [51].

High oil prices and strong climate change. For the oil price,
the high variant of the U.S. Energy Information Adminis-
tration projection was taken rather than the medium
variant which is used in the reference scenario. This means
an oil price of 1495 US$2008 t1, as compared to
909 US$2008 t1 in the medium variant. For climate change,
the Wþ scenario of the climate forecasts made by the Royal
Netherlands Meteorological Institute has been used [52]. In
this scenario, average temperatures rise by 2  C until 2050;
summers are drier and winters wetter; the frequency of
extreme weather events increases; and the sea level rises
with 20e35 cm by 2050. These conditions tend to decrease
the profitability of dairy production on marginal soils, while
that of bioenergy production is improved due to higher
energy prices. Other variables (economic growth, pop-
ulation, policies) are as in the reference scenario.

The biofuel policy scenario: this has been adapted from
a study on land-use change carried out for the European
Commission’s DG Environment [18]. In this scenario, poli-
cies promoting bioenergy and biofuels are implemented not
 Author's personal copy
48 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
only by the EU, but also in the U.S., Canada, Brazil, Japan and
South Africa. This raises the price that will be paid for bio-
energy compared to the reference scenario.

The soil and climate change policy scenario: this is based
on the previous one, but with additional policies for the
protection of soils and water balances. The sustainable
use of peat soils gets particular attention in this
scenario, but also water buffering acquires increased
importance.
5.2. Results
Figs. 2 and 3 show the results obtained with the Land Use
Scanner for the four scenarios described above. In addition,
the amount of arable farming and pasture area that it con-
verted into reed cultivation is provided in Table 4. In the
reference scenario, as we saw in the section on cost-benefit
analysis, the net benefit of reed is still slightly below that of
dairy production, the main competing form of land use.
Therefore, reed mainly shows up in a few small areas which
are marginal to pasture, situated on peat soils. These areas
would be insufficient to justify the location of a processing
plant for electricity generation and/or district heating. The
only economic option would be co-digestion for making
biogas, as a by-product of dairying.
In the other scenarios, the location of a processing plant
has been included in the modelling, meaning that areas suit-
able for reed but too small and dispersed to deliver to a pro-
cessing plant have been excluded. The capacity of an
economically viable plant would be 100,000 t a1 of reed [32],
Fig. 2 e Potential locations of reed according to the reference scenario (left) and the high oil price and climate change
scenario (right).
which means an area under reed of about 6700 ha within
a radius of about 8 km.
In these scenarios, three hotspots for reed cultivation can
be clearly identified, namely the provinces of Noord-Holland,
Zuid-Holland and Friesland. It can then be seen that the
outcome of the second and the third scenario are virtually
identical in terms spatial distribution and land-use conver-
sion. In the fourth scenario, reed becomes the most attractive
land-use option on virtually all peat soils as well as on some
other low-lying areas. That would also imply a conversion of
almost 30% and 10% of the area currently used as pastures and
arable farming, respectively.
6. Discussion
From the results section above, it can be concluded that reed
cultivation for energy purposes is not economically viable in
the Netherlands under current conditions. This is in line with
previous studies (e.g. [17,20]), which showed that in normal
conditions only a small fraction of agricultural land is ex-
pected to become available for bioenergy crops, due to high
competition for land in the Netherlands. However, future
conditions in terms of energy prices, climate change and
policies may make reed more attractive. Biofuel policies may
have a similar effect as high oil prices on increasing the
profitability of bioenergy production. Combining biofuel poli-
cies with environmental measures such as soil protection
and water management can even improve its economic
performance.
 Author's personal copy

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2 49

Fig. 3 e Potential locations of reed according to biofuel policy scenario (left) and soil and climate change policy scenario
(right).

Nevertheless, it is not to be expected that the situations
depicted in Figs. 2 and 3 would come to pass in reality.
Particularly in the fourth scenario, this would imply a massive
transformation of a historical grazing landscape into large
expanses of reed-fields, which will probably be considered
undesirable.
If conditions would be such as to make reed cultivation
competitive with dairying, a combination of pasture and reed-
growing might be feasible: restricting reed-growing to ear-
marked areas, or promoting multifunctional livestock-
keeping with energy production and with the ‘blue services’
of water buffering and water purification. Such earmarked
areas cannot be too small, however, because too much scat-
tering will make the transport of reed to the processing plant
prohibitively expensive. With the minimum of 6700 ha of reed
mentioned above and a maximum distance to the processing
facility of, say, 10 km, this would mean that in an area of
300 km2 25% of all land (or 30e40% of grassland) would be
converted into reed. If other wetland species capable of be
being used for energy purposes (e.g. willow) also find a place in
such a system, this could lead to an attractive, varied and
productive landscape.
A major environmental benefit of reed cultivation as
compared to other biomass crops such as reed canary grass is
that it does not require fertilizer. In some countries it is
already used for energy, although under subsidized condi-
tions, for instance in Finland [53]. There it is not cultivated, but
harvested in the wild e the problem in many areas is con-
taining the spread of reed rather than promoting its growth,
since it is so prolific.
Thus, the cultivation of reed can be envisaged under two
systems:

Dedicated reed production in specialized farms. This is
highly convenient especially for providing both large
quantities of reed and water purification services. However,
even in this system landscape considerations would require
that only part of the potential area would be actually used
for reed. The same considerations might make it attractive
to combine the production of reed with for instance willow

Table 4 e Land-use conversion for reed cultivation (in hectares).

Land-use conversion Reference High oil prices Biofuel policy Soil and climate
scenario and climate change change policy

Arable farming 2924 49,827 42,562 82,060

Pastures 21,203 207,826 198,218 420,477
 Author's personal copy
50 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
in short-rotation coppicing. To avoid a large-scale conver-
sion of pastoral landscape, other options could be consid-
ered such as only partly producing part the required
bioenergy feedstock domestically while importing the
remaining from other countries, thus taking advantage of
the existence of good port facilities in the areas where reed
is more attractive.

A combination of animal husbandry and reed cultivation.
The production of biogas would be the most appropriate
conversion path in this case. The livestock part of the farm
could be conventional dairying as practised today (where
soil conditions allow), or some lower-intensity form of
cattle-keeping using dual-purpose, hardy breeds rather
than the Holstein-Friesian type kept today. Sheep might
also play a role in such systems. Such farmers would derive
their income from a variety of sources: dairy, meat, energy
production, landscape and water services.
Conditions favouring reed cultivation are rising energy
prices and climate change, in particular a rising sea level. The
former will make bioenergy more profitable, the latter will
make dairying in low-lying fens with their peat soils even less
sustainable than it is today. Liberalization of agricultural
markets may put further pressure on dairying margins. Even
with more modest energy prices and more moderate climate
change, conditions favourable to reed production can arise as
a consequence of policies for stimulating bioenergy, protect-
ing water balances, mitigating climate change and conserving
vulnerable soils.
In such situations, reed cultivation can be an attractive
alternative particularly in the fens, but e being a crop with
a low unit value e only if it can yield other benefits beyond
bioenergy. These benefits are water purification, reduction of
greenhouse gas emissions, water buffering and landscape/
nature services. To the extent that policies will allow land-
users to reap part of these benefits, such land use will also
be commercially attractive. Our spatial simulation even
shows that policies may be needed to prevent reed cultivation
from overwhelming current pasture land use.
7. Conclusion
There is considerable interest in the potential for reed culti-
vation as a form of multifunctional land use, generating
biomass for energy as well as benefits in terms of water
retention and/or water purification, and possibly even
enhancing the value of nature and landscape. This paper
analyzes that potential for the Netherlands in terms of costs
and benefits, both to land-users and to society as a whole. It
also projects spatial patterns for such land use under different
scenarios. Several possible ways of converting biomass into
energy are explored, of which direct combustion either for
district heating or for energy generation appears the most
promising e at least with the present state of technology.
It turns out that reed cannot compete with present agri-
cultural land use under present conditions. However, these
conditions may change in decades to come: energy prices may
increase quite dramatically, climate change may make
present land use more difficult to sustain (in particular
pasture on peat soils), and technical progress may make
cellulosic ethanol an attractive option. All such changes will
create more favourable conditions for reed.
Peat soils are particularly suitable for this form of land use.
Moreover, on these soils there is an additional environmental
benefit to be had: these soils continuously emit carbon dioxide
under present land use (mostly grassland for dairy produc-
tion). Converting such land into reed-fields would imply
rewetting it, which would stop the emission of CO2 and, if
peat-forming conditions are revived, even lead to the storage
of carbon in the soil.
references
[1] Directive 2009/28/EC. On the promotion of the use of energy
from renewable sources and amending and subsequently
repealing Directives 2001/77/EC and 2003/30/EC.2009.04.23.
Off J European Union 2009;L 140:16e62.
[2] Roubanis N, Dahlström C, Noizette P. Renewable energy
statistics. European Union: Eurostat, vol. 7; 2010. Report No.:
56/2010.
[3] Petrou EC, Pappis CP. Biofuels: a survey on pros and cons.
Energy Fuel 2009;23:1055e66.
[4] Hamelinck CN, Faaij APC. Outlook for advanced biofuels.
Energy Policy 2006;34(17):3268e83.
[5] Banse M, Van Meijl H, Tabeau A, Woltjer G. Will EU biofuel
policies affect global agricultural markets? Eur Rev Agric
Econ 2008;35(2):117e41.
[6] Eickhout B, Notenboom J, Ros JPM, Van den Born GJ, Van
Oorschot M, Van Vuuren DP, et al. Local and global
consequences of the EU renewable directive for biofuels.
Bilthoven: Planbureau voor de Leefomgeving 2008;6:70.
Report No.: 500143001.
[7] Fargione J, Hill J, Tilman D, Polasky S, Hawthorne P. Land
clearing and the biofuel carbon debt. Science 2008;319(5867):
1235e8.
[8] Pimentel D, Patzek TW. Ethanol production using corn,
switchgrass, and wood; biodiesel production using
soybean and sunflower. Nat Resour Res 2005;14(1):65e76.
[9] Stephenson AL, von Blottnitz H, Brent AC, Dennis JS,
Scott SA. Global warming potential and fossil-energy
requirements of biodiesel production scenarios in South
Africa. Energy Fuel 2010;24:2489e99.
[10] Crutzen PJ, Mosier AR, Smith KA, Winiwarter W. N2O release
from agro-biofuel production negates global warming
reduction by replacing fossil fuels. Atmos Chem Phys 2008;
8(2):389e95.
[11] Smeets EMW, Faaij APC, Lewandowski IM, Turkenburg WC.
A bottom-up assessment and review of global bio-
energy potentials to 2050. Prog Energy Combust 2007;33(1):
56e106.
[12] van Dam J, Faaij APC, Hilbert J, Petruzzi H, Turkenburg WC.
Large-scale bioenergy production from soybeans and
switchgrass in Argentina Part B. Environmental and socio-
economic impacts on a regional level. Renew Sust Energy
Rev 2009;13(8):1679e709.
[13] Hilferink M, Rietveld P. Land use scanner: an integrated GIS
based model for long term projections of land use in urban
and rural areas. J Geogr Syst 1999;1(2):155e77.
[14] Daatselaar CHG, Hoogendam K, Poppe KJ. De economie van
het veenrietweidebedrijf. Een quickscan voor West-
Nederland [The economics of reed cultivation on peat soils e
an overview for the Western Netherlands]. Utrecht:
Innovatie Netwerk; 2009. 51 p. Report No.: 09.2.218.
 Author's personal copy
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
[15] Alcamo J, Kok K, Busch G, Priess J. Searching for the future of
land: scenarios from the local to global scale. In: Lambin E,
Geist H, editors. Land-use and land-cover change: local
processes and global impacts. Dordrecht: Springer; 2006. p.
137e55.
[16] Hoogwijk M, Faaij A, Eickhout B, de Vries B, Turkenburg W.
Potential of biomass energy out to 2100, for four IPCC SRES
land-use scenarios. Biomass Bioenergy 2005;29:225e57.
[17] Hellmann F, Verburg PH. Spatially explicit modelling of
biofuel crops in Europe. Biomass Bioenergy 2011;35(6):
2411e24.
[18] Perez-Soba M, Verburg PH, Koomen E, Hilferink M, Benito P,
Lesschen JP, et al. Land use modelling e implementation;
preserving and enhancing the environmental benefits
of land-use services. Final report to the European
commission, DG environment. Wageningen: Alterra
Wageningen UR/Geodan Next/Object Vision/BIOS/LEI and
PBL; 2010. 145 p.
[19] Van der Hilst F, Faaij A, van den Bos A, Hamelinck CN, van de
Staaij J, van den Heuvel E. Strategic evaluation large scale
sustainable bioenergy export from Mozambique to the
Netherlands. Utrecht: Ecofys/Utrecht University; 2011. 82 p.
Report No.: PEGENL084956.
[20] Van der Hilst F, Dornburg V, Sanders JPM, Elbersen B, Van
Dam JMC, Faaij APC. Potential, spatial distribution and
economic performance of regional biomass chains: the
North of the Netherlands as example. Agr Syst 2010;103(7):
403e17.
[21] Tittmann PW, Parker NC, Hart QJ, Jenkins BM. A spatially
explicit techno-economic model of bioenergy and biofuels
production in California. J Transp Geogr 2010;18(6):715e28.
[22] Giarola S, Zamboni A, Bezzo F. Spatially explicit
multi-objective optimisation for design and planning of
hybrid first and second generation biorefineries. Comput
Chem Eng 2011;35(9):1782e97.
[23] Irwin EG, Geoghegan J. Theory, data, methods: developing
spatially explicit economic models of land use change. Agr
Ecosyst Environ 2001;85(1e3):7e23.
[24] Londo HM. Energy farming in multiple land use. An
opportunity for energy crop introduction in the Netherlands
[dissertation]. Utrecht: Utrecht University, 2002.
[25] Borren W. Carbon exchange in Western Siberian watershed
mires and implication for the greenhouse effect: a spatial
temporal modelling approach [dissertation]. Utrecht: Utrecht
University, 2007.
[26] Wichtmann W, Tanneberger F, Wichmann S, Joosten H.
Paludiculture is paludifuture. Climate, biodiversity and
economic benefits from agriculture and forestry on rewetted
peatland. Peatlands Int 2010;(1):48e51.
[27] Blaeij A, Reinhard AJ. Een waterpark als alternatief: MKBA
aanleg multifunctioneel helofytenfiler op Waterpark Het
Lankheet [A waterpark as an alternative: a social cost-benefit
analysis of a multifunctional construction on wetlands in
Lankheet Waterpark]. The Hague: LEI; 2008. 75 p. Report No.:
2008-061.
[28] Ligtvoet W, Beugelink G, Brink C, Franken R, Kragt F.
Kwaliteit voor later: Ex Ante evaluatie Kaderrichtlijn Water
[Quality for later: Ex Ante evaluation of water framework
directive]. Bilthoven: Planbureau voor de Leefomgeving;
2008. 212 p. Report No.: 50014001.
[29] Ministry of Transport. Public works and water management.
Water in the Netherlands 2004e2005: facts and figures. The
Hague: Ministry of Transport, Public Works and Water
Management; 2004. 72 p.
[30] van Steen PJM, Pellenbarg P. Water management in the
Netherlands: introduction to the 2004 maps. Groningen:
Rijksuniversiteit Groningen; 2004. 3 p.
51
[31] Wichtmann W, Wichmann S. Environmental, social and
economic aspects of a sustainable biomass production. J Sust
Energy Environ 2011;(Special Issue):77e81.
[32] Grandiek N, van Herk J, Cronenberg C. De introductie van de
rieteconomie: Een duurzaam perspectief voor de
veenweidegebieden [The introduction of the reed economy:
A sustainable perspective for the Dutch peatlands]. Utrecht:
Innovatie Netwerk; 2007. 75 p. Report No.: 07.2.155.
[33] Duke JA. Ecosystematic data on economic plants. Q J Crude
Drug Res 1979;17(3e4):91e110.
[34] Best EPH, Zippin M, Dassen JHA. Growth and production of
Phragmites australis in lake Vechten (The Netherlands).
Aquat Ecol 1981;15(3):165e73.
[35] Windham L. Comparison of biomass production and
decomposition between Phragmites australis (common reed)
and Spartina patens (salt hay grass) in brackish tidal
marshes of New Jersey, USA. Wetlands 2001;21(2):179e88.
[36] Chiu KK, Ye ZH, Wong MH. Growth of Vetiveria zizanioides
and Phragmites australis on Pb/Zn and Cu mine tailings
amended with manure compost and sewage sludge:
a greenhouse study. Bioresour Technol 2006;97(1):1e14.
[37] Duke JA. Handbook of energy crops; 1983. West Lafayette
(Ind.): Purdue University, Center for New Crops & Plant
Products; Unpublished report, available at: http://www.hort.
purdue.edu/newcrop/duke_energy/Phragmites_australis.
html [accessed 10.06.11].
[38] Bioenergy Feedstock Information Network. Conversion
factors. Available at: http://bioenergy.ornl.gov/papers/misc/
energy_conv.html; 2011 [accessed 09.12.11].
[39] Gravalos I, Kateris D, Xyradakis P, Gialamas T, Loutridis S,
Augousti A, Georgiades A, Tsiropoulos Z. A study on calorific
energy values of biomass residue pellets for heating
purposes. Proceedings of the 43rd FORMEC conference on
Forest Engineering: Meeting the Needs of the Society and the
Environment; 2010 July 11e14, Padova, Italy. Available at:
http://www.tesaf.unipd.it/formec2010/01Proceedings.htm
[accessed 09.12.11].
[40] US Energy Information Administration. Analysis and
projections. Available at, http://www.eia.doe.gov/oiaf/
forecasting.html 2010; 2011 [accessed 09.12.11].
[41] Kuhlman T, Verburg R, Van Dijk J, Phan-Drost N. Biomass on
peat soils? Feasibility of bioenergy production under
a climate change scenario. In: Koomen E, Borsboom-van
Beurden J, editors. Land-use modelling in planning practice.
Heidelberg: Springer; 2011. p. 97e116.
[42] Koomen E, Kuhlman T, Groen J, Bouwman AA. Simulating
the future of agricultural land use in the Netherlands.
Tijdschrift voor Economische en Sociale Geografie (Journal of
Economic and Social Geography) 2005;96(2):218e24.
[43] Koomen E, Borsboom-van Beurden J. Land-use modeling in
planning practice. 1st ed. Heidelberg: Springer; 2011.
[44] Brouwer F, Huinink J. Opbrengstdervingspercentages voor
combinaties van bodemtypen en grondwatertrappen:
geactualiseerde HELP-tabellen en
opbrangstdepressiekaarten [Yield reduction for
combinations of soil types and groundwater level: HELP-
updated tables and yield reduction maps]. Wageningen:
Alterra; 2002. 38 p. Report No.: 429/MH/06-2002.
[45] Hellings SE, Gallagher JL. The effects of salinity and flooding
on Phragmites-australis. J Appl Ecol 1992;29(1):41e9.
[46] Coops H, van den Brink FWB, van der Velde G. Growth and
morphological responses of four helophyte species in an
experimental water-depth gradient. Aquatic Botany 1996;
54(1):11e24.
[47] Clevering OA. Effects of litter accumulation and water table
on morphology and productivity of Phragmites australis.
Wetlands Ecol Manage 1998;5:275e87.
 Author's personal copy
52 b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 4 1 e5 2
[48] Lenssen JPM, Menting FBJ, van der Putten WH, Blom CWPM.
Effects of sediment type and water level on biomass production
of wetland plant species. Aquat Bot 1999;64(2):151e65.
[49] Tulbure MG, Johnston CA. Environmental conditions
promoting non-native Phragmites australis expansion in
great Lakes Coastal wetlands. Wetlands 2010;30(3):577e87.
[50] International Panel on Climate Change. Emission scenarios.
Special report of the intergovernmental panel on climate
change. Cambridge: Cambridge University Press; 2000. 599 p.
[51] Janssen L, Okker V, Schuur J. Welvaart en Leefomgeving: Een
scenariostudie voor Nederland in 2040 [Prosperity, health
and safety: a scenario study for the Netherlands in 2040]. The
Hague/Bilthoven: Centraal Planbureau, Milieu- en
Natuurplanbureau, Ruimtelijk Planbureau; 2006. 239 p.
Report No.: 500082001.
[52] Van den Hurk B, Klein Tank A, Lenderink G, van
Oldenborgh GJ, Katsman C, van den Brink H, et al. WR 2006-
01: KNMI climate change scenarios 2006 for the Netherlands.
De Bilt: Koninklijk Nederlands Meteorologisch Instituut;
2006. 81 p.
[53] Komulainen M, Simi P, Hagelberg E, Ikonen I, Lyytinen S.
Reed energy e possibilities of using the common reed for
energy generation in Southern Finland. Turku: Turku
University of Applied Sciences; 2007. 78 p. Report No.: 67.