Field Crops Research 63 (1999) 3±11

Effect of irrigation and nitrogen fertilization on biomass yield and

ef®ciency of energy use in crop production of Miscanthus
L. Ercolia,*, M. Mariottib, A. Masonib, E. Bonaria
a
Scuola Superiore di Studi Universitari e di Perfezionamento S. Anna, via Carducci, 40, 56100, Pisa, Italy
b
Dipartimento di Agronomia e Gestione dell'Agroecosistema, via S. Michele degli Scalzi, 2, 56100, Pisa, Italy
Received 11 August 1998; accepted 20 March 1999

Abstract

The perennial C4 grass Miscanthus has been proposed as a biomass energy crop in Europe. Effects of crop age, irrigation and
nitrogen fertilization on biomass and energy yields and N content of Miscanthus were investigated and the energy costs of
production determined. After an establishment period of 1 year, cultivation of Miscanthus resulted in a dry matter production
of over 37 t haÿ1 yearÿ1 over a period of 4 years. Irrigation and nitrogen level greatly affected Miscanthus biomass yield. In
absence of N fertilization, irrigation did not modify biomass yield and the effect of irrigation increased with the increase in N
level. The average N response ranged from 37 to 50 kg biomass kgÿ1 N applied. Because the calori®c value of Miscanthus
biomass (16.5 MJ kgÿ1) was not affected by irrigation and N fertilization, energy production depended exclusively on biomass
yield. Maximum energy yield was 564 GJ haÿ1 yearÿ1. Without N supply and irrigation, energy yield was 291 GJ hÿ1. Net
energy yield, calculated as the difference between energy output and input, but without inclusion of drying costs, was
543 GJ haÿ1 with N fertilization and irrigation and 284 GJ haÿ1 without; the ratios of energy output to input in crop
production were 22 and 47, respectively. # 1999 Elsevier Science B.V. All rights reserved.

Keywords: Crop age; Energetic ef®ciency; Fertilization; Irrigation; Miscanthus; Nitrogen

1. Introduction (Schwarz, 1993; van der Werf et al., 1993; Hotz, 1996;
Himken et al., 1997; Venendaal et al., 1997).
Several plant species have been investigated for Miscanthus is a C4 rhizomatous perennial grass
their potential as biomass crops. As biomass conver- originating from Asia. In temperate climates, its
sion techniques have been advanced, interest has growth is limited by low temperature. In Europe it
focused on the cultivation of Miscanthus species, begins growth in April and continues until halted by
which have potential for high yields of biomass in frost in November. In winter the above-ground parts
European conditions with yields of 25 to are killed by frost. Regrowth occurs from crowns in
35 t haÿ1 yearÿ1 once the crop has been established spring. Nutrients and carbohydrates stored in rhi-
zomes during fall are mobilized to shoots in late
April/early May, supporting rapid growth. The plant
*Corresponding author. Fax: +39-50-883225 has an extensive root system that responds quickly to a
E-mail address: ercoli@sssup1.sssup.it (L. Ercoli) rapid demand for nutrients during spring growth, thus

0378-4290/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 3 7 8 - 4 2 9 0 ( 9 9 ) 0 0 0 2 2 - 2
4 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11

reducing the risk of nitrate leaching (Himken et al., and chemical properties were 34% sand, 21% silt,
1997). 45% clay, pH 7.2, 2.15% organic matter (Walkley-Black
Nursery plants are planted in spring and the crop method), 0.12% nitrogen (Kjeldahl method), 33 mg gÿ1
can be harvested annually for up to 15 years after available P (Olsen method), and 22 mg gÿ1 available K
establishment (Schwarz et al., 1994). Although max- (Dirks Scheffer method). During the trial period, no
imum above-ground dry matter yield is attained in late water table was observed within the top 1.2 m of soil.
summer, harvest can be delayed until February/March Age of crop at harvest was 1, 2, 3 or 4 years.
when the crop has its highest dry matter concentration Treatments were crop age, irrigation, and nitrogen
(Petrini et al., 1996; Himken et al., 1997). This is rate. Irrigation treatments were unirrigated and repla-
important for net energy production, as energy cement of 100% estimated evapotranspiration. Nitro-
requirements for drying of the plant material prior gen treatments were 0, 100 and 200 kg haÿ1 of N.
to combustion is less for standing dead material than These were applied in the establishment year (1992),
for green material. At harvest, shoots are mechanically as 50% preplant and 50% sidedressing when Mis-
cut near ground level. canthus plants were 0.35±0.45 m tall. In the following
The C4 photosynthetic pathway contributes to high years, N fertilizer was applied entirely at the start of
water-use ef®ciency and to a low nitrogen content of growth in spring (approximately on 15 April). Nitro-
the biomass. Long (1983) and Beale and Long (1997) gen was applied as urea.
calculated a requirement of 450 mm water and 92 kg The experimental design was a split plot with year
N haÿ1 for a crop producing an above-ground harvest as main plots, irrigation treatments as sub-plots and
of 15 t haÿ1 dry matter. It has been reported that the nitrogen rates as sub-sub-plots. Plots were replicated
effect of N fertilization on biomass yield is not high: three times and measured 6 m long and 5 m wide,
Schwarz et al. (1994), reported an increase of only separated by a minimum of 3 m. Miscanthus cv.
1.1 t haÿ1 in biomass dry weight (from 20.6 to Giganteus was used. Planting was carried out on 16
21.7 t haÿ1) with an increase in N level from 0 to April 1992 using potted plants produced by micro-
180 kg haÿ1 in a 3-year-old crop. Himken et al. propagation. At that time, the plants were about
(1997) calculated an annual removal of N from the ®eld 0.20 m tall. Miscanthus was grown in 0.50 m wide
(equivalent to the fertilizer demand of an established rows at a population of 400 00 plants haÿ1. The pre-
Miscanthus crop) around 50 to 70 kg N haÿ1 in a 4-year- vious crop was barley. Tillage was conducted in the
old crop. Christian et al. (1997) found, in 1-year-old autumn of 1991 and consisted of medium-depth
plants, that of the 117 kg haÿ1 N taken up by the crop, ploughing (30±40 cm). Seedbed preparation was con-
only 23 kg haÿ1 (38%) was derived from the fertilizer. ducted in spring, immediately before planting, by a
Long-term experience on Miscanthus is still lack- pass with a double-disking harrow and a pass with a
ing, especially about how to establish a good crop, ®eld cultivator. Preplant fertilizer was distributed at a
about its nutritional demands and how yield will vary rate of 87 kg P haÿ1 (triple superphosphate) and 83 kg
in the course of an estimated 10±20 years of usage. K haÿ1 (potassium sulphate). Plots were kept weed-
The objective of this research was to determine the free by handweeding. Water was distributed by drip
effects of irrigation and nitrogen fertilizer on yield and irrigation. After sidedressing, irrigation lines were
nitrogen content of Miscanthus sinensis cv. Giganteus. permanently installed in the centre of the interrow.
In addition, the use of energy resources was analyzed Flow application rate was 4L hÿ1 mÿ1 of tubing (1
to determine the energy cost of management techni- dripper per metre). The soil pro®le was presumed to be
ques and the energy balance of the crop. at ®eld capacity at planting. Irrigation treatment was
designed so that rainfall plus irrigation replaced a
2. Materials and methods portion of the soil moisture lost to estimated evapo-
transpiration. The amount of water given daily was
2.1. Crop culture based upon the potential evapotranspiration data (E0)
of the previous day, estimated from Class A pan
A ®eld study was conducted at Pisa (438400 N, evaporation. Actual evapotranspiration was calculated
108190 E) in Italy from 1992 to 1995. Soil physical as E ˆ kcE0 where kc is the crop coef®cient. Because
 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11 5

Table 1
Seasonal irrigation and precipitation during the four research years

Year Water treatment Seasonal irrigation (mm) Precipitationa (mm) Total (mm)

1992 Rainfed ± 399 399

Irrigated 224 399 623
1993 Rainfed ± 131 131
Irrigated 228 131 359
1994 Rainfed ± 201 201
Irrigated 250 201 451
1995 Rainfed ± 186 186
Irrigated 219 186 405
a
Total precipitation from planting or start of growth until harvest. The 110-year-average precipitation from 15 April to 30 September equals
330 mm.

Miscanthus kc is not known, we used a value for fabrication and repairs of machinery, for fertilizer
sorghum as reported by Tarantino and Onofri and planting material and for fuel consumption for
(1988), which increases from 0.33 at 20 days after the various operations. Energy costs for delivering the
emergence to 1.23 at anthesis and declines to 0.76 at production outside the ®eld and for storage were not
maturity. Irrigation, rainfall before and after the begin- calculated. In our research, Miscanthus was propa-
ning of irrigation, and total water supply are presented gated with in vitro micropropagated plants, but,
in Table 1. Irrigation periods were 24 July±8 August hypothesising establishment in an extensive cultiva-
1992, 28 July±16 August 1993, 30 June±22 August tion system, the best method of propagation is by
1994 and 27 June±9 August 1995. No pest infestation rhizome cuttings. Therefore, that method is consid-
was detected during the experimental period. ered in the determination of crop energy cost follow-
Harvesting was carried out at anthesis, when max- ing the approach suggested by Heichel (1980).
imum biomass yield occurred according to Petrini et Because the harvested plant material was anatomi-
al. (1996). Harvest dates were 14 October in 1992, 5 cally similar to the organs or tissues used for establish-
October in 1993, 8 October in 1994 and 10 October in ing the crop, the quantity of material used for
1995. Plants in a 1 m2 area were harvested by cutting propagation was subtracted from the total yield of
at ground level, separated into stems (including leaf the crop, with the output energy of the cropping
sheaths and in¯orescences) and leaf lamina, and system being the `net above seed requirements'. In
weighed to determine fresh weight. In all ®gures our analysis, the energy cost of propagation material
and tables, these fractions are referred to as stems and was considered negligible, since the biomass involved
leaves. The subsamples were placed in a forced-draft was less than 3% of the total yield of the crop.
oven at 758C for 72 h and were ground after determina- Energy inputs for machinery were determined,
tion of dry weight. Nitrogen concentration was deter- following Doering (1980), by estimating energy con-
mined by the micro-Kjeldahl method. A homogeneous sumption for the fabrication and the repairing of the
dried sample of the entire aerial plant part was ground to machinery utilized for Miscanthus cultivation, and by
pass through No. 4 sieve (4.75 mm), and burnt in an calculating the annual per hectare machinery cost. We
oxygen bomb calorimeter (Gallenkamp autobomb, Lei- assumed that machinery and implements were used on
cester, UK) to determine calori®c value. The energy 200 ha, and machine life was 10 years. In Table 2, the
yield of Miscanthus biomass was calculated by multi- machinery used in Miscanthus cultivation and the
plying the calori®c value per unit of biomass by the weight of each piece of machinery are reported.
above-ground dry matter yield. The energy cost for fertilizer manufacturing was
59.9 MJ kgÿ1 for N, 5.5 MJ kgÿ1 for P and
2.2. Energy analysis 5.6 MJ kgÿ1 for K (Lockeretz, 1980). Fuel costs of
various management operations were calculated
Energy analysis of Miscanthus biomass production (Table 3) by determining diesel consumption and by
was carried out by determining energy costs for multiplying those values by the heat of combustion of
6 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11

Table 2
Weight of machinery utilized for Miscanthus cultivation, annual per hectare machinery weight and energy consumption for the fabrication and
repair parts and materials of the machinery. We assumed that machinery and implements were used on 200 ha, and machine life was 10 years.
Tractors utilized were 204 kW for plowing, 74 kW for disk harrowing, 59 kW for rotary harrowing, and 37 kW for fertilization and harrowing

Operation Weight (kg) Energy costa

Tractor Implement Tractor ‡ Implement

kg haÿ1 yearÿ1 MJ haÿ1 yearÿ1

Plowing 6680 1350 4.0 301

Disk harrowing 4980 1510 3.2 241
Rotary harrowing 3960 650 2.3 173
Fertilization 50% N±P±K 1980 160 1.1 83
Fertilization 50% N 1980 160 1.1 83
Planting 2300 ± 1.2 90
Harvesting 1980 5800 3.9 294
a
The energy coefficient used in 75.36 MJ kgÿ1.

diesel fuel (45.8 MJ kgÿ1). While we used microirri- 3. Results

gation in the experiment in order to have accurate
water management, crop energy cost was based on the Precipitation during the 1992 growing season was
sprinkler irrigation method. The energy cost for irri- 69 mm greater than the 110-year-average (330 mm),
gation was determined by the formula reported by while those in 1993, 1994 and 1995 were lower by
Batty and Keller (1980) and adapted from Knutson et 199, 129 and 144 mm respectively (Table 1). The
al. (1977): energy cost (MJ) ˆ 550  L  A, where L amounts of water distributed by irrigation were
is the lift in meters and A is the amount of water about the same in all years, ranging from 219 to
pumped in m haÿ1. It was assumed that water was 250 mm.
lifted 10 m and delivered to the ®eld with a 20% loss in
conveyance and application. 3.1. Dry weight
All variables were analyzed with standard split-plot
analysis of variance techniques to test effects of year, Dry weight of Miscanthus biomass, stems and
irrigation, nitrogen supply, and their interactions. Sig- leaves were affected by year of harvest, irrigation
ni®cantly different means were separated at 0.05 and nitrogen level mean effects, and by irriga-
probability level by the least signi®cant difference tion  nitrogen level interaction, while it was not
test (Snedecor and Cochran, 1980). affected by the other interactions. Averaged over

Table 3
Annual fuel energetic cost for machine operations involved in Miscanthus production

N level Plowing Harrowing (MJ haÿ1) Fertilization (MJ haÿ1) Planting Harvesting Total
(kg haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1)
Disk Rotary P±K N

Establishment year
0 1924 651 504 111 837 3770 7797
100 1924 651 504 111 90 837 3770 7887
200 1924 651 504 111 137 837 3770 7934

Following yearsa
0 ± ± ± ± ± 3770 3770
100 ± ± ± ± 90 ± 3770 3860
200 ± ± ± ± 137 ± 3770 3907
a
Mean values of year 2, 3 and 4.
 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11 7

Table 4
Dry weight and nitrogen content of leaves, stems and total biomass
of Miscanthus observed over four production years. Values are
averaged over three N rates and irrigated/rainfed conditions

Year of harvest Leaves Stems Biomass

ÿ1
Dry weight (t ha )
1 5.0 a 10 a 15 a
2 5.6 ab 20 b 26 b
3 6.6 b 20 b 27 b
4 6.7 b 19 b 26 b

N content (kg haÿ1)

1 95 a 47 a 143 a
2 75 b 82 c 158 a
3 77 b 77 bc 144 a
4 74 b 70 b 143 a
Values within the same column followed by the same letter are not
significantly different at the 0.05 probability level.

Fig. 1. Dry weight (top) and nitrogen content (bottom) of rainfed

and irrigated Miscanthus crops in response to nitrogen fertilization.
irrigation and nitrogen levels, dry weight of biomass, Values are averaged over 4 years.
stems and leaves increased from the establishment
year to the following one and did not vary thereafter
(Table 4). 3.2. Nitrogen concentration and content
Nitrogen level and irrigation increased dry weight
of biomass. The effect of irrigation depended on Nitrogen concentrations of biomass, leaves, and
nitrogen level: in absence of nitrogen fertilization, stems were affected only by N fertilization. When
dry weight of biomass was not affected by irrigation N level increased from 0 to 200 kg haÿ1, N concen-
and increased with N level in both rainfed and irri-
gated crops. The increase was greater in irrigated
crops than in the rainfed crops, thus with 100 kg
N haÿ1 irrigation increased dry weight of biomass
by 3.7 t haÿ1 and with 200 kg N haÿ1 irrigation
increased dry weight of biomass by 9.8 t haÿ1
(Fig. 1). Maximum biomass of Miscanthus crop
was 37.5 t haÿ1, and was attained in the 3-year-old
crop with 200 kg N haÿ1 and irrigation.
The partitioning of biomass between leaves and
stems varied over years and was not affected either
by irrigation or nitrogen level. From the establishment
year to the following ones, the proportion of leaves
decreased and that of stems increased, and no sig-
ni®cant difference was detected among the older crops
(Fig. 2).
Water content of Miscanthus biomass, which is
important because it affects both transport and storage,
Fig. 2. Partitioning of Miscanthus dry weight (top) and nitrogen
was not affected by year, irrigation or nitrogen level. content (bottom) between leaves (empty area) and stems (filled
Averaged over all treatments, water content of leaves area) observed over four production years. Values are averaged
was about 65% and that of stems about 47%. over three N rates and irrigated/rainfed conditions.
8 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11

Table 5 (Table 6). The sum of machinery and fuel inputs

Nitrogen concentration of leaves, stems and total biomass of represents the cost of all management techniques
Miscanthus in response to three nitrogen levels. Values are
averaged over 4 years and irrigated/rainfed conditions
applied to the crop. After the establishment year, it
amounted to 100% of the total energy needed for crop
Nitrogen level Leaves Stems Biomass production in absence of irrigation and N fertilization,
(kg haÿ1) (g kgÿ1) (g kgÿ1) (g kgÿ1)
and to 25% with both irrigation and highest N level.
0 10.9 a 2.8 a 4.8 a The use of N fertilizer involved energy costs, each
100 13.7 b 3.9 b 6.4 b year, of 6 GJ haÿ1 with 100 kg N haÿ1 and of
200 14.8 c 4.8 c 7.5 c
12 GJ haÿ1 with 200 kg N haÿ1. These values repre-
Values within the same column followed by the same letter are not sent 32% and 48% of total energy inputs in the
significantly different at the 0.05 probability level. establishing year and 58% and 73% in the following
years, respectively. Irrigation required less energy:
tration, averaged over years and irrigation level, only 1.5 GJ haÿ1, representing 6±10% of total energy
increased from 4.8 to 7.5 g kg ÿ1 in biomass (Table 5). inputs in the establishing year and 6±27% in the
Nitrogen content of biomass, stems and leaves was following ones.
affected by irrigation  nitrogen interaction and N The average calori®c value of dry Miscanthus
content of stems and leaves was also affected by crop biomass (16.5 MJ kgÿ1) was not affected by nitrogen
age mean effect. Nitrogen content of leaves was supply or irrigation in any year of the trial. Therefore,
95 kg haÿ1 in the establishment year and decreased the effects of treatments on energy yield of the crop
to 74 kg haÿ1 in the following years (Table 4). By were equal to those on biomass yield. Maximum
contrast, nitrogen content of stems increased from energy yield of Miscanthus was 564 GJ haÿ1, obtained
47 kg haÿ1 in the establishment year to about with 200 kg N haÿ1 and irrigation. On the contrary,
75 kg haÿ1 in the following ones. As a consequence, without N supply and irrigation, energy yield was
N content of biomass was not signi®cantly affected 291 GJ haÿ1.
over years. The partitioning of N content between To evaluate the performance of the Miscanthus
leaves and stems was affected only by year of harvest bioenergy system, we considered net energy yield,
(Fig. 2). Over years, the quantity of N in leaves calculated as the difference between energy output
decreased and that in stems increased, but no signi®- and energy input per hectare, and ef®ciency of energy
cant differences were detected after year 1. production, calculated as the ratio between energy
Similar to dry weight, N content of biomass was not output and energy input per hectare. Both net energy
affected by irrigation in absence of nitrogen fertiliza- yield and ef®ciency of energy production were
tion, while N fertilizer increased N content of biomass affected by irrigation and N level. Without N fertiliza-
with a greater effect when the crop was irrigated. tion, net energy yield was not affected by irrigation,
Averaged over years, N content of biomass of the whereas it increased with the increase of N level in
rainfed crop was increased by the highest N level by both rainfed and irrigated crops. The increase was
92 kg haÿ1, and that of the irrigated crop by greater in the irrigated crop than in the rainfed one, in
171 kg haÿ1 (Fig. 1). The effects on N content of fact with 100 kg N haÿ1, irrigation increased net
stems and leaves was similar to that on biomass. energy yield by 60 GJ haÿ1 and, with 200 kg N haÿ1,
The maximum N content of Miscanthus crop was irrigation increased net energy yield by 160 GJ haÿ1
290 kg haÿ1, attained in the 3-year-old crop with (Fig. 3). Maximum net energy yield, of about
200 kg N haÿ1 and irrigation. 600 GJ haÿ1, was recorded in the 3-year-old crop
irrigated and fertilized with the highest N level.
3.3. Energy balance The ef®ciency of energy production decreased with
the increase in N fertilization but the effect was
The total energy input of Miscanthus production different whether the crop was rainfed or irrigated
ranged from 13 to 27 GJ haÿ1 in the establishing year (Fig. 3). Averaged over years, energy ef®ciency was
and from 4 to 18 GJ haÿ1 in the following years, greater in the rainfed crop than in the irrigated one in
depending on the irrigation or nitrogen fertilization absence of N fertilizer, whereas with 100 kg N haÿ1
 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11 9

Table 6
Annual energy inputs for Miscanthus cultivation in the establishment year and in the following years in response to irrigation and to three
nitrogen levels

Irrigation N level Machinery Fuel Phosphorus Potassium Nitrogen Irrigation Total

(kg haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1) (MJ haÿ1)

Establishment year
Rainfed 0 1182 7797 2512 1340 12 831
100 1265 7887 2512 1340 5 987 18 991
200 1265 7934 2512 1340 11 970 25 021

Irrigated 0 1182 7797 2512 1340 1479 14 310

100 1265 7887 2512 1340 5 987 1479 20 470
200 1265 7934 2512 1340 11 970 1479 26 500

Following yearsa
Rainfed 0 294 3770 ± ± 4 064
100 377 3907 ± ± 5 987 10 271
200 377 3979 ± ± 11 970 16 326

Irrigated 0 294 3770 ± ± 1534 5 598

100 377 3907 ± ± 5 987 1534 11 805
200 377 3979 ± ± 11 970 1534 17 860
a
Mean values of years 2, 3 and 4.

no difference was found owing to irrigation, and with ef®ciency. Maximum energy ef®ciency of over
200 kg N haÿ1 the ef®ciency was greater in the irri- 80 J Jÿ1 was recorded in the rainfed 2-, 3-, and 4-
gated crop. Thus, when nitrogen was not applied, it year-old crops without N fertilization.
was better not to irrigate, while with a high N level it If the analysis is carried out supposing a crop
was better to irrigate, since irrigation improved energy cycle length of 15 years, the costs of crop establish-
ment will be distributed in a longer period and their
incidence will decrease; likewise, the lower yield in
the establishment year is distributed in a longer period
and its in¯uence decreases. Consequently, an annual
maximum net energy yield of about 600 GJ haÿ1 and
an annual maximum energy ef®ciency of over 150
J Jÿ1 are expected, with irrigation and the highest
N level and without irrigation and N fertilization,
respectively.

4. Conclusions

After an establishing period of about 1 year, the

cultivation of Mscanthus resulted in a dry matter
production of over 37 t haÿ1 yearÿ1 during the next
4 years when the crop was well irrigated and fertilized.
In the absence of N fertilization, irrigation did not
Fig. 3. Net energy yield (top) and efficiency of energy production
modify biomass yield, the effect of irrigation
(bottom) of Miscanthus cultivation in response to nitrogen increased with the increase in N level. The effect
fertilization. Values are averaged over 4 years. can be estimated as ‡3.7 t haÿ1 with 100 kg N haÿ1
10 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11

and ‡9.8 t haÿ1 with 200 kg N haÿ1. Increase in Acknowledgements

nitrogen level also increased N concentration of bio-
mass, from 4.8 g kgÿ1 at the zero N level to 7.5 g kgÿ1 This research was supported by the Italian Ministry
at the 200 kg N level. However, even the highest N of Food Agriculture and Forestry Resources, Project
concentration is less than the critical limit of PRisCa. The work is to be attributed to all four authors
10 g kgÿ1 N for minimum N oxides emissions during in equal parts.
combustion of biomass fuels (Lewandowski and
Kicherer, 1997).
Water content of Miscanthus biomass was about References
53% in autumn, considerably higher than the max-
imum limit of 23% required to ensure good storability Batty, J.C., Keller, J., 1980. Energy requirements for irrigation. In:
of biomass (Lewandowski and Kicherer, 1997). A Pimentel, D. (Ed.), Handbook of Energy Utilization in
consequence of the high water content is that the Agriculture. CRC Press, Boca Raton, Florida, pp. 35±54.
Beale, C.V., Long, S.P., 1997. Seasonal dynamics of nutrient
harvested plant material must be dried. The water
accumulation and partitioning in the perennial C4-grass
content of Miscanthus at harvest depends on the Miscanthus  giganteus and Spartina cynosuroides. Biomass
harvesting season. In most countries, Miscanthus is Bioenergy 12, 419±428.
harvested in spring when the dead, overwintered Christian, D.G., Poulton, P.R., Riche, A.B., Yates, N.E., 1997. The
vegetative matter is at its driest (15±25%) and can recovery of 15N-labelled fertilizer applied to Mis-
be stored with less risk of fungal attack. On the other canthus  giganteus. Biomass Bioenergy 12, 21±24.
Doering, O.C. III, 1980. Accounting for energy in farm machinery
hand, autumn harvest of Miscanthus has environmen- and buildings. In: Pimentel, D. (Ed.), Handbook of Energy
tal bene®ts, such as less soil compaction, since the soil Utilization in Agriculture. CRC Press, Boca Raton, Florida, pp.
is wet at the end of winter and use of heavy machinery 9±24.
may damage the soil. Heichel, G.H., 1980. Assessing the fossil energy costs of
The calori®c value of Miscanthus biomass was propagating agricultural crops. In: Pimentel, D. (Ed.), Hand-
book of Energy Utilization in Agriculture. CRC Press, Boca
about 16.5 MJ kgÿ1 and was not affected by irrigation Raton, Florida, pp. 27±33.
and N fertilization. Therefore, energy production Himken, N., Lammel, J., Neukirchen, D., Czypionka-Krause, U.,
depended exclusively on biomass yield. Maximum Olfs, H.-W., 1997. Cultivation of Miscanthus under West
energy yield was 564 GJ haÿ1, obtained after the European conditions: Seasonal changes in dry matter produc-
tion, nutrient uptake and remobilization. Plant Soil 189, 117±
establishing year with 200 kg N haÿ1 and irrigation.
126.
On the contrary, without N supply and irrigation, Hotz, A., 1996. Screening of different Miscanthus cultivars in
energy yield was 291 GJ haÿ1. Net energy yield, respect of their productivity and usability as a raw material for
calculated as the difference between energy output energy and industry. In: Chartier, P., Ferrero, G.L., Henius,
and energy input per hectare, was 543 GJ haÿ1 in the U.M., Wiinblad, M. (Eds.), 9th European Bioenergy Conf.
former case the 284 GJ haÿ1 in the latter, assuming an Proc., 24±27 June 1996, Copenhagen, Denmark. Elsevier
Science, Oxford, UK, 2294 pp.
energetic input of 20.6 and 6.5 GJ haÿ1 respectively, Knutson, G.D., Curley, R.G., Roberts, E.B., Hagan, R.M.,
averaged over years of cultivation. The ratio of energy Cervinka, V., 1977. Pumping energy requirements for irrigation
output to energy input in crop production was 22±47, in California. Special Publication 3215. Division of Agricultur-
the range depending on whether irrigation or N ferti- al Science, University of California, Oakland.
lization were applied or not. These values are very Lewandowski, I., Kicherer, A., 1997. Combustion quality of
biomass: practical relevance and experiments to modify the
much higher than those determined on Miscanthus by biomass quality of Mascanthus  giganteus. Eur. J. Agron. 6,
Lewandowski et al. (1995). The difference probably 163±177.
depends on the fact that in our research Miscanthus Lewandowski, I., Kicherer, A., Vonier, P., 1995. CO2-balance for
was harvested when greatest biomass was attained. At the cultivation and combustion of Miscanthus. Biomass
this stage, the water content was at its highest value, Bioenergy 8, 81±90.
Lockeretz, W., 1980. Energy inputs for nitrogen, phosphorus, and
but the energy requirements for drying of the plant potash fertilizers. In: Pimentel, D. (Ed.), Handbook of Energy
material prior to combustion were not included in our Utilization in Agriculture. CRC Press, Boca Raton, Florida, pp.
analysis. 23±26.
 L. Ercoli et al. / Field Crops Research 63 (1999) 3±11 11

Long, S.P., 1983. C4 photosynthesis at low temperatures. Plant Cell Snedecor, G.W., Cochran, W.G., 1980. Statistical Methods, 7th ed.
Environ. 6, 345±363. Iowa State University Press, Ames.
Petrini, C., Bazzocchi, R., Bonari, E., Ercoli, L., Masoni, A., 1996. Tarantino, E., Onofri, M., 1988. Determinazione dei coefficienti
Effect of irrigation and nitrogen supply on biomass production colturali mediante lisimetri. Symposium Irrigazione e Ricerca
from Miscanthus in Northern-Central Italy. Agric. Med. 126, Proc., Bologna, Italy, 1988. Cartotecniche Meridionali, Foggia,
275±284. pp. 119±136.
Schwarz, H., 1993. Miscanthus sinensis `Giganteus' production on van der Werf, H.M.G., Meijer, W.J.M., Mathijssen, E.W.J.M.,
several sites in Austria. Biomass Bioenergy 5, 413±419. Darwinkel, A., 1993. Potential dry matter production of
Schwarz, H., Liebhard, P., Ehrendorfer, K., Ruckenmauer, P., Miscanthus sinensis in the Netherlands. Industrial Crops
1994. The effect of fertilization on yield and quality of Products 1, 203±210.
Miscanthus sinensis `Giganteus'. Industrial Crops Products 2, Venendaal, R., Jorgensen, U., Foster, C.A., 1997. European energy
153±159. crops: a synthesis. Biomass Bioenergy 13, 147±185.