ORIGINAL PAPER

Rui M. A. Machado Æ Maria do Rosario G. Oliveira

Tomato root distribution, yield and fruit quality under differentsubsurface drip irrigation regimes and depths

Received: 3 November 2004 / Accepted: 9 May 2005 / Published online: 2 August 2005� Springer-Verlag 2005

Abstract Tomato rooting patterns, yield and fruitquality were evaluated in a field trial where three irri-gation regimes [0.6 (DI), 0.9 (DII) and 1.2 ETc (DIII)]and three drip irrigation depths [surface (R0), subsur-face at 20 cm depth (RI) and subsurface at 40 cmdepth (RII)] were imposed following a split-plotexperimental design, with four replications. Thebehaviour of the root system in response to the irri-gation treatments was evaluated using minirhizotronsinstalled between two plants, near the plant row. Root-length intensity (La)—length of the root per unit ofminirhizotron surface area (cm cm�2)—was measuredat four crop stages. For all sampling dates, none of thefactors studied were found to influence La or rootingdepth significantly or the interaction between treat-ments. For all treatments most of the root system wasconcentrated in the top 40 cm of the soil profile, wherethe root-length density ranged from 0.5 cm cm�3 to1.4 cm cm�3 . The response of tomato fruits to an in-crease in the water applied was similar in quantitativeand qualitative terms for the different drip irrigationdepths. Water applied by drip irrigation had theopposite effect on commercial yield (t ha�1) and solublesolids (�Brix) (r=�0.82, P<0.001), however, yield interms of total soluble solids (t ha�1) was the same forthe 0.9 and 1.2 ETc. The increase in commercial yieldcan be described by the equation ðcommercial yield¼ �91:106ð%ET2

cÞ þ 264:34ð%ETcÞ� 55:973;R2 ¼ 0; 63;P\0:001Þ:

Introduction

Several investigations performed with different horti-cultural crops to analyse the influence of subsurfacedrip irrigation on crop yield show that when crops arefed by subsurface drips yields are equal to or greaterthan those obtained when fed by surface drips (Sam-mis 1980; Hutmacher et al. 1985; Phene et al. 1987;Bar-Yosef et al. 1991; Camp et al. 1993; El-Gindy andEl-Araby 1996; Davis et al. 1997; Ayars et al. 1999;Machado et al. 2003). This behaviour can be attrib-uted to factors affecting evaporation from topsoil(Camp 1998), such as the burying of the irrigationpipe with subsurface drip irrigation, which accordingto Phene (1991) and Phene et al. (1992) conduces thereduction of topsoil evaporation. However, severalother factors have an influence on evaporation fromtopsoil: the water content of the soil surface resultingeither from rainfall or irrigation, the degree of canopydevelopment during the season and the influence ofrainfall and irrigation on root growth and activity(Camp 1998). In processing tomato, during the firststages of crop growth, when the canopy is reduced,subsurface drip irrigation can increase the efficiency ofwater use when compared with surface irrigation(Machado 2002), due to a decrease in crop evapo-transpiration (ETc). However, during the first stages ofcrop development the yield response factor (ky) is low(Doorenbos and Kassam 1986). Similar yields usingsurface and subsurface drip irrigation could indicatethat evaporation is the same from the moment thecanopy effect begins to be felt. As the relationshipbetween water consumption and production shows asigmoidal curve, the hypothesis that there is a differentresponse to the water supplied can be proposed forintervals where the curve slope is higher. The aim ofthe present study is then to determine the effect ofdifferent water regimes supplied by drip irrigation atdifferent depths on tomato root development, distri-bution, yield and quality.

Communicated by A. Kassam

R. M. A. Machado (&) Æ M. do R. G. OliveiraInstituto de Ciencias Agrarias Mediterranicas,Universidade de Evora,Apartado 94, 7002-554,Evora, PortugalE-mail: rmam@uevora.ptFax: +351-266-760823

Irrig Sci (2005) 24: 15–24DOI 10.1007/s00271-005-0002-z

Materials and methods

Experimental site

The experiment was conducted on a Regosol Soil (TypicQuarzipsamments) at the Antonio Teixeira ResearchStation Coruche, Portugal. Soil and water characteris-tics and meteorological data observed during theexperiment are summarised in Tables 1, 2 and Fig. 2,respectively.

Experimental design and treatments

Three irrigation regimes [0.6 (DI), 0.9 (DII) and 1.2 ETc

(DIII)] and three drip irrigation depths [surface (R0),subsurface at 20 cm depth (RI) and subsurface at 40 cmdepth (RII)] were arranged in a split-plot experimentaldesign with four replications. Irrigation regime was de-fined as the primary factor and irrigation depth was thesecondary factor. The size of each plot was 5·7.5 m2,with four rows.

Irrigation was carried out every two days through-out the growing season. The volume of water appliedwas estimated from ETc (minus rainfall), measured twodays before irrigation. When rainfall exceeded the ETc

value, irrigation was suspended and the excess waterwas taken into account when calculating the sub-sequent irrigation volumes. ETc was estimated using thecrop coefficient (Kc) and the Penman Montheithevapotranspiration (ETo) reference data from a nearbyweather station (ETc =KcÆETo). Crop coefficients usedin this study were average values established by Doo-renbos and Kassam (1986) for the following cropstages: 0.75 for the development stage (from trans-plantation to beginning of fruit set); 1.15 for the mid-season stage (from the beginning of fruit set toblooming) and 0.88 for the late-season stage (fromblooming to fruit ripening, when 75% of the fruits werered or orange). To minimise the effect of differentirrigation treatments on the establishment of plantlets,all plants were sprinkler-irrigated at transplantation.Drip irrigation was initiated 11 days after transplanting

and terminated when 75% of fruits were red or orange.The total amount of water applied to the crop for everyirrigation treatment is presented in Table 3.

Soil preparation preceding the installation of irriga-tion tubes, consisted of 40–50 cm deep mouldboardploughing followed by two 10–15 cm disc-harrow

Table 1 Soil physical andchemical characteristics Depth (cm)

0–40 41–74 75–100Sand (%) 92.60 95.20 96.00Silt (%) 1.70 2.00 1.60Clay (%) 5.70 2.80 2.40Bulk density (g cm�3) 1.51 1.60 1.64Organic matter (%) 1.09 0.44 0.32pH (H2O) 5.70 6.20 6.20N (lg g�1) 3.45 2.05 2.18P (lg g�1) 81.84 73.04 43.12K (lg g�1) 69.72 68.06 78.02Ca2+ (cmolc kg

�1) 0.30 0.12 0.11Mg2+ (cmolc kg

�1) 0.18 0.15 0.09

Table 2 Means and standard error of the mean of the variouswater quality parameters

�xa r�xb

pH 6.5 0.09Electrical conductivity (lS cm�1) 347.6 6.09Bicarbonate (CaCO3) (mg l�1) 34.0 2.78Carbonate (CO3) (mg l�1) 0Nitrogen (N-NO3

�) (mg l�1) 74.0 10.1Nitrogen (N-NH4

+) (mg l�1) 0Sodium (Na+) (mg l�1) 37.8 0.58Potassium (K2O) (mg l�1) 24.8 0.50Chloride (Cl�) (mg l�1) 43.5 1.46Calcium (Ca2+) (mg l�1) 18.2 1.49Magnesium (Mg2+) (mg l�1) 13.5 0.15

a Data collected during crop development.b Standard error of the means.

Fig. 1 Soil water retention curves for 0–40 cm (—) and 41–100 cm(- - -) depths

16

operations. Fertilisers were applied (Table 4) beforetransplantation (along a 15 cm band) directly below therow and by fertigation, commencing on the third weekafter transplantation. Ca(NO3)2, KNO3 and H3PO4 wereapplied three times per week via fertigation, in accor-dance with the absorption rates estimated by Phene et al.(1986, 1987). Quantities of nutrients applied which in-cluded not only for soil nutrients but also for waternutrients content, were taken into account (Table 2).Fertiliser concentrations in irrigation water and theinjection rate were calculated to ensure that the electricalconductivity of water (Ecw) never exceeded 2.5 mS cm�1.Pressure-compensated RAM emitters (2.3 l h�1) (Net-afim Inc., Israel) were placed 40 cm apart.

Forty-day-old tomato seedlings, cv. ‘‘H3044’’, weretransplanted at 20 cm within the row and 150 cm be-tween rows, for a total of 33,333 plants ha�1 .

Soil water status

Soil water conditions were monitored using a neutronprobe weekly throughout the season. In the four repli-cations, one polyvinyl chloride (PVC) plastic access tube(with a diameter of 50 mm and a wall thickness of2 mm) per treatment was placed in the row line at adistance of 20 cm from the emitter, and soil water con-tent was quantified to a depth of 100 cm, at 20 cmintervals. Before carrying out counts in access tubes, tenstandard counts were performed. Count ratios werecalculated by dividing each tube count by the mean ofthe standard counts. The ratios were then converted to

soil water content using a calibration curve developedfor the site by Machado (2002). Using a neutron probe,measurements were always taken on days when no irri-gation was carried out.

Root measurement

Root distribution was estimated at four crop stages (atthe beginning of fruit set, at blooming, when 75% offruits were red or orange, and at harvest) using mini-rhizotrons with a length of 1.5 m and a diameter of5.2 cm. One tube per treatment was placed parallel toand 10 cm from the plant row, between two plants, at anangle of 30� to the vertical. At 10 cm intervals along thetube, photographs of roots intersecting the minirhizo-tron wall were taken with a 35 mm camera fitted to anendoscope. Root-length intensity (La) was estimated(Tennant 1975) and converted into root-length densityusing the regression equations defined in the calibrationprocedure described by Machado and Oliveira (2003).For statistical analysis, values for root-length intensity(cm cm�2) were transformed using the equationy ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

ðxþ 1Þp

:

Crop yield and quality

At the four crop stages, three plants from each subplotwere harvested and dried for 2–3 days at 70�C, andweighed for dry mass.

To evaluate yield and quality, all the fruits of plantsgrown in a 7.5 m2 area were hand harvested when �80%of the fruits were red or orange. After categorising theninto ‘‘mature’’, ‘‘green’’, ‘‘rotten’’ and ‘‘blossom-endrot’’ subgroups, the fruits were weighed for commercial-yield evaluation. From the mature fruits subset, a sam-ple of 2.5 kg was taken and passed through a 0.8 mmmesh sieve to separate the seeds and the epidermis fromthe juice. Soluble solids (�Brix) and pH were measured inthe homogenised juice.

Results and discussion

Soil water status

Regarding the overall weekly soil moisture pattern, foreach water application at the different emitter depths(Figs. 3 and 4), it is evident that the greatest differencesoccurred at a depth of 20 cm and especially in the DIIand DIII regimes. It is also evident that in all the situ-ations analysed, a sharp increase in soil moisture oc-curred 76 days after plantation (DAP), probablybecause irrigation was carried out during the afternoon.The normal procedure was to apply water during themorning, so observations using the neutron probe werecarried out sooner after applications.

Table 3 Water applied (irrigation + rainfall) per treatment

Treatments

DI DII DIII

Sprinkler irrigation (mm) 7.0 7.0 7.0Precipitation (mm) 76.1 76.1 76.1Drip irrigation (mm) 243.1 391.4 560.9Total water applied (mm) 326.2 474.5 644.0

Table 4 Crop fertilisation (kg ha�1)

Preplant

N 33.3P 20.4K 141.9Ca 95.8Mg 6.2S 33.12FertigationN 30.4P 8.6K 63.9Ca 12.3

17

Water content at a depth of 20 cm varied greatly as afunction of the localisation of irrigation tubes, althoughthis was less evident for treatments in which smallerquantities of water were supplied (DI) (Fig. 3). At RIIthere was no significant response to the increase of water

supply in terms of soil moisture, mainly when DII wascompared with DIII.

In general, at a depth of 40 cm (Fig. 3), for eachirrigation, soil moisture conditions at the differentemitter depths were similar, except for RII with DIII,which presented water content values significantly lowerthan those obtained for the other treatments: between 27and 55 DAP and at 68 DAP.

At depths of 60, 80 and 100 cm (Fig. 4), small dif-ferences between DII and DIII were recorded in terms ofsoil moisture. With reduced water application (DI),moisture at depths of 80 and 100 cm on dates between40 and 62 DAP was higher at RI. The soil moistureunder different irrigation regimes and emitters place-ment was often above values corresponding to the fieldcapacity (�9%) (Fig. 1). This can be explained by thefrequency of water application (on alternate days) whicheven with reduced applications could give rise to waterloss through percolation to depths where no roots werefound.

Root parameters

The combined effect of water application and emittersplacement contributed to a water status at 20 cm depth

Fig. 2 Daily precipitation and monthly air temperature during thegrowing season

Fig. 3 Volumetric water content (%) at 20 cm (a) and 40 cm (b)depth. (The vertical bars are the SE of the means); 0.6 ETc, DI; 0.9ETc, DII; 1.2 ETc, DIII; Surface, R0; 20 cm, RI; 40 cm, RII

18

of the soil profile distinct for the different treatments,thus providing conditions which could affect rootgrowth. However, La measured at different stages ofdevelopment, at a distance of 10 cm from the crop row,with some exceptions which did not enable distinct

behavioural patterns to be established, was in statisticalterms affected neither by the treatments studied nor bytheir interaction (Table 5).

In a study carried out on the same soil, using identicalculture techniques and drip surface irrigation, Machadoet al. (2000) observed that tomato root density in the croprow, evaluated on three dates throughout the seasonusingsoil–root cores, was also not affected by the quantity ofwater applied. Similar to the present study, the tomatowas transplanted.However, for the direct seeding tomato,

Fig. 4 Volumetric water content (%) at 60 cm (a), 80 cm (b), and100 cm (c) depth (the vertical bars are the SE of the means). 0.6ETc, DI; 0.9 ETc, DII, 1.2 ETc, DIII; Surface, R0; 20 cm, RI;40 cm, RII

19

the increase in the quantity of water applied led to greaterroot development in the top 30 cmof the soil, according toobservations reported by Bar-Yosef et al. (1980); May-nard et al. (1980) and Oliveira et al. (1996). However,Sanders et al. (1989b) observed contrasting results in atrial in which 0.35, 0.70 and 1.05 of ETo were applied, theroot density being the highest in the first 30 cm of soilwhere the lowest water regime was applied. In this studyhowever, a different irrigation system, the TravelingTrickle Irrigation System, was used.

Root-length density along the soil profile, estimatedfrom the regression equations defined for the minirhi-zotron calibration (Machado and Oliveira 2003), isshown in Figs. 5, 6. For different water regimes andirrigation depths, most of the root system was concen-trated within the top 40 cm of the soil profile, whereRLD reached 0.5–1.4 cm cm�3 .

The maximum rooting depth (1 m) was similar fordifferent treatments and was achieved between the stageof full development of the first blooming and the stagewhen 75% of fruits were red and orange (Figs. 5, 6).These results indicate that rooting depth was indepen-dent of water regime and drip irrigation depth. Locationof drip irrigation tubes did not influence the root systemdepth and this has been reported by Kamara et al. (1991)for cotton, Phene et al. (1991) for corn and Bryla et al.(2003) for faba bean, all using soil–root cores, andMachado et al. (2003) for processing tomato and usingminirhizotron data.

Crop production and fruit quality

Dry biomass production was significantly affected bywater application, the values being equal for DII and

Table 5 La at different depths, at four crop stages (data have been transformed)

Treatment Depth (cm)

0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 90–100

Beginning of fruit setDI 1.50 4.71 1.00 1.00 2.07 1.00DII 1.00 1.80 1.00 3.24 3.79 1.50DIII 1.00 3.88 2.53 5.25 2.43 2.23R0 1.00 4.71 1.00 3.48 2.97 1.00bRI 1.50 2.16 1.50 2.77 2.43 1.00bRII 1.00 3.52 2.03 3.24 2.89 2.73aF(D) 1.00NS 1.09NS 1.00NS 4.00NS 0.50NS 1.23NSF(R) 1.00NS 1.99NS 1.00NS 0.15NS 0.11NS 4.77*F(D · R) 1.00NS 3.81* 1.00NS 5.92** 3.78* 1.83NS

Blooming

DI 2.49 1.83 1.36 4.34 a 1.30 1.36 b 4.45DII 2.18 3.46 2.87 1.00 b 2.49 1.83 b 5.83DIII 5.43 3.44 1.60 1.41 b 3.70 6.69 a 5.18R0 3.81 5.18a 2.67a 3.03 1.30 4.11 4.75RI 5.58 1.00b 1.00b 2.72 3.70 2.12 4.88RII 4.71 2.55ab 2.16a 1.00 2.49 3.66 5.83F(D) 5.20NS 5.19NS 1.59NS 22.70** 2.07NS 2.66** 0.21NSF(R) 3.35NS 4.72* 4.63* 0.73NS 1.75NS 0.80NS 0.14NSF(D · R) 7.14** 2.10NS 2.47NS 0.54NS 4.96* 0.56NS 2.89NS

When 75% of the fruits were red or orange

DI 6.19 a 4.22 5.48 3.82 3.74 2.69 6.28 4.93 5.99 6.60DII 5.37 a 3.46 2.12 2.49 4.24 7.08 7.08 3.38 6.77 6.81DIII 1.72 b 5.03 5.60 5.69 5.06 4.74 3.02 3.41 4.40 4.73R0 4.02 3.46 4.04 5.82a 2.37 3.42 5.28 4.23 3.57b 3.17cRI 3.28 6.28 5.11 2.54b 5.40 5.88 3.46 3.58 5.35b 6.09bRII 5.98 2.97 4.05 3.64ab 5.27 5.21 7.64 3.91 8.25a 8.88aF(D) 22.11** 0.42NS 3.59NS 3.74NS 0.07NS 1.09NS 2.09NS 3.25NS 0.45NS 0.18NSF(R) 1.60NS 3.88NS 0.18NS 4.50* 2.82NS 1.08NS 0.75NS 0.05NS 9.18* 12.86**F(D · R) 3.09NS 7.20** 1.75NS 7.27** 4.98* 2.42NS 0.28NS 1.81NS 3.91* 3.17NS

Harvest

DI 2.35b 2.31 1.51 b 6.12 2.91 2.57 5.03 1.89 4.12 4.47DII 8.13a 2.66 3.33 ab 3.29 1.58 3.98 3.58 1.86 4.38 5.01DIII 6.79a 5.63 6.34a 5.21 3.10 7.07 7.93 6.26 4.78 6.66R0 6.21 4.90 2.70 4.91 2.67 4.05 5.01 1.89 2.49 3.76RI 6.20 2.94 3.92 5.37 2.54 5.29 5.57 3.82 4.58 5.91RII 4.85 2.76 4.55 4.34 2.38 4.28 5.95 4.30 6.21 6.47F(D) 27.15** 4.52NS 7.96* 0.70NS 0.92NS 3.81NS 2.51NS 4.32NS 0.13NS 0.42NSF(R) 0.27NS 1.27NS 0.86NS 0.12NS 0.02NS 0.23NS 0.11NS 1.86NS 1.98NS 0.63NSF(D · R) 0.98NS 0.43NS 1.24NS 1.54NS 0.20NS 0.21NS 0.59NS 4.67* 0.58NS 1.74NS

Within each column means with different letters are significantly different*P<0.05; **P<0.01 levels, respectively (LSD)

20

DIII regimes except at the stage when 75% of fruits werered and orange, when dry biomass was higher for DIII.DI presented lower values for the last three crop stages(Fig. 7). Lateral placement of drips did not have a sig-nificant effect on dry matter production (data notshown). Similar results were reported by Bryla et al.(2003) for faba bean.

The response of tomato to increasing levels of irri-gation was similar in quantitative and qualitative termsfor the different drip irrigation depths (Tables 6, 7).Total and commercial production increased with irri-gation from 0.6 to 1.1 ETc (Fig. 8). The same behaviourwas reported by Bar-Yosef et al. (1980); Sanders et al.(1989a); Oliveira et al. (1996) and Machado et al. (2000)for surface drip irrigation.

Water applied did not affect orange fruits yield nei-ther those with blossom-end rot (BER). The latter wasunexpected, as the tendency is for an increase in numberwith the reduction of the soil water potential, in accor-dance with the observations of Pill and Lambeth (1980)and Grierson and Kader (1986). This phenomenoncould be related to the cultivar used and/or the fact thatthe electrical conductivity of the irrigation water,

induced by fertilisation never rose above 2.5 mS cm�1 .Susceptibility to BER varies with the cultivar (Adamsand Ho 1992) and an excess of fertiliser or saline waterreduces Ca uptake and leads to BER (Ho 1998).

Rotten-fruit production rose with the increased waterapplied (Table 6). This is in accordance with the obser-vations recorded by Rudich et al. (1977); Williams andSistrunk (1979); Sanders et al. (1989a).

Regarding tomato fruit quality, �Brix and pH werenot significantly affected by the irrigation depth or bythe treatment interaction (Table 7). This is in agreementwith the observations of Davis et al. (1985); Phene et al.(1986, 1987) and Machado et al. (2003). The water ap-plied had a significant effect on the concentration ofsoluble solids (�Brix) and total soluble solids (t ha�1).Similar observations were recorded by Sanders et al.(1989a); May et al. (1990); Dumas et al. (1994) andMachado et al. (2000).

Water applied by irrigation had the converse effectson commercial yield and �Brix (Fig. 8). There was anegative relationship between fruit yield and �Brix(r=�0.82, P<0.001).

Fruit pH was not affected by the water applied. Incontrast, Sanders et al. (1989a) reported that fruit pHdecreased as irrigation rates increased. Total solublesolids (t ha�1) were significantly lower under DI andtreatments with greater irrigation water supplied be-haved similarly (Table 7).

Fig. 5 Root length density at beginning of fruit set and atblooming stages for different treatments (data represent averageof four replicates); 0.6 ETc, DI; 0.9 ETc, DII; 1.2 ETc, DIII;Surface, R0; 20 cm, RI; 40 cm, RII

21

Irrigation water use efficiency (IWUE), defined asthe ratio of crop yield to seasonal irrigation waterapplied, including rain (Howell 1994), was not signifi-

cantly affected by treatments, the IWUE values vary-ing from 0.202t ha�1 mm�1 to 0.228 t ha�1 mm�1

(Table 6).

Conclusions

The hypothesis that a different crop response wouldresult from different quantities of water applied bydrip irrigation at different depths was refuted. Theresults of this field study indicate that, at 10 cm fromthe plant row, the water regime and the depth ofirrigation tubes did not affect root-length intensity orrooting depth.

The response of tomato to the water supplied wassimilar in quantitative and qualitative terms for differentirrigation depths.

Commercial yield was higher for the treatment wherein the quantity of water applied was the greatest.However, the yield of total soluble solids (t ha�1) wasthe same for the irrigation treatments 0.9 and 1.2 ETc.Therefore, when tomato is grown for the production ofconcentrates for the industry, the quantity of water to beapplied can be 0.9 ETc, which will increase to the samequantity of total soluble material as that obtained withthe highest irrigation regime, thereby reducing costs andimproving the quality of raw materials.

Fig. 6 Root length density at 75% of the fruits red or orange andharvest stages for different treatments (data represent average offour replicates); 0.6 ETc, DI; 0.9 ETc, DII; 1.2 ETc, DIII; Surface,R0; 20 cm, RI; 40 cm, RII

Fig. 7 Total aboveground biomass in different crop stages for thethree irrigation regimes [DI (D), DII (o) and DIII (h)] (the verticalbars are the SE of the means)

22

References

Adams P, Ho LC (1992) The susceptibility of modern tomatocultivars to blossom-end rot in relation to salinity. J Hort Sci67:827–839

Ayars JE, Phene CJ, Hutmacher RB, Davis KR, Schoneman RA,Vail SS, Mead RM (1999) Subsurface drip irrigation of row

crops: a review of 15 years of research at the Water Manage-ment Research Laboratory. Agric Water Manage 42:1–27

Bar-Yosef B, Stammers C, Sagiv B (1980) Growth of trickle-irri-gated tomato as related to rooting volume and uptake of N andwater. Agron J 72:815–822

Bar-Yosef B, Martinez HJJ, Sagiv B, Levkovitch I, Markovitch,Phene CJ (1991) Processing tomato response to surface andsubsurface drip phosphorus fertigation Bard Project ScientificReport. Bet Dagan, Israel, pp 175–191

Bryla DR, Banuelos GS, Mitchell JP (2003) Water requirements ofsubsurface drip-irrigated faba bean in California. Irrig Sci22:31–37

Camp CR (1998) Subsurface drip irrigation: a review. Trans ASAE41(5):1353–1367

Camp CR, Garrett JT, Sadler EJ, Busscher WJ (1993) Microirri-gation management for double-cropped vegetables in a humidarea. Trans ASAE 36(6):1639–1644

Davis KR, Phene CJ, McCormick RL, Hutmacher RB, Meek DW(1985) Trickle frequency and installation depth effects ontomatoes. Proc Third Int Drip/Trickle Irrigation Congress,Fresno, California, pp 896–901

Doorenbos J, Kassam AH (1986) Yield response to water. FAO,Irrigation and Drainage Paper, 33, Rome

Dumas Y, Leoni C, Portas CAM, Bieche BJ (1994) Influence ofwater and nitrogen availability on yield and quality of pro-cessing tomato in the European Union countries. Acta Hort376:185–192

El-Gindy AM, El-Araby AM (1996) Vegetable crops to response tosurface and subsurface drip under calcareous soil. Proc IntConf on Evapotranspiration and Irrigation Scheduling, St Jo-seph, pp 1021–1028

Grierson D, Kader AA (1986) Fruit ripening and quality. In:Atherton JG, Rudich J (eds) The tomato crop. Chapman&Hall, New York, pp 241–280

Ho LC (1998) The physiological basis for improving tomato fruitquality. Acta Hort 487:33–40

Howell T (1994) Irrigation engineering, evapotranspiration In:Arntzem CJ, Ritter EM (eds) Encyclopaedia of agriculturalscience, vol 2. Academic, Orlando, FL, pp 591–600

Hutmacher RB, Vail SS, Muthamia JG, Mwaja V, Liu RC (1985)Effect of trickle irrigation frequency and installation depth ontomato growth and water status. The Proceedings of the ThirdInternational Drip/Trickle Irrigation Congress, Fresno, CA, pp798–803

Kamara L, Zartman R, Ramsey RH (1991) Cotton-root distribu-tion as a function of trickle irrigation emitter depth. Irrig Sci12:141–144

Machado RMA (2002) Estudos sobre a influencia da rega-gota-a-gota subsuperficial na dinamica de enraizamento, no rendi-mento fisico e na qualidade da materia-prima do tomatede industria. Tese de doutoramento, Universidade de Evora,Evora

Table 6 Yield components and IWUE

Treatment Yield (t ha�1) Fruit (t ha�1) IWUE (t ha�1 mm�1)b

Total Commerciala Red Orange Greens Rotten Blossonnd rot

DI 78.75c 69.84c 66.18c 3.65 6.91b 1.21b 0.80 0.214DII 123.75b 108.14b 102.36b 5.78 12.41a 1.92b 1.10 0.228DIII 141.67a 130.05a 126.18a 3.87 7.58b 3.51a 0.53 0.202R0 118.76 105.49 100.36 5.14 10.26 2.54 0.46 0.216RI 117.76 106.69 102.17 4.52 8.03 2.07 0.97 0.225RII 107.47 95.84 92.20 3.65 8.60 2.03 0.99 0.202F (D) 40.48*** 37.01*** 39.9*** 1.81NS 5.76* 21.09** 1.72NS 1.17 NSF (R) 1.48NS 1.43NS 1.12NS 1.82NS 1.11NS 1.07NS 1.15NS 0.92NSF (D x R) 1.79NS 1.73NS 1.63NS 0.61NS 0.83NS 0.82NS 0.38NS 0.89NS

a Red and orange fruits.b Commercial yield/applied water.Within each column means with different letters are significantly differ-ent.*P<0.05; **P<0.01; ***P<0.001 levels, respectively (LSD).

Table 7 �Brix, pH and total soluble solids (t ha�1)

Treatment �Brix pH Soluble solids (t ha�1)a

DI 6.07a 4.37 4.20bDII 5.10b 4.36 5.48aDIII 4.51c 4.32 5.85aR0 5.14 4.37 5.20RI 5.28 4.30 5.48RII 5.25 4.38 4.87F (D) 127.87** 0.28NS 8.49*F (R) 0.03NS 0.82NS 1.89NSF (D·R) 0.29NS 0.65NS 1.40NS

aCommercial yield (t ha�1)·�Brix/100Within each column means with different letters are significantlydifferent *P<0.05; **P<0.001 levels, respectively) (LSD).

Fig. 8 Relationship between commercial yield, �Brix and appliedwater ðCommercial yield ¼ �91:106ð%ET2

cÞ þ 264:34ð%ETcÞ�55:973;R2 ¼ 0:6265; P\0:001; �Brix ¼ 2:1435ð%ET2

cÞ � 6:4458ðETcÞ þ 9:165;R2 ¼ 0:7253; P\0:001Þ

23

Machado RMA, Oliveira MRG (2003) Comparison of tomato rootdistributions by minirhizotron and destructive sampling. PlantSoil 255(1):375–385

Machado RMA, Oliveira MRG, Portas CAM (2000) Effect of dripirrigation and fertilization on tomato rooting patterns. ActaHort 537:313–320

Machado RMA, Oliveira MRG, Portas CAM (2003) Tomato rootdistribution, yield and fruit quality under subsurface drip irri-gation. Plant Soil 255(1):333–341

May DM, Walcott TT, Peters D, Grimes D (1990) Moisture stressas it affects yields, soluble solids and viscosity of tomatoes. ActaHort 277:123–128

Maynard DN, Lorenz AO, Magnifico V (1980) Growth andpotassium partitioning in tomato. J Am Soc Hort Sci 105:79–82

Oliveira MRG, Calado AM, Portas CAM (1996) Tomato rootdistribution under drip irrigation. J Am Soc Hort Sci121(4):644–648

Phene CJ (1991) Advances in irrigation under water shortageconditions. The Proceedings of the Conference on Collabora-tive Research and Development Applications in Arid Lands,Santa Barbara, CA, pp 93–110

Phene CJ, Bar-Yosef B, Hutmacher RB, Patton SH, Davis KR,McCormick RL (1986) Fertilization of high yielding subsurfacetrickle irrigated tomatoes. The Proceedings of the 34th AnnualCalifornia Fertilizer Conference, Fresno, CA, pp 33–43

Phene CJ, Davis KR, Hutmacher RB, McCormick RL (1987)Advantages of subsurface irrigation for processing tomatoes.Acta Hort 200:101–113

Phene CJ, Davis KR, Hutmacher RB, Bar-Yosef B, Meek DW,Misaki J (1991) Effect of high frequency surface and subsurfacedrip irrigation on root distribution of sweet corn. Irrig Sci12(2):135–140

Phene CJ, Hutmacher RB, Ayars JE, Davis KR, Mead RM,Schoneman RA (1992) Maximizing water use efficiency withsubsurface drip irrigation. International Summer Meeting,Charlotte, 20p

Pill WG, Lambeth VN (1980) Effects of soil water regime andnitrogen form on blossom—end rot, yield, water relations, andelemental composition of tomato. J Am Soc Hort Sci105(5):730–734

Rudich J, Kalmar C, Geizenberg C, Harel S (1977) Low watertensions in defined growth stages of processing tomato plantsand their effects on yield and quality. J Hort Sci 52:391–399

Sammis TW (1980) Comparison of sprinkler, trickle, subsurface andfurrow irrigation methods for row crops. Agron J 72(5):701–704

Sanders DC, Howell TA, Hile MMS, Hodges L, Meek D, Phene CJ(1989a) Yield and quality of processing tomatoes in response toirrigation rate and schedule. J Am Soc Hort Sci 114(6):904–908

Sanders DC, Howell TA, Hile MMS, Hodges L, Phene CJ (1989b)Tomato root development affected by travelling trickle irriga-tion rate. Hort Sci 24(6):930–933

Tennant D (1975) A test of a modified line intersect method ofestimating root length. J Ecol 63:995–1001

Williams J, Sistrunk W (1979) Effects of cultivar, irrigation, ethe-phon and harvest date on the yield and quality of processingtomatoes. J Am Soc Hort Sci 104(4):435–439

24