Water requirements of poplar and willow vegetation filters grown in lysimeter under Mediterranean conditions: Results of the second rotation

DESALINATION

www.elsevier.com/locate/desalDesalination 247 (2009) 138–147

Water requirements of poplar and willow vegetationfilters grown in lysimeter under Mediterranean conditions:

Results of the second rotation

Chiara Pistocchi*, Werther Guidi, Emiliano Piccioni, Enrico Bonari

Land Lab – Scuola Superiore S. Anna Via S. Cecilia, 3, 56127 – Pisa, ItalyTel: +39 050883521; Fax: +39 050883 12; email: [email protected]

Received 18 September 2007; revised 15 January 2008; accepted 10 March 2008

Abstract

Poplar and willow short-rotation coppice have been shown to be suitable to improve wastewater quality invegetation filters. Nevertheless, water requirements of these species are still uncertain. The aim of this workwas to estimate crop evapotranspiration (ETc) and crop coefficient (kc) of these species grown under high fertil-ised (HF) and low fertilised (LF) conditions through the water balance of lysimeters. This paper deals with theresults of the first growing season after coppicing of the second rotation (2006) in comparison with those ofthe first season of the first rotation (2004). During the 2006 growing season (around 132 days), ETc was319 mm (LF) and 719 mm (HF) in poplar and 607 mm (LF) and 919 mm (HF) in willow, with maximum dailyvalues of 11.55 mm in HF poplar and 15.09 mm in HF willow. Ten-day ETc rates were in general higher thanthose found in the 2004 growing season. Maximum kc, reached in August and September, ranged between 1.2and 2.9 in LF treatment and between 2.5 and 4.0 in HF ones, and were higher than maximum kc values reachedduring the 2004 growing season. In both growing seasons evapotranspiration and kc seemed to be related more tothe availability of nutrients than to differences between the two species.

Keywords: Poplar; Willow; Short-rotation coppices; Evapotranspiration; Crop coefficient

1. Introduction

Short-rotation coppices (SRC) are high-densityplantations of woody crops which are grown toproduce biomass as an alternative to fossil fuelsfor renewable energy production [1,2]. During

the last decade SRC have also been shown to besuitable to improve wastewater quality in vegeta-tion filters which are SRC plantations irrigatedwith wastewater, usually as a complement toconventional treatments. Poplar and willow arethe most common species used for this purpose.In fact, the transpiration rate of these speciesis often higher than that of other crops [3], and*Corresponding author.

Presented at Multi Functions of Wetland Systems, International Conference of Multiple Roles of Wetlands, June 26–29, 2007, Legnario (Padova) Italy

0011-9164/08/$– See front matter � 2009 Elsevier B.V. All rights reserved

is more similar to that observed in wetland species[4]. For this reason, there is a considerable interestin using willow and poplar to treat high volumesof liquid wastes (e.g., municipal wastewater, land-fill leachate, agricultural drainage or stormwater)[5–8].

Pre-treated wastewaters usually contain alarge amount of nitrogen and phosphoruswhich may meet plant requirements, enhancingbiomass production without the need for addi-tional fertilisation. At the same time, the plant–soil system, which uptakes these elements,may improve the wastewater quality before itsrelease into water bodies. Moreover, since lackof water may limit plant growth [9], the extrawater may further enhance biomass production.The cost efficiency of willow and poplar vegeta-tion filters integrated with biomass production isreported to be higher than that of conventionaland, in many cases, unconventional tertiarywastewater treatments [10]. Furthermore, sincethese species are non-edible crops, possiblehygienic risks [11], deriving from applicationof wastewater, are reduced [12]. Despite theseseveral benefits, vegetation filters have so farbeen implemented to a limited degree andmostly in northern European countries.

However, to optimise wastewater reuse in veg-etation filters, from an economical (minimisingcosts) and environmental perspective (minimisingnutrients leaching), it is important to predict withaccuracy water requirements of these species.

Although previous studies have focused onvegetation filter efficiency [6,13], few of themhave investigated their water consumption duringthe whole growing season under well-wateredconditions and optimum nutrient supply. In addi-tion, most recent studies were conducted innorthern countries and there is no informationavailable about water requirements of SRCunder Mediterranean conditions [3,14]. Manystudies have investigated poplar or willow standtranspiration using either sap-flow methods[3,15,16] or a micro-meteorological approach

[9,14], whereas only few have been conductedin lysimeters [4,8]. The lysimeter method iseasy to replicate and allows an accurate measure-ment of evapotranspiration from the whole plantunder the most favourable conditions [17,18].Moreover, since vegetation filters are usually pro-vided with high volumes of nutrient-rich water,they are almost never affected by drought orshortage of nutrients. Thus, lysimeters seem tobe the most appropriate approach to study thewater requirements of this cropping system. Thelysimeter method also enables the estimation ofthe crop coefficient (kc), which is the ratiobetween crop evapotranspiration (ETc) and refer-ence evapotranspiration (ETo) [19]. ETc is theevapotranspiration from a specific, disease-free,well-fertilised crop, grown in large field, underoptimum soil and water conditions, and achievingfull production under the given climatic condi-tions. ETo refers to the evapotranspiration froma reference surface, is a climatic parameterexpressing the evaporation power of the atmos-phere and can be determined from meteorologicaldata. As a consequence, if kc for a specific crop isknown, it is possible to estimate its evapotranspi-ration rates by using the equation ETc = ETo kc. Inthe case of a vegetation filter ETc also representsthe maximum amount of wastewater that can besupplied to the plantation reducing environmentalrisks of pollution from nutrient leaching to thegroundwater.

Therefore, the aim of this work was to esti-mate ETc and Kc of willow and poplar SRC.We present here the results of the first year ofthe second rotation, after coppicing (2006), incomparison with those obtained from the firstyear of the first rotation (2004).

2. Materials and methods

2.1. Lysimeter set-up and plant establishment

The experiment was carried out at San Piero aGrado Pisa – Italy (438N 108E and altitude

C. Pistocchi et al. / Desalination 247 (2009) 138–147 139

3 m a.s.l.). The region’s climate is Mediterraneanwith a mean annual rainfall of 916 mm with twopeaks in spring and autumn. In 2002, twelvedrainage lysimeters were set up at the experimen-tal site. Each lysimeter consisted of a concretebox of 1.70 m depth and 1.03� 1.03 m2 surfaceand was equipped with an automatic drip irriga-tion system to maintain the soil moisture contentaround field capacity (Fig. 1). Each lysimeter wasfilled with topsoil whose characteristics areshown in Table 1.

In spring 2003, six lysimeters, randomlychosen, were planted with cuttings of Populusdeltoides(clone Lux) and the other six with cut-tings of Salix alba (clone SI62-059). Further cut-tings were planted around each lysimeter tosimulate a density of 10,000 plants ha�1 and toavoid the clothesline effect. The experimentallayout is shown in Fig. 2.

We adopted a 2-year rotation. Plants wereharvested at the end of the establishment year(2003), and at the end of the first rotation (2005).

In the first rotation cycle (2004–2005), fertil-ised (F) and unfertilised (NF) conditions werecompared using either a solution of nitrogenand phosphorus (20 mg l�1) or tap water

Irrigation pipeSoil surface

Drainage pit

Geotextile layer

Water tableGravelFloating sensor

Fig. 1. Lysimeter installed at the experimental site.

Table 1

Average lysimeter soil properties

Clay % 12.6Silt % 36.4Sand % 51.0pH 8.23Organic matter % 1.64Total N (ppm) 1.15Total P (Olsen) (ppm) 8.60EC (microS/100 g) 331.0

140 C. Pistocchi et al. / Desalination 247 (2009) 138–147

assigned to three lysimeters per species. Duringthe following rotation (2006–2007) two differentfertilisation levels were compared. Thus, whilethe high fertilised (HF) lysimeters were kept atthe same concentration level as the previousrotation of 20 mg l�1 the unfertilised were irri-gated with a solution of 10 mg l�1 (LF). Thenutritive solution was obtained by mixing a bal-anced fertiliser (FERTER 21.21.21 + MICRO,Agroqualita, N:P:K ratio 21:21:21) with tapwater. Therefore, HF treatment being the samein both cycles, only data from this treatmenthave been compared in this paper. The concentra-tion of 20 mg l�1 of N and P (expressed as P2O5)was chosen because it was considered a well-bal-anced nutrient supply, meeting the plants’ needs,and it was similar to N and P concentrations thattreated wastewater needs to have to be used forirrigation purposes, according to the Italian legis-lation in force when the trial started [20].

2.2. ETc and kc calculation

Irrigation and drainage volumes were meas-ured daily during the growing season. Rainfall

was measured by a pluviometer located in theexperimental field. ETo was estimated by thePenman-Monteith – FAO method [19] using10-day time intervals, with daily meteorologicaldata (comprising air temperature, relativehumidity, wind speed and solar radiation)obtained by an automatic weather station locatedat S. Piero a Grado. Ten-day ETc of both specieswas calculated through the water balance of thelysimeters, as follows:

ETc ¼ IþR�D

where I was the amount of water provided byirrigation, R rainfall occurred and D the amountof water lost by drainage. Ten-day kc for thewhole growing season was determined by calcu-lating the ETc/ETo ratio.

Since at the end of the growing seasonplants were not harvested to estimate biomassproduction stem height and diameter, thenumber of shoots per stump of each plant wasmeasured and stem basal area (SBA) wasdetermined.

Lysimeter

2.50 m

0.4 m

WillowLF

PoplarHF

WillowHF

PoplarHF

PoplarHF

PoplarLF

WillowHF

PoplarLF

WillowLF

WillowLF

PoplarLF

WillowHF

Fig. 2. Experimental layout of the trial: high fertilised (HF) and low fertilised (LF) treatment.


2.3. Statistics

All data were subject to a two-way analysisof variance (ANOVA), according to a com-pletely randomised experimental design, to testthe effects of species and fertilisation treatments,after applying Bartlett’s test for homogeneity ofvariances. P < 0.05 threshold was chosen toidentify significant differences among treat-ments. The statistical analysis was performedby using R software package (R Foundation,http://www.r-project.org).

3. Results

3.1. Meteorological conditions

Monthly temperature and rainfall of the 2006growing season are shown in Fig. 3, togetherwith the 16-year average monthly values for theS. Piero a Grado weather station. During theexperiment (from June to November) total rain-fall was 616 mm, around 100 mm higher than

the average for the same period and this wasmainly due to the abundant rainfall in Augustand September. Maximum and minimum temper-atures were above the average in July and belowin August; maximum temperature was higherthan the average also in the following months.This resulted in a longer growing season, com-pared to the previous; thus, water consumptionwas measured until 20 November. ETo, calcu-lated by Penman-Monteith equation for thestudy period, was approximately 537 mm.

3.2. ETc, crop coefficient and growth parameters

The main results of the trial are shown inTable 2.

The 10-day ETc pattern is represented inFig. 4a. During the first 40 days of measurements,ETc was not significantly different among thetreatments and ranged from average values of2.20 (LF) and 3.36 (HF) mm day�1 in poplar to3.31 (LF) and 4.52 (HF) mm day�1 in willow.

Jun

Rainfall (2006)Tmin (2006)

Average rainfallTmax (2006)

Average TminAverage Tmax

Jul Aug Sep Oct Nov0

50

100

150

200

250

0

15

10

5

20

25

30

35

Rai

nfal

l (m

in)

Tem

pera

ture

(�C

)

Fig. 3. Monthly temperature and rainfall during the trial in 2006 and average monthly values (1990–2006) at San Pieroa Grado, Pisa.


In the middle part of the growing season ETc

increased rapidly, reaching maximum dailyvalues in the first 10 days of August for HFtreatments (11.55 mm in poplar and 15.09 mmin willow) and in the second ones for low fertil-ised (LF) treatments (6.71 mm in poplar and9.78 mm in willow). On 16 August the irriga-tion system accidentally broke down and conse-quentially water consumption could not bemeasured properly until 4 September. Howeverduring the last part of the season, ETc decreasedin all treatments up to daily values of 0.63 (HF)and 0.24 mm (LF) in poplar and 1.32 (HF) and

1.77 mm (LF) in willow. The whole seasonalevapotranspiration (around 132 days) was719 mm (HF) and 319 mm (LF) in poplar and919 mm (HF) and 607 mm (LF) in willow, andits extent was only affected by fertilisationtreatment (Table 2).

Kc followed the same pattern as ETc (Fig. 4b),showing three main stages. At the beginning ofthe measurement period, kc was 0.20 and 0.34,respectively, for low- and high-fertilised poplar,and 0.42 and 0.43 for low- and high-fertilised willow. In the central part of the season,with the exception of the above-mentioned

Table 2

Effect of treatments on poplar and willow seasonal ETc and growth parameters measured at the end of the 2006 growing

season. P-value < 0.05 denotes statistically significant effects

Means Statistical

Poplar Willow P-value

Parameters (LF) (HF) (LF) (HF) Species (Df = 1) Fertilisation (Df = 1) Interaction (Df = 1)

Total ETc (mm) 319 719 607 919 0.0926 0.0268 0.7289Stem height (cm) 197 198 142 194 0.1658 0.2056 0.2221Diameter (mm) 12.2 14.0 7.4 11.1 0.0165 0.066 0.4588Number of shoots 4 7 17 23 0.0002 0.0768 0.5272SBA (cm2) 6.4 13.3 9.4 27.5 0.0043 0.0005 0.0306

02J 3J 1JU 2JU 3JU 1A 2A 3A

Ten-days Ten-days

1S 2S 3S 1O 2O 3O 1N 2N

2

4

6

8

10

12

14

16

ET

c (m

m d

ay �

1 )

0.02J 3J 1JU 2JU 3JU 1A 2A 3A 1S 2S 3S 1O 2O 3O 1N 2N

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0Willow HFWillow LFPoplar HFPoplar LF

Kc

(a) (b)

Fig. 4. (a) Ten-day average ETc rates (+SE) and (b) kc (+SE); dashed rectangles show the irrigation system break-down period.


breakdown period, kc reached a peak for all thetreatments that ranged from 1.23–2.95 in LFlysimeters to 2.54–4.03 in the HF. Afterwards,from the first 10 days of October to the secondhalf of November, kc showed average values of0.40 (LF) and 0.80 (HF) in poplar and 1.50(LF) and 1.70 (HF) in willow.

Plant growth depended mainly on differencesbetween the species including stem diameterand number of shoots per lysimeter, whichwere significantly higher in willow than in pop-lar. SBA was affected by both treatments andinteraction was significant as well.

4. Discussion

Ten-day ETc rates were significantly affectedby fertilisation in the middle part of the summer,whereas later in the growing season, significantdifferences occurred between species when wil-low showed higher ETc rates than poplar. Thiscan possibly be explained because in that periodpoplar was affected by leaf rust (Melampsorasp.) and the leaf area was partially reduced.Ten-day ETc can be compared to the first rota-tion [21,22], although it should be noticed that,between the two rotations, only the data refer-ring to high fertilisation treatment can be takeninto account. In both growing seasons, whethercompared with the unfertilised (2004) or lowfertilised (2006) condition, fertilisation seemedto increase evapotranspiration. In addition,both in 2004 and in 2006 10-day ETc temporalpatterns of both species were similar and maxi-mum ETc rates were registered in the middleand late summer, as reported also by otherauthors [8,14]. This is likely determined by thecombined effect of high evaporation power ofthe atmosphere reached in this period (i.e.,high values of ETo), and increased plant sizeand leaf area, as observed in other studies[8,16,23]. Afterwards ETc rates started todecrease at the beginning of the autumn andceased in late autumn since light and temperature

requirements of the trees were no longer satisfied.Willow transpiration has been measured by otherresearchers using different approaches. For exam-ple, Hall et al. [3], using sap flow method,registered mean daily transpiration rates around6 mm day�1 for Salix burjatica and Populustrichocarpa� deltoides (3-year-old stems on4-year-old stumps) in June and July beforewater stress occurred. This value is comparablewith average ETc rates of the same period inour study. In addition, maximum transpirationrate found by these authors (10.7 mm day�1) forboth species was also quite similar to, or slightlylower than in the case of poplar, the valuesobserved during the corresponding period in ourtrial. However, it has to be pointed out that inthe above-mentioned study the stand was notirrigated and the values reported did not includesoil evaporation which, in turn, was metered inour trial. The higher ETc rates reached at thebeginning of August and September can be com-pared to those reported by Zalesny et al. [16], dur-ing a sampling period of 18 days (25 July to11 August 2003). In this study, average transpira-tion rate of 11.3 mm day�1 was measured on a4-year-old fertilised hybrid poplar stand, whichis very close to the values observed on HF poplarsin August 2006. Moreover, in a lysimeter-basedtrial Pauliukonis and Schneider [4] observed inSalix babylonica evapotranspiration rates whichranged between 10 and 23 mm day�1, during astudy period of 2 months. These data are similarto ours, although some differences occurredmainly because of the larger size of trees reachedin the former trial compared with the currentexperiment.

Seasonal ETc seemed to be significantlyaffected by fertilisation, but not by the species(Table 2). This partially disagrees with theresults of the 2004 growing season when ETc

was significantly affected by both treatments(Table 3). However, 2004 and 2006 seasonalevapotranspiration was related to SBA, whichis an index of biomass production.


Although seasonal ETc of the two species washigher in 2004 than in 2006, this is not in contrastwith the higher 10-day ETc rates found in 2006compared to those found in 2004, consideringthe different length of the measurement season,the different weather conditions (i.e., from Aprilto October total ETo was 745 mm in 2004,while in 2006 it was 671 mm) and the lack of20 days of measurements between August andSeptember. Seasonal ETc of HF willow wasslightly higher than 870 mm observed by Martinand Stephen [8] on Salix viminalis grown inlysimeters the first year after coppice. For poplargrown in lysimeters seasonal ETc were not foundin the literature and hence a comparison with ourdata was not possible.

The Kc values showed the same pattern as10-day ETc values, in both 2004 and 2006 grow-ing seasons, although in 2006 maximum kc val-ues were reached in the second 10 days ofAugust for willow and in the second 10 daysof September for poplar, whereas in 2004 kc ofboth species reached a peak in the first 10 daysof September [21,22]. Growth parameters atthe end of the 2004 growing seasons are shownin Table 3.

Significant differences occurred in 2004between stem diameter of the fertilised and

unfertilised treatments and between the numberof shoots in both poplar and willow. In 2006the number of shoots showed the same pattern,while average stem diameter differed onlybetween the species and not between the fertil-isation treatments.

Stem diameter and height were in generalhigher at the end of the 2004 growing seasoncompared with those reached at the end of2006, whereas SBA was in general higher in2006 than in 2004. This can be explained consid-ering the higher number of shoots produced bythe trees in 2006. In both years SBA valuesdecreased in this order: HF willow > HF poplar> LF/NF willow > LF/NF poplar, though in 2006differences between the fertilisation treatments,as expected, were reduced (69% in 2004 and46% in 2006 for poplar, 74% in 2004 and 66%in 2006 for willow), due to the change in thetreatment. Moreover, SBA was significantlyaffected by both species and fertilisation inboth years, and in 2006, the interaction betweenthe two factors was also significant. This maysuggest that, under optimum water supply condi-tions, fertilisation can differently affect biomassproduction, with a less pronounced influence inpoplar than in willow, as observed also byother authors [24].

Table 3

Effect of treatments on poplar and willow seasonal ETc and growth parameters in the 2004 growing season [21,22].

P-value < 0.05 denotes statistically significant effects

Means Statistical

Poplar Willow P-value

Parameters NF HF NF HF Species (Df = 1) Fertilisation (Df = 1) Interaction (Df = 1)

Total ETc (mm) 590 725 620 1190 0.0044 0.0001 0.0916Stem height (cm) 227 254 149 294 0.6692 0.0810 0.2091Diameter (mm) 15.4 25.7 8.9 25.0 0.3374 0.0059 0.4342Number of shoots 2 3 8 6 0.0000 0.5037 0.1062SBA (cm2) 3.8 12.3 6.3 24.4 0.0112 0.0003 0.0647


4. Conclusions

The trial allowed us to investigate ETc and kc

of poplar and willow SRC under field condi-tions. The results can be useful to predict withmore accuracy the volumes of wastewater tobe provided to a vegetation filter grown underMediterranean climatic conditions, reducingthe risk of nutrients losses to the environmentand the costs for distributing the wastewaters.ETc seemed to be dependent more on availabil-ity of nutrients than on the species. This, theoret-ically, enables both species to be used for suchpurposes and allows a greater choice amongthe clones, to spread these multifunctional bio-energy systems also in Mediterranean regions.

Moreover, it could be interesting to betterunderstand some further aspects, such as wateruse efficiency of the two species and the amountof nutrients retained by the system (i.e. in litter,soil organic matter and plant tissues).

Acknowledgements

We wish to thank Dr. Marcello Bertolacci,from Laboratorio Nazionale di Irrigazione – Uni-versita di Pisa, who hosted parts of the laboratorymeasurements and Mr. Paolo Giovanetti for hisessential technical support. We wish to thankalso Dr. Giorgio Ragaglini and Dr. CristianoTozzini, from Land Lab – Scuola SuperioreSant’Anna, for their useful suggestions duringthe data processing.

References

[1] M. Labrecque and T. Teodorescu, High biomassyields achieved by Salix clones in SRIC followingtwo 3-years coppice rotations on abandonedfarmland in southern Quebec, Canada, BiomassBioenerg, 15(2003), 71–74.

[2] E. Bonari, G. Picchi, A. Fraga, M. Ginanni,W. Guidi and E. Piccioni, Comparison of thethree coppice intervals on a nine years poplarbiomass production, In: Proc. of IPC 22nd Ses-sion: The Contribution of Poplars and Willows to

Sustainable Forestry and Rural Development, 28November–2 December 2004, Santiago, Chile.

[3] R.L. Hall, S.J. Allen, P.T.W. Rosier andR. Hopkins, Transpiration from coppiced poplarand willow measured using sap-flow methods,Agr. Forest Meteor, 90(1998), pp. 275–290.

[4] N. Pauliukonis and R. Schneider, Temporal pat-terns in evapotranspiration from lysimeter withthree common wetland plant species in the easternUnited States, Aquat Bot, 71(2001), pp. 35–46.

[5] S. Elowson, Willow as a vegetation filter for clean-ing of polluted drainage water from agriculturalland, Biomass Bioenerg, 16(1999), 281–290.

[6] P.G. Aronsson and L.F. Bergstrom, Nitrate leach-ing from lysimeter-grown short rotation willowcoppice in relation to N-application, irrigation andsoil type, Biomass Bioenerg, 21(2001), 155–164.

[7] G.R. Alker, A.R. Godley and J.E. Hallett, LandfillLeachate Management Using Short Rotation Cop-pice: Operational Guide, WRc Report numberCO5127, Swindon, 2002.

[8] P.J. Martin and W. Stephens, Willow growth inresponse to nutrients and moisture on a clay land-fill cap soil. II: water use, Bioresource Technol,97(2006), 449–458.

[9] A. Lindroth and A. B�ath, Assessment of regionalwillow coppice yield in Sweden on basis of wateravailability, Forest Ecol Manag, 121(1999), 57–65.

[10] P. B€orjesson and G. Berndes, The prospect forwillow plantations for wastewater treatment inSweden, Biomass Bioenerg, 30(2006), 428–438.

[11] G. Kamizoulis, Setting health based targets forwater reuse (in agriculture), Desalination,218(2008), 154–163.

[12] A. Carlander, P. Aronsson, G. Allestam, T.A.Stenstrom and K. Perttu, Transport and retentionof bacteriophages in two types of willow-croppedlysimeters, J Environ Sci Heal A, 35(2000),1477–1492.

[13] I. Dimitriou and P.Aronsson, Nitrogen leachingfrom short-rotation willow coppice after intensiveirrigation with wastewater, Biomass Bioenerg,26(2004), 433–441.

[14] G. Persson, Willow stand evapotranspiration simu-lated for Swedish soils, Agr Water Manage,28(1995), 271–293.

[15] S.J. Allen, R.L. Hall and P.T.W. Rosier, Transpira-tion by two poplar varieties grown as coppice for bio-mass production, Tree Physiol, 19(1999), 493–501.

[16] R.S. Zalesny, A.H. Wiese, E.O. Bauer and D.E.Riemenschneider, Sapflow of hybrid poplar


(Populus nigra x P. maximowiczii A. Henry‘‘NM6’’) during phytoremediation of landfillleachate, Biomass Bioenerg, 30(2006), 784–793.

[17] A. Aboukhaled, Lysimeters, FAO Irrigation anddrainage paper 39, Rome, 1982.

[18] R.G. Allen, J.H. Prueger and R.W. Hill, Evapo-transpiration from isolated stands of hydrophytes:Cattail and bulrush, Trans. ASAE (Am. Soc.Agric. Eng), 35(4) (1992), 1191–119.

[19] R.G. Allen, L.S. Pereira, D. Raes and M. Smith,Crop Evapotranspiration, Guidelines for Comput-ing Crop Water Requirements, FAO Irrigation anddrainage paper 56, Rome, 1998.

[20] Ministero dell’Ambiente e della Tutela delTerritorio. Decreto 12 giugno 2003, n. 185: regola-mento recante norme tecniche per il riutilizzo delleacque reflue in attuazione dell’articolo 26, comma2, del decreto legislativo 11 maggio 1999, n. 152.G.U. n. 169 del 23-7-2003.

[21] W. Guidi, E. Bonari and M. Bertolacci, Water con-sumption of poplar and willow short rotation forestryused as vegetation filter: Preliminary results. Inte-grated Land and Water Resources Management:Towards Sustainable Rural Development, 15–19May 2005, Frankfurt, Germany- Slubice, Poland.

[22] W. Guidi, E. Piccioni and E. Bonari, Evapotranspi-ration and crop coefficient of poplar and willowshort rotation coppice used as vegetation filter,Bioresource Technol, 99(2008), 4832–4840.

[23] A. Pellis, I. Laureysens and R. Cuelemans, Growthand production of a short rotation coppice cultureof poplar I. Clonal differences in leaf characteris-tics in relation to biomass production, BiomassBioenerg, 27(2004), 9–19.

[24] V. Scholz and R. Ellerbrock, The growth produc-tivity, and environmental impact of the cultivationof energy crops on sandy soil in Germany, BiomassBioenerg, 23(2002), 81–92.


Water requirements of poplar and willow vegetation filters grown in lysimeter under Mediterranean conditions: Results of the second rotation

Documents