Revista Brasileira de Agrometeorologia, Santa Maria, v. 11, n. 1, p. 1-6,2003Recebido para publicação em 27/05/2002. Aprovado em 20/12/2002.

AMT-82>t-P~AFt. 2<:)0--::'~3/5/2Co~

, ,

ISSN 0104-1347

Solar radiation interception and its relation withtranspiration in different coffee canopy layers

Interceptação da radiação solar e sua relação com a transpiraçãoem diferentes extratos foliares de um cafeeiro

Fabio Ricardo Marin':", Alailson Venceslau Santiago':", Evandro Zanini Righi'r', Paulo César Sentelhas':",Luiz Roberto Angelocci'é, Selma Regina Maggiotto'v e José Ricardo Macedo Pezzopane'v

Abstract - Solar radiation distribution within a coffee plant and its effects on the transpiration rate at differentplant layers were studied. Net radiation (Rn), photosynthetically active radiation (PAR) and sap flow (SF)were measured at three heights (0.2, 1.10, and 1.55 m above the ground) within a coffee tree during twentydays. 1t was observed that about 60% ofthe Rn and PAR incident on the plant were extinguished along the firstlayer. Extinction coefficient for Rn and PAR presented a trend of reduction through the planto The relationbetween Rn, expressed per leaf area unit, and SF indicated a significant difference among plant layers. However,analysing the transpiraton rate per leaf area unit, there was a similarity between the plant layers, independentof their exposure to solar radiation. The different patterns of interaction between plant layers and radiantenergy caused different ratios between absorbed energy and water lost, resulting that the intermediate andlowest layers had the highest efficiency in using the energy for transpiration.

Key words: sap flow. net radiation, photosynthetically active radiation, coffea arabica

Resumo - A distribuição da radiação solar em um cafeeiro e seus efeitos sobre a taxa de transpiração dediferentes camadas foliares foram estudados. Para tanto, mediu-se o saldo de radiação (Rn), a radiaçãofotossinteticamente ativa (PAR) e o fluxo de seiva (SF) em três alturas (0,2; 1,10 e 1,55m acima do solo) nacopa de um cafeeiro adulto durante vinte dias. Observou-se que cerca de 60% de Rn e PAR que atingem aplanta é extinta ao longo do primeiro extrato. O coeficiente de extinção apresentou uma expressiva tendênciade redução ao longo da planta. A relação entre Rn, expressa em unidade de área fo lia r, e SF indicaram umaexpressiva diferença entre os extratos foliares, havendo, entretanto, similaridade entre os extratos quando seavaliou a taxa de transpiração por unidade de área foliar. Os diferentes padrões de interação dos extratosfoliares com a energia radiante resultaram em diferenças significativas na razão entre energia absorvida eperda d'água, sendo que os extratos intermediário e inferior tiveram maior eficiência no uso da energia paratranspiração.

Palavras-chave: fluxo de seiva, saldo de radiação, radiação fotossinteticamente ativa, coffea arabica.

Introduction

Plant transpiration and photosynthesis ratesare mainly determined by the interception of solarradiation by the canopy, and several studies were

carried out to characterize the radiation as adeterminant factor of agricultural yield (AUBERTIN& PETER, 1961; YAO & SHAW, 1964;PENDLETON et al., 1966; SAKAMOTO & SHAW,

'Departamento de Ciências Exatas, Setor de Agrometeorologia, ESALQ, Universidade de São Paulo, Av. Pádua Dias, 11, Caixa Postal09, 13418-900, Piracicaba, SP, Brasil.

2Sponsored by FAPESP3Sponsored by CAPES'Corresponding Author: farmarin&esalq.usp.brsSponsored by CNPq6Instituto Tecnológico SIMEPAR/LEMMA - Curitiba - PR

2

used, representing the average condition of theorchard, and it was about 1.8 m high, with a maximumcanopy diameter of about 1.5 m (Figure 1). This plantreceived an extra amount of water (about 5 mm perday) to ensure its transpiration was at a maximumrate.

MARIN, F.R. et aI. - Solar radiation interception and its relation with transpiration ...

1967), or just to characterize the solar radiation regimeabove and below the vegetation (PEREIRA et aI.,1982; MACHADO et aI., 1985).

These kinds of studies are more difficult forsparse crops, due the different types of geometry oforchards (spacing and arrangement of the trees) andalso due to the' canopy geometry. The few studiesrealized for sparse crops had focused only thequantitative aspect of the radiation, without theestablishment of ecophysiological relationships(TURREL & AUSTIN, 1965; GREENE & GERBER,1967; PALMER, 1977). NUTMAN (1937, 1941),tried to establish a relation between radiant energyand transpiration rate and also with stomatal cIosureof young coffee plants cultivated in vases andsubmitted to different radiative conditions, findingsome relationships that could be applied to understandthe performance of young coffee plants submitted todifferent amounts of radiation. However, for adultcoffee plants in field conditions such relations werestill not studied.

Therefore, due the few studies approachingthis aspect in adult coffee plants, this work had theobjective of quantifying the radiation intercepted inthree layers of an adult coffee tree in a high densitysystem, evaluating the variation of the extinctioncoefficient and available radiant energy (Rner), andestablishing the relationship between Rner andtranspiration rate in each layer and also for the wholecanopy.

Material and Methods

The experiment was carried out in a non shadedirrigated coffee plantation (Coffea arabica L.), culti-var Mundo Novo Apuatã, at the Agricultural College"Luiz de Queiroz", University of São Paulo,Piracicaba, Brazil (Lat. 22°43' S, Long. 47°30' W,560 m a.m.s.I.). Trees were four years old, spaced2.5m x 1.0m (the direction of the hedgerows was N-S) and the total area was 2750 m". Managementpractices were done regularly, such as weed control,fertilizer application and phytosanitary control. Thesoil is classified as an Alfisol ("Terra RoxaEstruturada", series Luiz de Queiroz).

The experiment was performed from 17 Aprilto 24 May, 2001, being thirty-three days used for so-lar radiation analysis, and twenty days for sap flowstudies, as described below. One healthy plant was

Netradiation (Rn) and photosynthetically activeradiation (PAR) were measured at four levels abovecanopy, at 2.20 m above the ground, and within onetree, at 1.55, 1.10 and 0.20 m above the ground (Figure1), using net radiometers (Q*7.1, REBS Inc.) andsilicon photovoltaic detectors (LII90SB QuantumSensor, Li-Cor, Inc.). The sensors height representedthe following foliage layers: LI' the top layer, above1.55 m; L2' between 1.10 and 1.55 m; and L3' between1.10 m and 0.20 m. At each every three days, theposition of the Rn and PAR sensors in relation to thecompass directions was altered to minimize thedifferences in amount of radiation between east andwest faces of hedgerow.

To calculate the extinction coefficient (k) ofeach plant layer and for the whole plant, hourlyaverages of Rn and PAR were used when solarelevation was greater than 20° (PEREIRA et aI.,

----------~.. ~OAOm area=I.30 m2

Layer 1

OA5 rn Layer 2SGB25

0.90m Layer 3

SGB50

p-nWm: :

---_._~~

I •• area_I.80m2 ~ I

c!ib. PAR scnsor-+Rn~cnsor

Figure 1. Schematic representation of the coffeeplant studied and sensors location: sapflow sensors (SGB50, SGB25 e SGA13),Rn and PAR sensors, at positions 1, 2 and3.

"

Rev. Bras. Agrometeorologia, v. 11, n. 1, p. 1-6, 2003

3

1982). For the LI (the highest layer), k was calculatedusing the measures taken above the layer, i.e., abovethe canopy, and below the layer. The same procedurewas used for L2and L3' and for the whole plant, whenthe measures at 2.20m and 0.20m were used. Thefollowing expression was applied to calculate k:

(1)

where k is the extinction coefficient of the three layers,or the whole tree (dimensionless), Rb is the radiationmeasured below the layer, or below the tree (W rrr'),Rn is the radiation measured above the layer, or abovethe tree (W rrr"), and IAl is the leaf area index (rrf rrr')of the layer or plant. Radiation measurements usedin eq. 1 were either Rn or PAR, and therefore k wascaJculated for both Rn and PAR. Although k for Rn isnot usually calculated, it was used here to obtain arelation between transpiration and available radiantenergy, and as certain sirnilarity between k for netand incorning solar radiation may be expected, suchvalues become useful.

To estimate the energy available at each layerper leaf area unit, the net radiation was expressed interms of leaf area and then was called effective netradiation, calculated by:

Rn ; -RnbRn .r = --"'---=--ej LAI

where Rnef is the effective net radiation (MJ m\afday'),

The leaf area was deterrnined by counting all\eaves (N) of each layer and measuring the width (i,m) and length (L, m) of 15% of them. The followingequation was used to calculate the total leaf area ofeach layer:

LA=iLN F

where IA is the leaf area of the layer (m") and F is acoefficient that converts the product between the in-dividuallongest length (L) and width (i) to individu-al leaf area (LA). F was obtained from a linearregression analysis between the IA{ measured in anarea meter (LI31 00 Area Meter, Li -Cor Inc.) and theproduct L x i of 50 leaves randomly selected from thesame plant where PAR and Rn measurements were

done. Was obtained F = 0.703 (R2 = 0.99). For thewhole plant, IA was the sum of areas of each layer.The LAl of each layer was calculated using theprojected area of its base (Table 1).

Assuming the sap flow integrated for 24 h as agood approximation of the daily transpiration (7) ateach plant layer (VALANCOGNE & NARS, 1993;TREJO-CHANDIA et aI., 1997), the sap flow (SF)was measured using commercial sensors (SGB50,SGB25 and SBAI3, Dynamax Inc.), installed at1.55m, 1.10m and 0.20m above the ground (Figure1). The power supply dissipated ranged from 0.2 to2 W, as a continuous current, depending on the sensorspecification. Sap flow values were calculatedaccording to the general heat balance equation(SAKURATANI, 1981; BAKER & VAN BAVEL,1987), following the procedure proposed byVALANCOGNE & NASR (1993), and accumulatedfor 24 h, as follow:

SF = P-qa -qrcp dt (4)

(2)

where SF is the sap flow (g S·I),P is the power appliedto the sensor (W), q is the heat flux conductedavertically in the trunk (W), q, is the heat fluxconducted radially in the trunk (W), cp is the specificheat of water (4.186 J g' °C·I), and dt is the saptemperature increment along the sensor COC).

Four sheets of alurninized paper were used forthermal insulation and a flexible silicon layer was usedbetween the thermal insulation and the trunk torninirnize moisture effects on the measurements. Theelectrical signals from the sensors were sampled every10 s, and averaged every 15 min and stored in adatalogger (CRI0X, Campbell Scientific Inc.)connected with a multiplexer board (AM416,Campbell Scientific Inc.). Data were analysed usingdescriptive statistics and linear regression.

(3)

Table 1. Leaf area (LA) and leaf area index (LAI) foreach layer and for the whole coffee plant.

1.535.198.6515.37

1.152.944.895.29

4

between SF and Rn, indicated a difference amongfoliage layers, ranged from 0.09 L.MJ·I in LI, to 5.66L.MJ-I in L3 (Figure 3). This large difference showedthe importance of the upper layer in keeping the wholeplant water balance, by absorbing most of the solarradiation, allowing the others foliage layers keep amore efficient synthesis of carbohidrate along the day,as the plant water conditions are more adequate forstomatal opening (NUTMAN, 1937; CANNEL,1985).

MARIN, F.R. et a!. - Solar radiation interception and its relation with transpiration ...

Results and discussion

Solar radiation measured above the plant (at2.20 m above the ground) indicated c1ear skies duringthe measurements period with Rn values at 12:00reached 400 W m? and PAR reached values as highas 300 W m', 'Rn during the experiment was around10 MJ m' day', while PAR reached an average of7.1 MJ m? day'. About 60% of the Rn and PARreaching the plant were extinct in LI (Figure 2), inspite of larger variability of the measures at 1.55mwhen compared to others heights, what was probablydue the little leaf area of LI leading to direct sunexposure of the net radiometer sensor.

The extinction coefficient at each layerpresented large variability within coffee plant (TabIe2), witch occurred mainly in the first layer due toproblems with Rn measurements mentioned above.The k values for Rn decreased from the top to thebottom of the canopy while for PAR was observed anirregular variation of k through the canopy. For wholeplant, Rn showed a lower value of k than PAR, sinceRn remaining close to 10 W m? below the canopywhile PAR tended to zero in this position, probablybecause the low extinction of infrared short wavecomponent of Rn by foliage (ROSS, 1975; PEREI-RA et aI., 1982).

By relating Rnef and SF is possible to makeinferences about the amount of radiant energy neededby the plant to transpirate 1 liter of water, and also tounderstand the influence of radiation on the waterbalance of the coffee plant and on the conditions forstomatal regulation of gas exchange. The ratio

A 1-+-2,20 -8-1,55 -f.I-1,10 ~0,201500

400N 300'E

~ 200

100

9:00 10:0011:00 12:00 13:00 14:0015:00 16:00

Local Time

The ratio between SF and Rn'f for the wholecanopy was similar to L2 (Figure 4), with an averagevalue of 0.47 L.MJ-I. Although this SFlRn

efvalue is

also dependent on others environmental variables,linear regression analisys between SF and Rnef hadan R2 = 0.45. This R2 value indicates that waterconsumption of coffee plants can be estimated usingmeteorological variables, if the plant leaf area isknown.

The transpiration rate in L3 represented 60%of the total amount transpired by the plant (Figure5A). The slope of the regression lines on Figure 5Aagreed very well with the ratio of the leaf area at eachlayer to the total leaf area (Table 1), indicating thatthere was a similarity among layers transpiration rate,independent of their exposure to solar radiation (Fi-gure 5B).

On leaf area basis, the transpiration rate of LItended to be larger than transpiration rate of L2 andL3' but alllayers presented a high variability, makingit difficult to find differences among them (Figure5B). The limited amount of radiant energy and thepresumable similarity of air temperature and humidity

B -+-2,20 -e-1,55 -f.I-1,10 ~0,20400~========================='300

N

'~ 200~

100

9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00

Local Time

Figure 2. Average day-time fluctuation of Rn (A), and PAR (B), measured above the canopy (2.2 m above theground), and within the coffee canopy, at 1.55 m (bottom of L), at 1.10 m (bottom of Lz) and at 0.20 m(bottom of L3)'

Rev. Bras. Agrometeorologia, v. 11, n. 1, p. 1-6,2003

5

Table 2. Average daily extinction coefficient (k) ofcoffee for net radiation (Rn) andphotosyntetically active radiation (PAR), ateach plant layer, and for the whole cofeeplant. Number in brackets are standarddeviation of the mean.

Rn PAR

LI 0.92 [0.74] 0.79 [0.26]L2 0.44 [0.28] 0.82 [0.15]L, 0.11 [0.03] 0.32 [0.11]

Plant 0.55 [0.08] 0.93 [0.12]

conditions may explain the similiar transpiration ratesof Lj and L3' However, in spite ofthe large amount ofradiant energy absorbed by LI' its transpiration ratewas close to the transpiration of the others layers,and this may be a consequence of stomatal closure.

Similar results were presented by NUTMAN(1941), indicating the plant ability of keeping theirstomatal open, even at low radiation levels, and reducethe water vapour lost when the conditions induce hightranspiration rates. It was observed that, on theaverage, the sum of RnefatLz and L3represented 15%ofthe available energy at LI' while transpiration ratesat these layers were about 80% of the one at LI'

When alI data were pooled, the regression linehad a slope close to unity (1.04), differently from thevalues obtained when each layer were analysedseparately (1.24 for LI' 0.72 for r., and 1.06 for L3)'Linear regression analysis indicated that there is nostatistical significance (10% leveI) when comparinglayers versus canopy transpiration, which lays

1.0

0.8

Cii~ 0.6"!E;

'C? 0.4=I-

0.2

0.0

0.0

I O L1 O L2 f> L31

2.0 4.0 6.0

Figure 3. Relationship between coffee tree transpiration(D and effective net radiation (Rn,) at eachlayer of coffee plant.

emphasis in the inference about stomatal regulationof water vapour loss.

Conclusions

Coffee tree layers have different availableamounts of PAR and Rn, which lead the upper layerof leaves having the largest values of k and availableenergy. However, transpiration rates are not statisticalydifferent among layers, indicating that the upper layerhas the highest resistance to vapour transfer toatmosphere, while the other two leaf layers are moreefficient in using energy for transpiration.

Acknowledgements

The authors are grateful to Dr. Nereu AugustoStreck for the valuable suggestions on the manuscriptand for correcting English text.

References

AUBERTIN, F.M.; PETER, D.B. Net radiationdetermination in a com field. Agronorny Journal,Madison, v. 53, p. 269-272, 1961.

BAKER, J.M.; VAN BAVEL, C.H.M. Measurements ofmass flow of water in stems of herbaceous plants.Plant, Cell and Environrnent, Oxford, v. 10, n. 6,p.777-782, 1987.

CANNEL, M.G.R. Physiology of coffee crop. In:CLIFFORD, N.M.; WILSON, K.C. Coffee: Botany,

0.6

oO

DOO

DO

Oc~ 0.4

'O:::'- 0.2I-

y = 0.47 46x

R2 = 0.4468

0.0 +----,-----,------,-----,---,0.0 0.2 0.4 0.6 0.8 1.0

Figure 4. Relationship between transpiration (D andeffective net radiation (Rnef) for the wholecoffee canopy.

6

B0.8

IO L1 O L2 t:. L31O

c 0.6-ªl t:.~ t:.

E;;, 0.4dQ;>-.s~ 02

O

MARIN, F.R. et aI. - Solar radiation interception and its relation with transpiration ...

A8

I o L1 O L2 t:. L31

6

0.0 +-----,------,-------r------,0.0 0.2 0.60.4 0.8

Figure 5. Relationship between the coffee canopy transpiration (Taump) and the transpiration at three layers of thecoffee canopy (Tia,.,): A. transpiration on a volume basis; B. transpiration on a leaf area basis.

biochemistry and production beans and beverage,London: Croom Helm, 1985. p. 108-134

GREENE, B.A.; GERBER, J.F Radiant energydistribution in citrus trees. Proceedings of theAmerican Society for Horticultural Science, Geneva,v. 90, p. 77-85, 1967.

MACHADO, E.C. et aI. Relações radiométricas de umacultura de cana-de-açúcar. Bragantia, Campinas, v. 44,n. 1, p. 229-238, 1985.

NUTMAN, FJ. Studies on the physiology of Coffeaarabica L. 11.Stomatal movements in relation to naturalconditions. Annals of Botany, Oxford, v. 1, n. 4,p. 681-693, 1937.

NUTMAN, FJ. Studies on the physiology of Coffeaarabica L. III. Transpiration rates of whole trees inrelation to natural environmental conditions. Annals ofBotany, Oxford, v. 5, n. 17, p. 59-81, 1941.

PALMER, J. Diurnal light interception and computermodel of light interception by hedgerow-appleorchards. Journal of Applied Ecology, Oxford, v. 14,p.601-614, 1977.

PENDLETON, r.w., PETERS, D.B.; PEEK, W. Role ofreflected light in the com ecosystem. AgronomyJournal, Madison, v. 59, p. 395-397. 1966.

PEREIRA, A.R.; MACHADO, E.C.; CAMARGO,M.P.B. Solar radiation regime in three cassava (Manihotsculenta CRANTZ) canopies. Agricultural and ForestMeteorology, Amsterdam, v. 26, n. 1, p. 1-10, 1982.

ROSS, J. Radiative transfer in plant communities. In:MONTEITH, J.L. (ed): Vegetation and atmosphere,v.l, London: Academic Press, 1975. p. 13-52.

SAKAMOTO, C.M.; SHAW, R.H. Apparentphotosynthesis in field soybean canopies. AgronomyJournal, Madison, v. 59, p. 73-75. 1967.

SAKURATANI, T. A heat balance method formeasuring water sap flow in the stem of intact plant.Journal of Agricultural Meteorology, Tokyo, v. 39,n. 1, p. 9-17, 1981.

TREJO-CHANDIA, J.E.; ANGELOCCI, L.R.;OLIVEIRA, R.F Aplicação do método do balanço decalor na determinação da transpiração de mudas delimoeiro. Scientia Agricola, Piracicaba, v. 54, n. 3,p.221-231, 1997.

TURREL, FM.; AUSTIN, S.W. Comparative nocturnalthermal budget of large and small trees. Ecology,Brooklyn, v. 46, p. 25-34, 1965.

VALANCOGNE, C.; NASR, Z. A heat balance methodfor measuring sap flow in small trees. In: BORGHETTI,M.; GRACE, J.; RASCHI, A. (eds.) Water transport inplants under cIimatic stress, Cambridge: CambridgeUniversity Press, 1993. p. 166-173.

YAO, A.Y.N.; SHAW, R.H. Effect of plant populationand planting pattern of corn on the distribution of netradiation. Agronomy Journal, Madison, v. 56, p. 165-169, 1964.