- Email: [email protected]
Effects of planting density on the productivity and water use of tea (Camellia sinensis L.) clones
agricultural water management 90 (2007) 224–232
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/agwat
Effects of planting density on the productivity and water use of tea (Camellia sinensis L.) clones I. Measurement of water use in young tea using sap flow meters with a stem heat balance method Julius M. Kigalu * Tea Research Institute of Tanzania, Ngwazi Tea Research Station, P.O. Box 2177, Dar es Salaam, Tanzania
article info
abstract
Article history:
Sap flow meters based on the stem heat balance method were used to measure the mass
Accepted 12 March 2007
flow rates or water use in young potted tea (Camellia sinensis L.) plants of clones AHP S15/10
Published on line 4 May 2007
and BBK35. The meters were constructed on site and installed onto the stem or branch sections of field growing plants in an experiment originally designed to study the effects of
Keywords:
plant population density and drought on the productivity and water use of young tea clones.
Evapotranspiration
The objective of the study was to use the SHB method as a first attempt to use sap flow
Irrigation
meters for determining the water use of young tea growing in the field under well watered
Plant density
conditions in Tanzania. The results are reported and recommendation made for further
Sap flow
work on using the technique. # 2007 Elsevier B.V. All rights reserved.
Stem heat balance method Tea Transpiration Water use
1.
Introduction
Since early in the 1990s, the tea industry in Tanzania continues to expand with substantial new planted areas especially in the Southern Highlands where tea plantations are predominantly irrigated during the long dry seasons. However, despite the extensive tea field expansions there is yet little information on plant population density and the effect on crop productivity and water use of young tea. High density planting can increase initial yields of young tea, encouraging early establishment of full ground cover and minimizing water losses due to surface evaporation and runoff. Although sap flow meters have been used to measure the water use of many crop plant species adoption of this technology in tea is very limited.
There are significant developments of techniques to measure directly sap flow in plants in order to quantify with precision the total amount of water transpired (Swanson, 1994). Briefly, there are two methods for sap flow measurements: the heat pulse velocity (HPV) based on the determination of sap velocity within a stem section (Huber, 1932; Dugas, 1990) which is invasive and may be destructive to the plant, and the stem heat balance (SHB) which estimates mass flow rates (Cermak et al., 1973; Sakuratani, 1984; Baker and van Bavel, 1987; Steinberg et al., 1989; Ishida et al., 1991; Batho et al., 1994; Boersma and Weibel, 1995). In a detailed review, Swanson (1994) reported that in recent years, techniques have been developed to measure the sap flow rate in a number of woody plant species including herbaceous plants (Sakuratani, 1984) and poplar (Populus spp.) clones (Souch, 1996), providing
* Tel.: +255 22 2122033; fax: +255 22 2113838. E-mail addresses: [email protected], [email protected]. 0378-3774/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.agwat.2007.03.005
agricultural water management 90 (2007) 224–232
accurate quantification of the amount of water transpired. The main advantage of the SHB technique is that sap flow meters are non-destructive to the plants and have lower costs than other methods, including weighing lysimeters (Cermak et al., 1973, 1984; Sakuratani, 1981; Swanson, 1994). The SHB method uses a small heater wrapped around the plant stem or branch section to provide a heat input into the section. Heat storage in the section may also be significant on trees with stem diameters greater than 25 mm, but this is negligibly small at smaller diameters (Batho, 1993; Batho et al., 1994; Grime et al., 1995). Thus if a known amount of heat is supplied into a stem section and a constant amount of sap (water) flows through the section during the process of transpiration, the temperature of the sapwood will reach a steady value which is inversely proportional to the water flow (Cermak et al., 1984; Swanson, 1994; Sakuratani, 1984). Therefore, under ideal conditions the amount of heat energy carried upwards out of the stem section (by conduction) by the mass flow of water will be equal to the amount of heat energy input to the section (Cermak et al., 1984; Batho et al., 1994). The objective of the current study was to use the SHB method described in detail by Weibel and Devos (1994), Batho et al. (1994) and Kigalu et al. (1995) to measure sap flow or transpiration rates in young tea plants in the field.
2.
Materials and methods
2.1.
Description of the site, climate and soils
The field trials reported here were conducted at Ngwazi Tea Research Station, NTRS (latitude 88320 S, longitude 358100 E, altitude: 1840 m a.s.l.) in the Mufindi District in Southern Tanzania. A full description of the climate, the weather and soils of the site is provided by Stephens and Carr (1991a,b), Burgess (1992), Burgess and Carr (1996), Kigalu (1997). Briefly, the climate of the area can be divided into three main seasons based on rainfall and temperature. More than 95% of the annual rainfall, ranging from 800 to 1100 mm occurs between the end of November and May. The corresponding monthly mean air temperature is 16–19 8C. The dry season can be divided into two: cool (13–16 8C) from June to August and warm (16–19 8C) from September to November. The experimental area is on weathered soils classified as a Xanthic ferralsol (Baillie and Burton, 1993; FAO-UNESCO, 1988) with a kaolinitic sandy clay down to 0.15 m depth and friable clay below. The mean soil pH was 5.4 and 5.8 at 0.15 and 0.70 m depths, respectively, which is suitable for growing tea (pH 4.9– 5.6). The volumetric water content of the soil at a water potential of 10 kPa were low for clay soils and more typical sandy loam, corresponding to field capacity (FC; 10 kPa), increased with depth from 24.7% in the top 0.15 m to 32.5% at 1.80 m where the silt and clay content was higher. The available water content held between field capacity (FC; 10 kPa) and permanent wilting point (PWP; 1500 kPa) ranged from 110 to 122 mm m1 1 in the top 2 m of soil to 93 mm m1 at 2.70 m. Regarding management of the tea bushes in the main field experiment before the field sap flow trials, fertiliser was applied in February and July 1995 on a per unit area basis at the rate of 100 kg N ha1 as N:P2O5:K2O in the ratio of 2:1:2. Also, in
225
1995 foliar zinc oxide (ZnO) was also uniformly applied at the rate of 4.5 kg ZnO ha1 in three equal splits to correct for zinc deficiency. The experimental field was kept weed free by regular hand weeding and herbicide application. At NTRS, overhead sprinkler irrigation of young tea may start in mid May to the end of November/December in bad years when rains stop early and start late. Irrigation scheduling and calculation of the potential soil water deficit (SWD, mm) are based on the soil water balance Eq. (1) described in detail at the site (Stephens and Carr, 1989; Burgess, 1992): SWDi ¼ SWDi1 Ri Ii þ Di þ ETi
(1)
where SWDi1 is the soil deficit on the previous ((i 1)th) day, Ri the rainfall, Ii the irrigation applied, Di the drainage and ETi is the crop evapotranspiration all recorded on the ith day and measured in mm. The value of Ii on each date of irrigation was recorded in 56 catch cans which were spaced symmetrically across the experiment. Any irrigation or rainfall received when the soil was still at field capacity (i.e., SWD = 0) was assumed to drain instantly through the soil profile. Movement of surface water (run-off) between plots or treatments was assumed negligible as it was prevented by micro-catchments dug between the rows of tea bushes, and by the accumulated mulch and tea leaf liter. The value of ETi was calculated from the daily evaporation pan measurement (Epan, mm) multiplied by a crop factor Kc. The value of Epan measured at NTRS gives a good estimate of the water loss from a mature stand of tea with or close to complete ground cover, implying negligible evaporation from the soil surface (Stephens and Carr, 1991a,b; Burgess, 1992). It was assumed that losses due to transpiration were directly proportional to the ground cover by the canopy, and Kc was assumed to be equal to 1.0 on the day of irrigation due to soil evaporation losses (Dagg, 1970; Ritchie, 1972). The soil water deficit (SWD; mm), incident solar radiation (S; MJ m2 d1) and wind speed (u; m s1) were calculated daily from measurements recorded at a nearby NTRS meteorological weather station.
2.2.
Features of the sap flow meters
This study used the SHB technique described in detail by Weibel and Devos (1994) and Batho et al. (1994) for the first time in tea in Tanzania. Briefly, the sap flow meters were constructed on site based on the SHM method following the procedures described by Batho et al. (1994) as briefly illustrated in Fig. 1, representing a schematic diagram of a sap flow meter attached to a stem or branch section with symbols briefly described as follows. The sap flow rate (F; g s1) was calculated from the energy balance across the sap flow meter (Eq. (2)) which states that: F¼
½Pin Q cd Q r þ Q s ðCs dTsap Þ
(2)
where using the symbols illustrated in Fig. 1: Qf = [Pin Qcd Qr + Qs] is the amount of heat (W) transported in the moving sap at differential temperature dTsap given by dTsap = [dTa + dTb]/2 (8C), where dTa and dTb are temperature differentials of the sap up and down stream measured by thermocouples a and b, respectively. Pin is the heater power
226
agricultural water management 90 (2007) 224–232
dThc; 8C) and one combined heater/flux meter (Fig. 1). An analogue heater control unit was used to maintain the constant differential temperature between the heated stem segment and the unheated stem below the meter (dThc) at 2 8C. The thermocouples were taped directly around the bark of the stem of the tea plants with PTFE tape. The combined heater/flux meter provided the necessary heat to the sap and measured radial heat loss (dTr; 8C) (Fig. 1). Measurements of transpiration were made with the sap flow meters in selected clone density treatment combinations (Table 1) of well-irrigated tea in a field experiment (labelled N12).
2.3.
Fig. 1 – Schematic diagram of a sap flow meter showing the combined heater/fluxmeter and the thermocouples, which measured temperature differentials. The heater control unit maintained the constant temperature differential dThc at 2 8C. The temperature differentials (dTa and dTb) were used to calculate the components of heat balance in the stem segment as described in detail below (modified from Kigalu et al., 1995).
input (W). Qcd is the heat conduction loss along the stem up and down stream (W); Qcd = [(dTb dTa) Kst A]/dz, where A is stem section cross-sectional area (m2), Kst is the thermal conductivity constant of the stem sapwood (W m1 K1). In this study, the value of Kst was assumed to be 0.54 W m1 K1 (Baker and van Bavel, 1987; Sakuratani, 1981), and dz is the distance between thermocouples a and b (m). Qr is the radial heat conduction loss (W) = dTr Ksh, where dTr is the thermopile radial temperature differential (8C), Ksh is a constant called the sheath conductance of the gauge system (W K1) calculoss Cs/ lated from Ksh = [(Pin Qcd Qs) (weight 3600)] dTsap, where weight loss is derived from sap flow per hour (Qf). Assuming that Qs is negligibly small and that Qf = 0 at night (Sakuratani, 1981), then Ksh = [Pin Qcd Qs]/ dTr. Qs is the heat stored in the stem section (W); negligible for small stem (<30 mm) diameters. Therefore, if Qs is very small, obtain Qf = [Pin Qcd Qr] (W), and sap flow rate = F = [Pin Qcd Qr]/(Cs dTsap) (g s1) or F = Qf 3600/ 1 (dTsap Cs) (g h ). Cs is the specific heat capacity of the sap or water (4.186 J g1 8C1). dThc is the temperature differential between the heater and the stem section (8C). Each sap flow meter consisted of three pairs of thermocouples (to measure differential temperatures dTa, dTb and
Field measurements
The main field experiment described in detail by Burgess and Kigalu (1993) and Kigalu (1997) comprised of two broad-leaved clones of contrasting growth behaviour. These are clones AHP S15/10 and BBK35 with spreading and upright or erect habit, respectively. Both clones originated from Kenya and are of scientific and economic importance in East and Central Africa. These were planted in six plant population density treatments, ranging from the lowest (D1: 8333 plants ha1 at a spacing of 1.00 m 1.20 m; below the usual commercial density of 10,417 plants ha1 corresponding to a spacing of 0.80 m 1.20 m) to the highest density (D6: 83,333 plants ha1 at a spacing of 0.60 m 0.20 m). Irrigation was applied under seven varying treatments, ranging from non-irrigated (I0) to well or full irrigation (I6) conditions. In fully irrigated plots, the irrigation water amounts applied reduced the cumulative potential SWD to zero (i.e., the soil was reduced to field capacity) for each irrigation date. An initial trial was conducted for 2 days (1–2 March 1995) in potted plants to compare the trends in transpiration rates with time of the day between one plant of each of the two clones (AHP S15/10 and BBK35) concurrently. This was followed by field measurements of sap flow comprising of three series of trials. In the first trial (9–12 November 1995) four plants from each clone (AHP S15/10 and BBK35) were selected from neighbouring sub-plots of the same density (D2: at a spacing of 0.6 mm 1.33 m; 12,531 plants ha1). The second and third series were conducted from 22 to 24 November and from 4 to 6 December in 1995 to quantify the effects of planting density on the transpiration rates of fully irrigated (I6) plants of each of the clones AHP S15/10 (series two) and BBK35 (series three) at the lowest (D1) and highest (D6) densities, using four sap flow meters (i.e., four plants) at each density. Sap flow was adjusted by dividing the rate of sap flow (g h1) by the total area of the leaves (m2) on the stem or suitable branch on which the gauge was attached as described by Valancogne et al. (1993). The results were then adjusted using two methods. Firstly, they were multiplied by the product of the total leaf area (m2) of the whole plant and divided by the area for each plant (m2), i.e., the leaf area index (LAI) calculated as reported by Pethiyagoda and Ragendram (1965), to obtain transpiration per unit leaf area using Eq. (3): Tl ¼
X f i =LAi LAI n
ði ¼ 1; 2; . . . ; nÞ
(3)
227
agricultural water management 90 (2007) 224–232
Table 1 – Series trials and dates of measurements of daily transpiration (T) by (a) both clones AHP S15/10 and BBK35 at the same density treatment D2, and by (b) clone AHP S15/10 and (c) clone BBK35 at the same density treatments D1 and D6 Transpiration (T; mm day1)
Series trials and dates
Clone S15/10
AHP
ETo (mm day1)
Clone BBK35
Kc = (T/ETo)/C for canopy cover AHP S15/10
BBK35
(1) Both clones at density D2 9 November 1995 10 November 1995 11 November 1995 12 November 1995
2.1 2.3 3.1 3.3
1.9 2.1 2.7 2.6
5.9 6.5 6.5 6.1
0.43 0.42 0.57 0.65
0.66 0.66 0.87 0.89
Mean S. E. () (n = 4)
2.7 0.56
2.3 0.38
6.3 0.30
0.51 0.11
0.76 0.12
(2) Clone AHP S15/10 density 22 November 1995 23 November 1995 24 November 1995
D1 4.3 5.3 5.5
D6 2.2 3.3 3.6
5.6 6.1 6.6
D1 0.92 1.04 0.99
D6 0.41 0.57 0.58
Mean S. E () (n = 4)
5.0 0.65
3.1 0.71
6.1 0.50
0.98 0.06
0.53 0.09
(3) Clone BBK35 density 4 December 1995 5 December 1995a 6 December 1995a
D1
Mean S. E. () (n = 3)
D6 2.6 2.3 1.6
1.9 2.8 2.7
D1 5.2 5.4 2.9
D6 0.86 0.74 0.94
0.40 0.56 1.01
2.2 0.49
2.5 0.49
4.5 1.39
0.85 0.12
0.65 0.32
The daily reference evapotranspiration (ETo) was estimated from the Penman–Monteith equation (Monteith, 1965) using the DAILYET computer programme (Hess, 1997), and the corresponding calculated crop coefficients (Kc) adjusted for canopy cover from well-watered plants of both clones. The measurements were recorded on selected days in November/December 1995. The values of canopy cover (C; %) for trial series (a) were 84 and 48% for clones AHP S15/10 and BBK35 at plant density treatment D2; (b) 84 and 95% for clone AHP S15/10 at density D1 and D6; and (c) 58 and 92% for clone BBK35 at density treatments D1 and D6, respectively. a Very cloudy days; and first rain in the afternoon (14.5 mm) on 6th December 1995 associated with low values of short-wave solar radiation (10 MJ m2 day1), wind speed (1.6 m s1), T and ETo.
where Tl is the adjusted transpiration (g m2 leaf area h1), fi the measured sap flow (g h1), and LAi is the leaf area (m2) of ith plant. Secondly, the sap flow rates (g h1) were adjusted on a unit ground area basis by multiplying by the reciprocal of the area per plant (m2) associated with each gauge–plant combination (Ham et al., 1990; Sakuratani, 1987) as in Eq. (4): Td ¼
X ð f ri Þ i ; n
ði ¼ 1; 2; . . . ; nÞ
(4)
where Td is the transpiration (g m2 h1), and ri is the reciprocal of the area per plant (m2) associated with the gauge–plant ith combination. Measurements of crop ground cover of the selected plants were taken immediately before commencement of sap flow data recording of each trial using the method described in detail by Burgess (1992) and by Kigalu (1997), and as reported by Kigalu and Burgess (1994). The daily transpiration rates (T; g m2 day1) were estimated from the cumulative values of the rates derived from Eq. (5) over a 24 h period. The equivalent mass of a kilogram of water in a metre square of ground area was then expressed in millimetre (mm) depth so that transpiration could be given in mm day1. The results were then adjusted based on the mean crop ground cover (C; %) of the plants within the sub-plot, by dividing by the proportion of ground cover to obtain estimates of the daily crop evapo-
transpiration (ETc; mm day1) (Dagg, 1967, 1970; Guetierrez and Meinzer, 1994) as in Eq. (5), used by Dagg (1970) in tea: ETc ¼
T C
(5)
Transpiration rates were adjusted on a per unit ground area basis using Eqs. (3) and (4), allowing for calculations and comparisons of the daily crop evapotranspiration rates (ETc; mm day1; Eq. (5)) with the daily reference evapotranspiration (ETo; mm day1), estimated using the DAILYET programme (Hess, 1997). Daily evapotranspiration rates were calculated using the approach described by Dagg (1970, 1967) for determining monthly evapotranspiration rates (Et) from young tea crop, which varied from 40 to 115 mm per month, but were satisfactorily predicted (R2 = 85%), both in wet and dry months, by Eq. (6) (Dagg, 1970): Et ¼ Eo ½0:9a þ ð1 aÞ 0:9 n
(6)
where Eo is the Penman estimate of open-water evaporation, a the fraction of the soil covered by the crop at noon, and n is the fractional number of rain days per month. Dagg (1970, 1967) further reported the relationship between ETC and ETo adjusted for crop ground cover as in Eq. (7). ETc ¼ 0:9ETo C
(7)
228
agricultural water management 90 (2007) 224–232
Fig. 2 – Comparison of transpiration rate per unit leaf area of young potted tea plants of clones AHP S15/10 and BBK35 measured by sap flow meters during the period 1–2 March 1995. Mean of two plants from each clone.
where the ratio ETc/(ETo C) gave a reasonable estimate of the transpiration factor, or crop factor Kc, over the area covered by the canopy whilst they are not under serious moisture stress. The crop factor or coefficient (Kc) for plant cover was estimated from the ratio of ETc to the reference crop evapotranspiration (ETo; mm day1), calculated with the Penman–Monteith equation (Monteith, 1965) using the daily evapotranspiration (DAILYET) computer programme (Hess, 1997) developed at Cranfield University, Silsoe College, UK.
3.
Results
3.1.
Clone differences in sap flow
For the potted plants both clones had a similar pattern or trend of transpiration rates, but the rate of transpiration (g h1 m2 leaf area) of clone AHP S15/10 was larger than that of clone BBK35 for most part of the day especially from 1100 to 1600 local mean time when the difference was more than twice at times (Fig. 2). In November 1995 the average crop cover of plants of clones AHP S15/10 and BBK35 in the field were 84 and 48%, respectively, whilst the leaf area indices ranged from 1.3 to 1.9 (mean: 1.6) and from 2.1 to 2.9 (mean: 2.5) for clones BBK35 and AHP S15/10, respectively. During the field trials all four gauges worked properly and the trends of sap flow rates per unit leaf area during the day that were observed from the field trials are shown in Figs. 3–5, and were similar to the trend for the potted plants in March 1995 (Fig. 2). Adjusting the water loss (measured using a sap flow meter) from each plant on a per unit leaf area basis indicated that water use by clone AHP S15/10 was greater than that from clone BBK35 from about 08:00 until 17:00 h on all the 4 days of measurements, but only twice as large for very short periods (Fig. 3), but transpiration per unit leaf area for clone AHP S15/10 was greater than that from clone BBK35.
3.2.
Density differences
The water use per unit leaf area by clone AHP S15/10 at the lowest plant density treatment (D1; with mean crop cover: 84%) was consistently larger than that at the highest population density (D6; mean crop cover: 95%), both from
Fig. 3 – Comparison of transpiration rate per unit leaf area of tea plants of clones AHP S15/10 and BBK35 measured at the same density D2 in the field using sap flow meters during the period 9–12 November 1995 (n = 4 for both clones AHP S15/10 and BBK35). The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
agricultural water management 90 (2007) 224–232
229
Fig. 4 – Comparison of transpiration rate per unit leaf area of fully irrigated (I6) tea plants of clone AHP S15/10 at density treatments D1 and D6 measured in the field using sap flow meters, 21–24 November 1995. The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
fully irrigated (I6) sub-plots (Fig. 4). The plants at the lowest density (D1) started transpiring earlier and continued longer than those at highest density (D6). Sap flow increased sharply from zero at around 07:00 h, and remained constant during the day before declining from 15:00 h onwards. Shade effects from tall trees below the experimental area might have been the cause for the changes in diurnal patterns of transpiration rates of clone AHP S15/10 in the highest density (D6). Fig. 5 shows the water use per unit leaf area by clone BBK35 at the same densities (D1 and D6, with mean crop cover of 58 and 92%, respectively) and irrigation treatments (I6). These results can be compared with those of clone AHP S15/10 (Fig. 4). The results were of a similar trend but the sap flow rate per unit leaf area for plants under the lowest density (D1) was larger
than under the highest (D6). However, contrary to clone AHP S15/10, on each of the 3 days of measurements, transpiration of clone BBK35 began at about 08:00 h and increased rapidly to peak values at 10:00–12:00 h before declining in a more or less constant way from then onwards. Although the measurements were recorded on different days, the absolute rates of transpiration per unit leaf area from clone BBK35 were about two thirds of those from clone AHP S15/10. The water use by the plants of clone AHP S15/10 (range: 2.1– 3.3 mm day1; mean: 2.7 0.56 mm day1) during the first series of measurements was consistently larger than that by clone BBK35 (range: 1.9–2.6 mm day1; mean: 2.3 0.38 mm day1), taken at the same time from the same density D2 (Table 1). Clone AHP S15/10 had relatively large Kc values (mean:
230
agricultural water management 90 (2007) 224–232
Fig. 5 – Comparison of transpiration rate per unit leaf area of fully irrigated (I6) tea plants of clone BBK35 at density treatments D1 and D6 measured in the field using sap flow meters, 4–6 December 1995. The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
0.98 and 0.53 for D1 and D6, respectively), adjusted for crop cover, compared to the corresponding values of 0.85 and 0.65 for clone BBK35. In Figs. 3–5 the soil water deficit values that are given were estimated from a nearby meteorological station. In future further research is required to quantify transpiration rates and the corresponding actual soil water deficits in the field to establish possible differences in SWD between plots within the trial resulting from differences in transpiration rate between clones and densities.
4.
Discussion
4.1.
Differences between clones
The findings presented in Table 1 and Figs. 3–5 indicated that the transpiration rates (T; mm day1) and the water use by young plants of clone AHP S15/10 were larger than those by clone BBK35 on all the days at all the densities (D1, D2 and D6) investigated in the three consecutive series of field trials. Using sap flow meters, based on the stem heat balance method, for the first time in tea in Tanzania, it was shown that
the crop coefficient Kc, derived from the ratio of ETc to ETo and adjusted for plant canopy cover, for young tea clones AHP S15/ 10 and BBK35 at the same plant density (D2) was 0.51 and 0.76, respectively.
4.2.
Density differences
The average crop coefficient (Kc) value, adjusted for plant canopy cover, for clone AHP S15/10 was 0.98 ( 0.06), ranging from 0.92 to 1.04 for the plants at the lowest density treatment D1, and 0.53 0.09 (range: 0.41–0.58) at the highest (D6). The corresponding values for clone BBK35 were 0.85 0.12 (range: 0.86–0.94) and 0.65 0.32 (range: 0.40–1.01). The fraction of radiation intercepted by the plants varies with solar angle (Dagg, 1970) and cloud cover especially in the morning and evenings, and this may have effects on the transpiration rate. Clone AHP S15/10, having a spreading growth habit, had a larger transpiring total leaf area (leaf area index) than clone BBK35. Unexpected shade could have some influence could have some influence on the diurnal patterns of transpiration rate for clone AHP S15/10 at density D6. However, high rates were observed in the afternoon for clone AHP S15/ 10, having the larger crop cover, at the three density
agricultural water management 90 (2007) 224–232
treatments (D1, D2 and D6), whilst the values for clone BBK35 declined earlier during the day, probably suggesting clonal differences in the mechanisms by which each clone avoids plant water deficits (Jones, 1992; Loomis and Connor, 1992).
4.3.
Mechanisms
Transpiration is influenced indirectly by xylem leaf water potential through mediation of stomatal aperture of plants (Jones, 1978, 1992). However, the relationships between sap flow and incident short-wave radiation, vapour pressure deficit and stomatal conductance (McNaughton and Jarvis, 1983; Jarvis, 1985; Jarvis and McNaughton, 1986), xylem water potential (Jones, 1978; Loomis and Connor, 1992) were not investigated in the present study. The effects of light, temperature and photosynthetic demand rates (Squire, 1977; Sakai, 1987; Smith et al., 1993; Burgess and Carr, 1997) were not covered. No doubt, these require detailed quantification during subsequent sap flow measurements in tea using this technique. These authors, however, provided a mechanism which can enable scaling up transpiration from leaf to region similar to the work by Guetierrez and Meinzer (1994). This study demonstrated that the water use per unit leaf area by young tea varied with clone (being larger by clone AHP S15/10 than clone BBK35), and was density dependent. The values were consistently larger at the lowest density treatment (D1) than at the highest (D6) for both clones under similar environmental conditions. Compared with results from previous studies, the Kc values from clones AHP S15/10 and BBK35 at the lowest (D1) density treatment (Table 1) compared very closely with those reported by Squire and Callander (1981) for seedling tea with almost full ground cover, and also by Dagg (1967) for cashew trees in Southern Tanzania, respectively. However, further research is required to quantify in detail the mechanisms behind trends in sap flow rates, and to use this technique to measure accurately the water use of young tea, including measurements at varying drought stress levels and under different environmental conditions.
5.
Conclusions
This was the first attempt to use sap flow meters for determining the water use of young tea in a tea growing area in Tanzania. The transpiration rates by clone AHP S15/10 remained high and stable in the afternoon compared with those by clone BBK35 with declining values as the day progressed beyond 1700 h. The water use of young tea clone AHP S15/10 at plant densities D1 and D6 was higher than the corresponding values for clone BBK35. In the third year after field planting, clone AHP S15/10 had the largest ETc to ETo ratio (0.98) at the lowest density treatment (D1) compared with a value of 0.53 at the highest density (D6). The corresponding values for clone BBK35 were 0.85 and 0.65. Further research is required to quantify in detail the mechanisms behind the observed trends in sap flow rates, and to use the SHB technique to measure accurately the
231
water use of young tea in response to planting density at varying drought stress levels and under different environmental conditions.
Acknowledgements At the time of this experiment, Ngwazi Tea Research Unit (NTRU) (now Ngwazi Tea Research Station, NTRS) was funded by a consortium of tea growers with varying interests with the support of the Tanzanian Ministry of Agriculture and Cooperatives. Now NTRS is under the Tea Research Institute of Tanzania (TRIT), an NGO supported by the entire Tanzanian tea industry, the Government of Tanzania and willing donors. This work was part of a Ph.D. programme for the author at Cranfield University at Silsoe College, UK. I thank the Tanzanian tea industry, the British Council—Chevening Scholarships, my supervisors, Clare Souch, Hereward Corley and Wilson Ng’etich and all staff at NTRS for the encouragement and support, and Professor Bruno Ndunguru, TRIT Executive Director, for the permission to publish.
references
Baillie, I.C., Burton, R.G.O., 1993. Ngwazi Estate, Mufindi, Tanzania. Report on Land and Water Resources, with Special Reference to the Development of Irrigated Tea. Part I. Land and Soil Suitability. Soil Survey and Land Research Centre, Silsoe College, UK. Baker, J.M., van Bavel, C.H.M., 1987. Measurement of mass flow of water in the stems of herbaceous plants. Plant Cell Environ. 10, 777–782. Batho, P.F., 1993. Comparative measurements of sap flow and mass loss in a young apple tree. M.Sc. Thesis. Silsoe College, Cranfield Institute of Technology, UK. Batho, P.F., Weibel, F.P., Souch, C.A., 1994. Construction manual for a sap flow meter on the basis of the stem heat balance method, second ed. Internal Report. Silsoe College, Cranfield University, UK. Boersma, K., Weibel, F.P., 1995. An improved stem heat balance method using analogue heat control. Agricult. Forest Meteorol. 75 (1–3), 191–208. Burgess, P.J., 1992. Responses of tea clones to drought in Southern Tanzania. Unpublished Ph.D. Thesis. Silsoe College, Cranfield Institute of Technology, UK. Burgess, P.J., Kigalu, J.M., 1993. The rationale of a new plant density experiment. Ngwazi Tea Res. Unit Q. Rep. 11, 3–7. Burgess, P.J., Carr, M.K.V., 1996. Responses of young tea (Camellia sinensis) clones to drought and temperature. I. Yield and yield distribution. Exp. Agricult. 32, 357–372. Burgess, P.J., Carr, M.K.V., 1997. Responses of young tea (Camellia sinensis) clones to drought and temperature. 3. Shoot extension and development. Exp. Agricult. 33, 367–383. Cermak, J., Deml, J., Penka, M., 1973. A new method of sap flow determination in full-grown trees based on stem heat balance with direct electric heating of xylem. Biol. Plant. 18, 105–110. Cermak, J., Kucera, J., Zidek, V., 1984. Xylem water flow in a crack willow tree (Salix fragilis L.) in relation to diurnal changes in environment. Oecologia (Berlin) 64, 145–151. Dagg, M., 1967. Cashew nut production in Southern Tanzania. V. Water balance of cashew trees in relation to spacing. East Afr. Agricult. Forestry J. 33, 88–94.
232
agricultural water management 90 (2007) 224–232
Dagg, M., 1970. A study of the water use of tea in East Africa using a hydraulic lysimeter. Agricult. Meteorol. 7, 303–320. Dugas, W.A., 1990. Comparative measurement of stem flow and transpiration in cotton. Theor. Appl. Climatol. 42, 215–221. FAO-UNESCO, 1988. Map of the World, Revised Legend. FAO, Rome, Italy. Grime, V.L., Morrison, J.I.L., Simmonds, L.P., 1995. Including the heat storage term in sap flow measurements with the stem balance method. Agricult. Forest Meteorol. 74, 1–25. Guetierrez, M.V., Meinzer, F.C., 1994. Estimating water use and irrigation requirements of coffee in Hawaii. J. Am. Soc. Horticult. Sci. 119 (3), 652–657. Ham, J.M., Heilman, J.L., Lascano, R.J., 1990. Determination of soil water evaporation and transpiration from energy balance and stem flow measurements. Agricult. Forest Meteorol. 52, 287–301. Hess, T.M., 1997. DAILYET. A computer programme for calculating the daily reference evapotranspiration (ETo). Cranfield University, Silsoe College, UK (unpublished). Huber, H., 1932. Beobachtung und Messung pflanzlicher saftstrome. Berichte Deutsche Botanische gesellschaft 50, 89–109. Ishida, T., Campbell, G.S., Calissendorff, C., 1991. Improved heat balance for determining sap flow rate. Agricult. Forest Meteorol. 56, 35–48. Jarvis, P.G., 1985. Coupling of transpiration to the atmosphere in horticultural crops: the ‘‘omega factor’’. Octa. Horticult. 171, 187–205. Jarvis, P.G., McNaughton, K.G., 1986. Stomatal control of transpiration: scaling up from leaf to region. Adv. Ecol. Res. 15, 1–49. Jones, H.G., 1978. Modelling diurnal trends of leaf water potential in transpiring wheat. J. Appl. Ecol. 15, 613–626. Jones, H.G., 1992. Plants and Microclimate: A Quantitative Approach to Environmental Plant Physiology, second ed. Cambridge University Press, Cambridge, UK. Kigalu, J.M., 1997. Effects of planting density on the productivity and water use of young tea (Camellia sinensis L.) clones in Southern Tanzania. Unpublished Ph.D. Thesis. Silsoe College, Cranfield University, UK. Kigalu, J.M., Burgess, P.J., 1994. Effects of clone and plant density on the yield and crop cover of young tea. Ngwazi Tea Res. Unit Q. Rep. 18, 17–22. Kigalu, J.M., Nixon, D.J., Souch, C.A., 1995. Measurement of transpiration in young tea using sap flow meters. Ngwazi Tea Res. Unit Q. Rep. 20, 15–22. Loomis, R.S., Connor, D.J., 1992. Crop Ecology Productivity and Management in Agricultural Systems. Cambridge University Press, Cambridge, UK. McNaughton, K.G., Jarvis, P.G., 1983. Predicting effects of vegetation changes on transpiration and evaporation. In: Kozlowski, T.T. (Ed.), Water Deficits and Plant Growth, vol. 7. Academic Press, New York, pp. 1–47. Monteith, J.L., 1965. Evaporation and environment. Symp. Soc. Exp. Biol. 19, 205–234.
Pethiyagoda, V., Ragendram, N.S., 1965. The determination of leaf areas in tea. Tea Q. 36, 48–58. Ritchie, J.T., 1972. Model for predicting evaporation from a row crop with incomplete cover. Water Resour. Res. 8, 1204– 1213. Sakai, S., 1987. Studies on photosynthesis and dry matter production of tea plants. Bulletin of the National Research Institute of Tea. National Research Institute of Tea, Kasanaga Shizuoka, Japan. Sakuratani, T., 1981. A heat balance method for measuring water flux in the stem of intact plants. J. Agricult. Meteorol. 37, 9–17. Sakuratani, T., 1984. Improvement of the probe for measuring water flow rate in intact plants with the stem heat balance method. J. Agricult. Meteorol. 40, 273–277. Sakuratani, T., 1987. Studies on evapotranspiration from crops. 2. Separate estimation of transpiration and evaporation from a soybean field without water shortage. J. Agricult. Meteorol. 42 (4), 309–317. Smith, B.G., Stephens, W., Burgess, P.J., Carr, M.K.V., 1993. Effects of light, temperature, irrigation and fertiliser on photosynthetic rate in tea (Camellia sinensis). Exp. Agricult. 29, 291–306. Souch, C.A., 1996. Water use and productivity in three poplar clones. Ph.D. Thesis. School of Agriculture Food and Environment (SAFE), Cranfield University, Silsoe, UK. Squire, G.R., 1977. Seasonal changes in the photosynthesis of tea (Camellia sinensis L.). J. Appl. Ecol. 14, 303–306. Squire, G.R., Callander, B.A., 1981. Tea plantations. In: Kozlowski, T.T. (Ed.), Water Deficits and Plant Growth, vol. VI. Academic Press, London, pp. 471–510. Steinberg, S., van Bavel, C.H.M., McFarland, M.J., 1989. A gauge to measure mass flow rate of sap in stems and trunks of woody plants. J. Am. Soc. Horticult. Sci. 114 (3), 466–472. Stephens, W., Carr, M.K.V., 1989. A water stress index for tea. Exp. Agricult. 25, 545–558. Stephens, W., Carr, M.K.V., 1991a. Responses of tea (camellia sinensis) to irrigation and fertiliser. I. Yield. Exp. Agricult. 27, 177–191. Stephens, W., Carr, M.K.V., 1991b. Responses of tea (camellia sinensis) to irrigation and fertiliser. II. Water use. Exp. Agricult. 27, 193–210. Swanson, R.H., 1994. Significant historical developments in thermal methods for measuring sap flow in trees. Agricult. Forest Meteorol. 72, 113–132. Valancogne, C., Ameglio, L., Angelocci, L., Cruiziat, P., 1993. Evaluation of transpiration of apple trees and measurement of daily course of water flow within the main branches of walnut trees. In: Borghetti, M., Grace, J., Raschi, A. (Eds.), Water Transport in Plants under Climatic Stress. Cambridge University Press, Cambridge, UK, pp. 219–227. Weibel, F.P., Devos, J.A., 1994. Transpiration measurements on apple trees with an improved stem heat-balance method. Plant Soil 166 (2), 203–219.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/agwat
Effects of planting density on the productivity and water use of tea (Camellia sinensis L.) clones I. Measurement of water use in young tea using sap flow meters with a stem heat balance method Julius M. Kigalu * Tea Research Institute of Tanzania, Ngwazi Tea Research Station, P.O. Box 2177, Dar es Salaam, Tanzania
article info
abstract
Article history:
Sap flow meters based on the stem heat balance method were used to measure the mass
Accepted 12 March 2007
flow rates or water use in young potted tea (Camellia sinensis L.) plants of clones AHP S15/10
Published on line 4 May 2007
and BBK35. The meters were constructed on site and installed onto the stem or branch sections of field growing plants in an experiment originally designed to study the effects of
Keywords:
plant population density and drought on the productivity and water use of young tea clones.
Evapotranspiration
The objective of the study was to use the SHB method as a first attempt to use sap flow
Irrigation
meters for determining the water use of young tea growing in the field under well watered
Plant density
conditions in Tanzania. The results are reported and recommendation made for further
Sap flow
work on using the technique. # 2007 Elsevier B.V. All rights reserved.
Stem heat balance method Tea Transpiration Water use
1.
Introduction
Since early in the 1990s, the tea industry in Tanzania continues to expand with substantial new planted areas especially in the Southern Highlands where tea plantations are predominantly irrigated during the long dry seasons. However, despite the extensive tea field expansions there is yet little information on plant population density and the effect on crop productivity and water use of young tea. High density planting can increase initial yields of young tea, encouraging early establishment of full ground cover and minimizing water losses due to surface evaporation and runoff. Although sap flow meters have been used to measure the water use of many crop plant species adoption of this technology in tea is very limited.
There are significant developments of techniques to measure directly sap flow in plants in order to quantify with precision the total amount of water transpired (Swanson, 1994). Briefly, there are two methods for sap flow measurements: the heat pulse velocity (HPV) based on the determination of sap velocity within a stem section (Huber, 1932; Dugas, 1990) which is invasive and may be destructive to the plant, and the stem heat balance (SHB) which estimates mass flow rates (Cermak et al., 1973; Sakuratani, 1984; Baker and van Bavel, 1987; Steinberg et al., 1989; Ishida et al., 1991; Batho et al., 1994; Boersma and Weibel, 1995). In a detailed review, Swanson (1994) reported that in recent years, techniques have been developed to measure the sap flow rate in a number of woody plant species including herbaceous plants (Sakuratani, 1984) and poplar (Populus spp.) clones (Souch, 1996), providing
* Tel.: +255 22 2122033; fax: +255 22 2113838. E-mail addresses: [email protected], [email protected]. 0378-3774/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.agwat.2007.03.005
agricultural water management 90 (2007) 224–232
accurate quantification of the amount of water transpired. The main advantage of the SHB technique is that sap flow meters are non-destructive to the plants and have lower costs than other methods, including weighing lysimeters (Cermak et al., 1973, 1984; Sakuratani, 1981; Swanson, 1994). The SHB method uses a small heater wrapped around the plant stem or branch section to provide a heat input into the section. Heat storage in the section may also be significant on trees with stem diameters greater than 25 mm, but this is negligibly small at smaller diameters (Batho, 1993; Batho et al., 1994; Grime et al., 1995). Thus if a known amount of heat is supplied into a stem section and a constant amount of sap (water) flows through the section during the process of transpiration, the temperature of the sapwood will reach a steady value which is inversely proportional to the water flow (Cermak et al., 1984; Swanson, 1994; Sakuratani, 1984). Therefore, under ideal conditions the amount of heat energy carried upwards out of the stem section (by conduction) by the mass flow of water will be equal to the amount of heat energy input to the section (Cermak et al., 1984; Batho et al., 1994). The objective of the current study was to use the SHB method described in detail by Weibel and Devos (1994), Batho et al. (1994) and Kigalu et al. (1995) to measure sap flow or transpiration rates in young tea plants in the field.
2.
Materials and methods
2.1.
Description of the site, climate and soils
The field trials reported here were conducted at Ngwazi Tea Research Station, NTRS (latitude 88320 S, longitude 358100 E, altitude: 1840 m a.s.l.) in the Mufindi District in Southern Tanzania. A full description of the climate, the weather and soils of the site is provided by Stephens and Carr (1991a,b), Burgess (1992), Burgess and Carr (1996), Kigalu (1997). Briefly, the climate of the area can be divided into three main seasons based on rainfall and temperature. More than 95% of the annual rainfall, ranging from 800 to 1100 mm occurs between the end of November and May. The corresponding monthly mean air temperature is 16–19 8C. The dry season can be divided into two: cool (13–16 8C) from June to August and warm (16–19 8C) from September to November. The experimental area is on weathered soils classified as a Xanthic ferralsol (Baillie and Burton, 1993; FAO-UNESCO, 1988) with a kaolinitic sandy clay down to 0.15 m depth and friable clay below. The mean soil pH was 5.4 and 5.8 at 0.15 and 0.70 m depths, respectively, which is suitable for growing tea (pH 4.9– 5.6). The volumetric water content of the soil at a water potential of 10 kPa were low for clay soils and more typical sandy loam, corresponding to field capacity (FC; 10 kPa), increased with depth from 24.7% in the top 0.15 m to 32.5% at 1.80 m where the silt and clay content was higher. The available water content held between field capacity (FC; 10 kPa) and permanent wilting point (PWP; 1500 kPa) ranged from 110 to 122 mm m1 1 in the top 2 m of soil to 93 mm m1 at 2.70 m. Regarding management of the tea bushes in the main field experiment before the field sap flow trials, fertiliser was applied in February and July 1995 on a per unit area basis at the rate of 100 kg N ha1 as N:P2O5:K2O in the ratio of 2:1:2. Also, in
225
1995 foliar zinc oxide (ZnO) was also uniformly applied at the rate of 4.5 kg ZnO ha1 in three equal splits to correct for zinc deficiency. The experimental field was kept weed free by regular hand weeding and herbicide application. At NTRS, overhead sprinkler irrigation of young tea may start in mid May to the end of November/December in bad years when rains stop early and start late. Irrigation scheduling and calculation of the potential soil water deficit (SWD, mm) are based on the soil water balance Eq. (1) described in detail at the site (Stephens and Carr, 1989; Burgess, 1992): SWDi ¼ SWDi1 Ri Ii þ Di þ ETi
(1)
where SWDi1 is the soil deficit on the previous ((i 1)th) day, Ri the rainfall, Ii the irrigation applied, Di the drainage and ETi is the crop evapotranspiration all recorded on the ith day and measured in mm. The value of Ii on each date of irrigation was recorded in 56 catch cans which were spaced symmetrically across the experiment. Any irrigation or rainfall received when the soil was still at field capacity (i.e., SWD = 0) was assumed to drain instantly through the soil profile. Movement of surface water (run-off) between plots or treatments was assumed negligible as it was prevented by micro-catchments dug between the rows of tea bushes, and by the accumulated mulch and tea leaf liter. The value of ETi was calculated from the daily evaporation pan measurement (Epan, mm) multiplied by a crop factor Kc. The value of Epan measured at NTRS gives a good estimate of the water loss from a mature stand of tea with or close to complete ground cover, implying negligible evaporation from the soil surface (Stephens and Carr, 1991a,b; Burgess, 1992). It was assumed that losses due to transpiration were directly proportional to the ground cover by the canopy, and Kc was assumed to be equal to 1.0 on the day of irrigation due to soil evaporation losses (Dagg, 1970; Ritchie, 1972). The soil water deficit (SWD; mm), incident solar radiation (S; MJ m2 d1) and wind speed (u; m s1) were calculated daily from measurements recorded at a nearby NTRS meteorological weather station.
2.2.
Features of the sap flow meters
This study used the SHB technique described in detail by Weibel and Devos (1994) and Batho et al. (1994) for the first time in tea in Tanzania. Briefly, the sap flow meters were constructed on site based on the SHM method following the procedures described by Batho et al. (1994) as briefly illustrated in Fig. 1, representing a schematic diagram of a sap flow meter attached to a stem or branch section with symbols briefly described as follows. The sap flow rate (F; g s1) was calculated from the energy balance across the sap flow meter (Eq. (2)) which states that: F¼
½Pin Q cd Q r þ Q s ðCs dTsap Þ
(2)
where using the symbols illustrated in Fig. 1: Qf = [Pin Qcd Qr + Qs] is the amount of heat (W) transported in the moving sap at differential temperature dTsap given by dTsap = [dTa + dTb]/2 (8C), where dTa and dTb are temperature differentials of the sap up and down stream measured by thermocouples a and b, respectively. Pin is the heater power
226
agricultural water management 90 (2007) 224–232
dThc; 8C) and one combined heater/flux meter (Fig. 1). An analogue heater control unit was used to maintain the constant differential temperature between the heated stem segment and the unheated stem below the meter (dThc) at 2 8C. The thermocouples were taped directly around the bark of the stem of the tea plants with PTFE tape. The combined heater/flux meter provided the necessary heat to the sap and measured radial heat loss (dTr; 8C) (Fig. 1). Measurements of transpiration were made with the sap flow meters in selected clone density treatment combinations (Table 1) of well-irrigated tea in a field experiment (labelled N12).
2.3.
Fig. 1 – Schematic diagram of a sap flow meter showing the combined heater/fluxmeter and the thermocouples, which measured temperature differentials. The heater control unit maintained the constant temperature differential dThc at 2 8C. The temperature differentials (dTa and dTb) were used to calculate the components of heat balance in the stem segment as described in detail below (modified from Kigalu et al., 1995).
input (W). Qcd is the heat conduction loss along the stem up and down stream (W); Qcd = [(dTb dTa) Kst A]/dz, where A is stem section cross-sectional area (m2), Kst is the thermal conductivity constant of the stem sapwood (W m1 K1). In this study, the value of Kst was assumed to be 0.54 W m1 K1 (Baker and van Bavel, 1987; Sakuratani, 1981), and dz is the distance between thermocouples a and b (m). Qr is the radial heat conduction loss (W) = dTr Ksh, where dTr is the thermopile radial temperature differential (8C), Ksh is a constant called the sheath conductance of the gauge system (W K1) calculoss Cs/ lated from Ksh = [(Pin Qcd Qs) (weight 3600)] dTsap, where weight loss is derived from sap flow per hour (Qf). Assuming that Qs is negligibly small and that Qf = 0 at night (Sakuratani, 1981), then Ksh = [Pin Qcd Qs]/ dTr. Qs is the heat stored in the stem section (W); negligible for small stem (<30 mm) diameters. Therefore, if Qs is very small, obtain Qf = [Pin Qcd Qr] (W), and sap flow rate = F = [Pin Qcd Qr]/(Cs dTsap) (g s1) or F = Qf 3600/ 1 (dTsap Cs) (g h ). Cs is the specific heat capacity of the sap or water (4.186 J g1 8C1). dThc is the temperature differential between the heater and the stem section (8C). Each sap flow meter consisted of three pairs of thermocouples (to measure differential temperatures dTa, dTb and
Field measurements
The main field experiment described in detail by Burgess and Kigalu (1993) and Kigalu (1997) comprised of two broad-leaved clones of contrasting growth behaviour. These are clones AHP S15/10 and BBK35 with spreading and upright or erect habit, respectively. Both clones originated from Kenya and are of scientific and economic importance in East and Central Africa. These were planted in six plant population density treatments, ranging from the lowest (D1: 8333 plants ha1 at a spacing of 1.00 m 1.20 m; below the usual commercial density of 10,417 plants ha1 corresponding to a spacing of 0.80 m 1.20 m) to the highest density (D6: 83,333 plants ha1 at a spacing of 0.60 m 0.20 m). Irrigation was applied under seven varying treatments, ranging from non-irrigated (I0) to well or full irrigation (I6) conditions. In fully irrigated plots, the irrigation water amounts applied reduced the cumulative potential SWD to zero (i.e., the soil was reduced to field capacity) for each irrigation date. An initial trial was conducted for 2 days (1–2 March 1995) in potted plants to compare the trends in transpiration rates with time of the day between one plant of each of the two clones (AHP S15/10 and BBK35) concurrently. This was followed by field measurements of sap flow comprising of three series of trials. In the first trial (9–12 November 1995) four plants from each clone (AHP S15/10 and BBK35) were selected from neighbouring sub-plots of the same density (D2: at a spacing of 0.6 mm 1.33 m; 12,531 plants ha1). The second and third series were conducted from 22 to 24 November and from 4 to 6 December in 1995 to quantify the effects of planting density on the transpiration rates of fully irrigated (I6) plants of each of the clones AHP S15/10 (series two) and BBK35 (series three) at the lowest (D1) and highest (D6) densities, using four sap flow meters (i.e., four plants) at each density. Sap flow was adjusted by dividing the rate of sap flow (g h1) by the total area of the leaves (m2) on the stem or suitable branch on which the gauge was attached as described by Valancogne et al. (1993). The results were then adjusted using two methods. Firstly, they were multiplied by the product of the total leaf area (m2) of the whole plant and divided by the area for each plant (m2), i.e., the leaf area index (LAI) calculated as reported by Pethiyagoda and Ragendram (1965), to obtain transpiration per unit leaf area using Eq. (3): Tl ¼
X f i =LAi LAI n
ði ¼ 1; 2; . . . ; nÞ
(3)
227
agricultural water management 90 (2007) 224–232
Table 1 – Series trials and dates of measurements of daily transpiration (T) by (a) both clones AHP S15/10 and BBK35 at the same density treatment D2, and by (b) clone AHP S15/10 and (c) clone BBK35 at the same density treatments D1 and D6 Transpiration (T; mm day1)
Series trials and dates
Clone S15/10
AHP
ETo (mm day1)
Clone BBK35
Kc = (T/ETo)/C for canopy cover AHP S15/10
BBK35
(1) Both clones at density D2 9 November 1995 10 November 1995 11 November 1995 12 November 1995
2.1 2.3 3.1 3.3
1.9 2.1 2.7 2.6
5.9 6.5 6.5 6.1
0.43 0.42 0.57 0.65
0.66 0.66 0.87 0.89
Mean S. E. () (n = 4)
2.7 0.56
2.3 0.38
6.3 0.30
0.51 0.11
0.76 0.12
(2) Clone AHP S15/10 density 22 November 1995 23 November 1995 24 November 1995
D1 4.3 5.3 5.5
D6 2.2 3.3 3.6
5.6 6.1 6.6
D1 0.92 1.04 0.99
D6 0.41 0.57 0.58
Mean S. E () (n = 4)
5.0 0.65
3.1 0.71
6.1 0.50
0.98 0.06
0.53 0.09
(3) Clone BBK35 density 4 December 1995 5 December 1995a 6 December 1995a
D1
Mean S. E. () (n = 3)
D6 2.6 2.3 1.6
1.9 2.8 2.7
D1 5.2 5.4 2.9
D6 0.86 0.74 0.94
0.40 0.56 1.01
2.2 0.49
2.5 0.49
4.5 1.39
0.85 0.12
0.65 0.32
The daily reference evapotranspiration (ETo) was estimated from the Penman–Monteith equation (Monteith, 1965) using the DAILYET computer programme (Hess, 1997), and the corresponding calculated crop coefficients (Kc) adjusted for canopy cover from well-watered plants of both clones. The measurements were recorded on selected days in November/December 1995. The values of canopy cover (C; %) for trial series (a) were 84 and 48% for clones AHP S15/10 and BBK35 at plant density treatment D2; (b) 84 and 95% for clone AHP S15/10 at density D1 and D6; and (c) 58 and 92% for clone BBK35 at density treatments D1 and D6, respectively. a Very cloudy days; and first rain in the afternoon (14.5 mm) on 6th December 1995 associated with low values of short-wave solar radiation (10 MJ m2 day1), wind speed (1.6 m s1), T and ETo.
where Tl is the adjusted transpiration (g m2 leaf area h1), fi the measured sap flow (g h1), and LAi is the leaf area (m2) of ith plant. Secondly, the sap flow rates (g h1) were adjusted on a unit ground area basis by multiplying by the reciprocal of the area per plant (m2) associated with each gauge–plant combination (Ham et al., 1990; Sakuratani, 1987) as in Eq. (4): Td ¼
X ð f ri Þ i ; n
ði ¼ 1; 2; . . . ; nÞ
(4)
where Td is the transpiration (g m2 h1), and ri is the reciprocal of the area per plant (m2) associated with the gauge–plant ith combination. Measurements of crop ground cover of the selected plants were taken immediately before commencement of sap flow data recording of each trial using the method described in detail by Burgess (1992) and by Kigalu (1997), and as reported by Kigalu and Burgess (1994). The daily transpiration rates (T; g m2 day1) were estimated from the cumulative values of the rates derived from Eq. (5) over a 24 h period. The equivalent mass of a kilogram of water in a metre square of ground area was then expressed in millimetre (mm) depth so that transpiration could be given in mm day1. The results were then adjusted based on the mean crop ground cover (C; %) of the plants within the sub-plot, by dividing by the proportion of ground cover to obtain estimates of the daily crop evapo-
transpiration (ETc; mm day1) (Dagg, 1967, 1970; Guetierrez and Meinzer, 1994) as in Eq. (5), used by Dagg (1970) in tea: ETc ¼
T C
(5)
Transpiration rates were adjusted on a per unit ground area basis using Eqs. (3) and (4), allowing for calculations and comparisons of the daily crop evapotranspiration rates (ETc; mm day1; Eq. (5)) with the daily reference evapotranspiration (ETo; mm day1), estimated using the DAILYET programme (Hess, 1997). Daily evapotranspiration rates were calculated using the approach described by Dagg (1970, 1967) for determining monthly evapotranspiration rates (Et) from young tea crop, which varied from 40 to 115 mm per month, but were satisfactorily predicted (R2 = 85%), both in wet and dry months, by Eq. (6) (Dagg, 1970): Et ¼ Eo ½0:9a þ ð1 aÞ 0:9 n
(6)
where Eo is the Penman estimate of open-water evaporation, a the fraction of the soil covered by the crop at noon, and n is the fractional number of rain days per month. Dagg (1970, 1967) further reported the relationship between ETC and ETo adjusted for crop ground cover as in Eq. (7). ETc ¼ 0:9ETo C
(7)
228
agricultural water management 90 (2007) 224–232
Fig. 2 – Comparison of transpiration rate per unit leaf area of young potted tea plants of clones AHP S15/10 and BBK35 measured by sap flow meters during the period 1–2 March 1995. Mean of two plants from each clone.
where the ratio ETc/(ETo C) gave a reasonable estimate of the transpiration factor, or crop factor Kc, over the area covered by the canopy whilst they are not under serious moisture stress. The crop factor or coefficient (Kc) for plant cover was estimated from the ratio of ETc to the reference crop evapotranspiration (ETo; mm day1), calculated with the Penman–Monteith equation (Monteith, 1965) using the daily evapotranspiration (DAILYET) computer programme (Hess, 1997) developed at Cranfield University, Silsoe College, UK.
3.
Results
3.1.
Clone differences in sap flow
For the potted plants both clones had a similar pattern or trend of transpiration rates, but the rate of transpiration (g h1 m2 leaf area) of clone AHP S15/10 was larger than that of clone BBK35 for most part of the day especially from 1100 to 1600 local mean time when the difference was more than twice at times (Fig. 2). In November 1995 the average crop cover of plants of clones AHP S15/10 and BBK35 in the field were 84 and 48%, respectively, whilst the leaf area indices ranged from 1.3 to 1.9 (mean: 1.6) and from 2.1 to 2.9 (mean: 2.5) for clones BBK35 and AHP S15/10, respectively. During the field trials all four gauges worked properly and the trends of sap flow rates per unit leaf area during the day that were observed from the field trials are shown in Figs. 3–5, and were similar to the trend for the potted plants in March 1995 (Fig. 2). Adjusting the water loss (measured using a sap flow meter) from each plant on a per unit leaf area basis indicated that water use by clone AHP S15/10 was greater than that from clone BBK35 from about 08:00 until 17:00 h on all the 4 days of measurements, but only twice as large for very short periods (Fig. 3), but transpiration per unit leaf area for clone AHP S15/10 was greater than that from clone BBK35.
3.2.
Density differences
The water use per unit leaf area by clone AHP S15/10 at the lowest plant density treatment (D1; with mean crop cover: 84%) was consistently larger than that at the highest population density (D6; mean crop cover: 95%), both from
Fig. 3 – Comparison of transpiration rate per unit leaf area of tea plants of clones AHP S15/10 and BBK35 measured at the same density D2 in the field using sap flow meters during the period 9–12 November 1995 (n = 4 for both clones AHP S15/10 and BBK35). The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
agricultural water management 90 (2007) 224–232
229
Fig. 4 – Comparison of transpiration rate per unit leaf area of fully irrigated (I6) tea plants of clone AHP S15/10 at density treatments D1 and D6 measured in the field using sap flow meters, 21–24 November 1995. The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
fully irrigated (I6) sub-plots (Fig. 4). The plants at the lowest density (D1) started transpiring earlier and continued longer than those at highest density (D6). Sap flow increased sharply from zero at around 07:00 h, and remained constant during the day before declining from 15:00 h onwards. Shade effects from tall trees below the experimental area might have been the cause for the changes in diurnal patterns of transpiration rates of clone AHP S15/10 in the highest density (D6). Fig. 5 shows the water use per unit leaf area by clone BBK35 at the same densities (D1 and D6, with mean crop cover of 58 and 92%, respectively) and irrigation treatments (I6). These results can be compared with those of clone AHP S15/10 (Fig. 4). The results were of a similar trend but the sap flow rate per unit leaf area for plants under the lowest density (D1) was larger
than under the highest (D6). However, contrary to clone AHP S15/10, on each of the 3 days of measurements, transpiration of clone BBK35 began at about 08:00 h and increased rapidly to peak values at 10:00–12:00 h before declining in a more or less constant way from then onwards. Although the measurements were recorded on different days, the absolute rates of transpiration per unit leaf area from clone BBK35 were about two thirds of those from clone AHP S15/10. The water use by the plants of clone AHP S15/10 (range: 2.1– 3.3 mm day1; mean: 2.7 0.56 mm day1) during the first series of measurements was consistently larger than that by clone BBK35 (range: 1.9–2.6 mm day1; mean: 2.3 0.38 mm day1), taken at the same time from the same density D2 (Table 1). Clone AHP S15/10 had relatively large Kc values (mean:
230
agricultural water management 90 (2007) 224–232
Fig. 5 – Comparison of transpiration rate per unit leaf area of fully irrigated (I6) tea plants of clone BBK35 at density treatments D1 and D6 measured in the field using sap flow meters, 4–6 December 1995. The daily incident short-wave solar radiation (S; MJ mS2 dayS1), the potential soil water deficit (SWD; mm) and the wind speed (u; m sS1), at a height of 2 m above the surface of a well-watered short healthy grass, measured from the meteorological station are also shown.
0.98 and 0.53 for D1 and D6, respectively), adjusted for crop cover, compared to the corresponding values of 0.85 and 0.65 for clone BBK35. In Figs. 3–5 the soil water deficit values that are given were estimated from a nearby meteorological station. In future further research is required to quantify transpiration rates and the corresponding actual soil water deficits in the field to establish possible differences in SWD between plots within the trial resulting from differences in transpiration rate between clones and densities.
4.
Discussion
4.1.
Differences between clones
The findings presented in Table 1 and Figs. 3–5 indicated that the transpiration rates (T; mm day1) and the water use by young plants of clone AHP S15/10 were larger than those by clone BBK35 on all the days at all the densities (D1, D2 and D6) investigated in the three consecutive series of field trials. Using sap flow meters, based on the stem heat balance method, for the first time in tea in Tanzania, it was shown that
the crop coefficient Kc, derived from the ratio of ETc to ETo and adjusted for plant canopy cover, for young tea clones AHP S15/ 10 and BBK35 at the same plant density (D2) was 0.51 and 0.76, respectively.
4.2.
Density differences
The average crop coefficient (Kc) value, adjusted for plant canopy cover, for clone AHP S15/10 was 0.98 ( 0.06), ranging from 0.92 to 1.04 for the plants at the lowest density treatment D1, and 0.53 0.09 (range: 0.41–0.58) at the highest (D6). The corresponding values for clone BBK35 were 0.85 0.12 (range: 0.86–0.94) and 0.65 0.32 (range: 0.40–1.01). The fraction of radiation intercepted by the plants varies with solar angle (Dagg, 1970) and cloud cover especially in the morning and evenings, and this may have effects on the transpiration rate. Clone AHP S15/10, having a spreading growth habit, had a larger transpiring total leaf area (leaf area index) than clone BBK35. Unexpected shade could have some influence could have some influence on the diurnal patterns of transpiration rate for clone AHP S15/10 at density D6. However, high rates were observed in the afternoon for clone AHP S15/ 10, having the larger crop cover, at the three density
agricultural water management 90 (2007) 224–232
treatments (D1, D2 and D6), whilst the values for clone BBK35 declined earlier during the day, probably suggesting clonal differences in the mechanisms by which each clone avoids plant water deficits (Jones, 1992; Loomis and Connor, 1992).
4.3.
Mechanisms
Transpiration is influenced indirectly by xylem leaf water potential through mediation of stomatal aperture of plants (Jones, 1978, 1992). However, the relationships between sap flow and incident short-wave radiation, vapour pressure deficit and stomatal conductance (McNaughton and Jarvis, 1983; Jarvis, 1985; Jarvis and McNaughton, 1986), xylem water potential (Jones, 1978; Loomis and Connor, 1992) were not investigated in the present study. The effects of light, temperature and photosynthetic demand rates (Squire, 1977; Sakai, 1987; Smith et al., 1993; Burgess and Carr, 1997) were not covered. No doubt, these require detailed quantification during subsequent sap flow measurements in tea using this technique. These authors, however, provided a mechanism which can enable scaling up transpiration from leaf to region similar to the work by Guetierrez and Meinzer (1994). This study demonstrated that the water use per unit leaf area by young tea varied with clone (being larger by clone AHP S15/10 than clone BBK35), and was density dependent. The values were consistently larger at the lowest density treatment (D1) than at the highest (D6) for both clones under similar environmental conditions. Compared with results from previous studies, the Kc values from clones AHP S15/10 and BBK35 at the lowest (D1) density treatment (Table 1) compared very closely with those reported by Squire and Callander (1981) for seedling tea with almost full ground cover, and also by Dagg (1967) for cashew trees in Southern Tanzania, respectively. However, further research is required to quantify in detail the mechanisms behind trends in sap flow rates, and to use this technique to measure accurately the water use of young tea, including measurements at varying drought stress levels and under different environmental conditions.
5.
Conclusions
This was the first attempt to use sap flow meters for determining the water use of young tea in a tea growing area in Tanzania. The transpiration rates by clone AHP S15/10 remained high and stable in the afternoon compared with those by clone BBK35 with declining values as the day progressed beyond 1700 h. The water use of young tea clone AHP S15/10 at plant densities D1 and D6 was higher than the corresponding values for clone BBK35. In the third year after field planting, clone AHP S15/10 had the largest ETc to ETo ratio (0.98) at the lowest density treatment (D1) compared with a value of 0.53 at the highest density (D6). The corresponding values for clone BBK35 were 0.85 and 0.65. Further research is required to quantify in detail the mechanisms behind the observed trends in sap flow rates, and to use the SHB technique to measure accurately the
231
water use of young tea in response to planting density at varying drought stress levels and under different environmental conditions.
Acknowledgements At the time of this experiment, Ngwazi Tea Research Unit (NTRU) (now Ngwazi Tea Research Station, NTRS) was funded by a consortium of tea growers with varying interests with the support of the Tanzanian Ministry of Agriculture and Cooperatives. Now NTRS is under the Tea Research Institute of Tanzania (TRIT), an NGO supported by the entire Tanzanian tea industry, the Government of Tanzania and willing donors. This work was part of a Ph.D. programme for the author at Cranfield University at Silsoe College, UK. I thank the Tanzanian tea industry, the British Council—Chevening Scholarships, my supervisors, Clare Souch, Hereward Corley and Wilson Ng’etich and all staff at NTRS for the encouragement and support, and Professor Bruno Ndunguru, TRIT Executive Director, for the permission to publish.
references
Baillie, I.C., Burton, R.G.O., 1993. Ngwazi Estate, Mufindi, Tanzania. Report on Land and Water Resources, with Special Reference to the Development of Irrigated Tea. Part I. Land and Soil Suitability. Soil Survey and Land Research Centre, Silsoe College, UK. Baker, J.M., van Bavel, C.H.M., 1987. Measurement of mass flow of water in the stems of herbaceous plants. Plant Cell Environ. 10, 777–782. Batho, P.F., 1993. Comparative measurements of sap flow and mass loss in a young apple tree. M.Sc. Thesis. Silsoe College, Cranfield Institute of Technology, UK. Batho, P.F., Weibel, F.P., Souch, C.A., 1994. Construction manual for a sap flow meter on the basis of the stem heat balance method, second ed. Internal Report. Silsoe College, Cranfield University, UK. Boersma, K., Weibel, F.P., 1995. An improved stem heat balance method using analogue heat control. Agricult. Forest Meteorol. 75 (1–3), 191–208. Burgess, P.J., 1992. Responses of tea clones to drought in Southern Tanzania. Unpublished Ph.D. Thesis. Silsoe College, Cranfield Institute of Technology, UK. Burgess, P.J., Kigalu, J.M., 1993. The rationale of a new plant density experiment. Ngwazi Tea Res. Unit Q. Rep. 11, 3–7. Burgess, P.J., Carr, M.K.V., 1996. Responses of young tea (Camellia sinensis) clones to drought and temperature. I. Yield and yield distribution. Exp. Agricult. 32, 357–372. Burgess, P.J., Carr, M.K.V., 1997. Responses of young tea (Camellia sinensis) clones to drought and temperature. 3. Shoot extension and development. Exp. Agricult. 33, 367–383. Cermak, J., Deml, J., Penka, M., 1973. A new method of sap flow determination in full-grown trees based on stem heat balance with direct electric heating of xylem. Biol. Plant. 18, 105–110. Cermak, J., Kucera, J., Zidek, V., 1984. Xylem water flow in a crack willow tree (Salix fragilis L.) in relation to diurnal changes in environment. Oecologia (Berlin) 64, 145–151. Dagg, M., 1967. Cashew nut production in Southern Tanzania. V. Water balance of cashew trees in relation to spacing. East Afr. Agricult. Forestry J. 33, 88–94.
232
agricultural water management 90 (2007) 224–232
Dagg, M., 1970. A study of the water use of tea in East Africa using a hydraulic lysimeter. Agricult. Meteorol. 7, 303–320. Dugas, W.A., 1990. Comparative measurement of stem flow and transpiration in cotton. Theor. Appl. Climatol. 42, 215–221. FAO-UNESCO, 1988. Map of the World, Revised Legend. FAO, Rome, Italy. Grime, V.L., Morrison, J.I.L., Simmonds, L.P., 1995. Including the heat storage term in sap flow measurements with the stem balance method. Agricult. Forest Meteorol. 74, 1–25. Guetierrez, M.V., Meinzer, F.C., 1994. Estimating water use and irrigation requirements of coffee in Hawaii. J. Am. Soc. Horticult. Sci. 119 (3), 652–657. Ham, J.M., Heilman, J.L., Lascano, R.J., 1990. Determination of soil water evaporation and transpiration from energy balance and stem flow measurements. Agricult. Forest Meteorol. 52, 287–301. Hess, T.M., 1997. DAILYET. A computer programme for calculating the daily reference evapotranspiration (ETo). Cranfield University, Silsoe College, UK (unpublished). Huber, H., 1932. Beobachtung und Messung pflanzlicher saftstrome. Berichte Deutsche Botanische gesellschaft 50, 89–109. Ishida, T., Campbell, G.S., Calissendorff, C., 1991. Improved heat balance for determining sap flow rate. Agricult. Forest Meteorol. 56, 35–48. Jarvis, P.G., 1985. Coupling of transpiration to the atmosphere in horticultural crops: the ‘‘omega factor’’. Octa. Horticult. 171, 187–205. Jarvis, P.G., McNaughton, K.G., 1986. Stomatal control of transpiration: scaling up from leaf to region. Adv. Ecol. Res. 15, 1–49. Jones, H.G., 1978. Modelling diurnal trends of leaf water potential in transpiring wheat. J. Appl. Ecol. 15, 613–626. Jones, H.G., 1992. Plants and Microclimate: A Quantitative Approach to Environmental Plant Physiology, second ed. Cambridge University Press, Cambridge, UK. Kigalu, J.M., 1997. Effects of planting density on the productivity and water use of young tea (Camellia sinensis L.) clones in Southern Tanzania. Unpublished Ph.D. Thesis. Silsoe College, Cranfield University, UK. Kigalu, J.M., Burgess, P.J., 1994. Effects of clone and plant density on the yield and crop cover of young tea. Ngwazi Tea Res. Unit Q. Rep. 18, 17–22. Kigalu, J.M., Nixon, D.J., Souch, C.A., 1995. Measurement of transpiration in young tea using sap flow meters. Ngwazi Tea Res. Unit Q. Rep. 20, 15–22. Loomis, R.S., Connor, D.J., 1992. Crop Ecology Productivity and Management in Agricultural Systems. Cambridge University Press, Cambridge, UK. McNaughton, K.G., Jarvis, P.G., 1983. Predicting effects of vegetation changes on transpiration and evaporation. In: Kozlowski, T.T. (Ed.), Water Deficits and Plant Growth, vol. 7. Academic Press, New York, pp. 1–47. Monteith, J.L., 1965. Evaporation and environment. Symp. Soc. Exp. Biol. 19, 205–234.
Pethiyagoda, V., Ragendram, N.S., 1965. The determination of leaf areas in tea. Tea Q. 36, 48–58. Ritchie, J.T., 1972. Model for predicting evaporation from a row crop with incomplete cover. Water Resour. Res. 8, 1204– 1213. Sakai, S., 1987. Studies on photosynthesis and dry matter production of tea plants. Bulletin of the National Research Institute of Tea. National Research Institute of Tea, Kasanaga Shizuoka, Japan. Sakuratani, T., 1981. A heat balance method for measuring water flux in the stem of intact plants. J. Agricult. Meteorol. 37, 9–17. Sakuratani, T., 1984. Improvement of the probe for measuring water flow rate in intact plants with the stem heat balance method. J. Agricult. Meteorol. 40, 273–277. Sakuratani, T., 1987. Studies on evapotranspiration from crops. 2. Separate estimation of transpiration and evaporation from a soybean field without water shortage. J. Agricult. Meteorol. 42 (4), 309–317. Smith, B.G., Stephens, W., Burgess, P.J., Carr, M.K.V., 1993. Effects of light, temperature, irrigation and fertiliser on photosynthetic rate in tea (Camellia sinensis). Exp. Agricult. 29, 291–306. Souch, C.A., 1996. Water use and productivity in three poplar clones. Ph.D. Thesis. School of Agriculture Food and Environment (SAFE), Cranfield University, Silsoe, UK. Squire, G.R., 1977. Seasonal changes in the photosynthesis of tea (Camellia sinensis L.). J. Appl. Ecol. 14, 303–306. Squire, G.R., Callander, B.A., 1981. Tea plantations. In: Kozlowski, T.T. (Ed.), Water Deficits and Plant Growth, vol. VI. Academic Press, London, pp. 471–510. Steinberg, S., van Bavel, C.H.M., McFarland, M.J., 1989. A gauge to measure mass flow rate of sap in stems and trunks of woody plants. J. Am. Soc. Horticult. Sci. 114 (3), 466–472. Stephens, W., Carr, M.K.V., 1989. A water stress index for tea. Exp. Agricult. 25, 545–558. Stephens, W., Carr, M.K.V., 1991a. Responses of tea (camellia sinensis) to irrigation and fertiliser. I. Yield. Exp. Agricult. 27, 177–191. Stephens, W., Carr, M.K.V., 1991b. Responses of tea (camellia sinensis) to irrigation and fertiliser. II. Water use. Exp. Agricult. 27, 193–210. Swanson, R.H., 1994. Significant historical developments in thermal methods for measuring sap flow in trees. Agricult. Forest Meteorol. 72, 113–132. Valancogne, C., Ameglio, L., Angelocci, L., Cruiziat, P., 1993. Evaluation of transpiration of apple trees and measurement of daily course of water flow within the main branches of walnut trees. In: Borghetti, M., Grace, J., Raschi, A. (Eds.), Water Transport in Plants under Climatic Stress. Cambridge University Press, Cambridge, UK, pp. 219–227. Weibel, F.P., Devos, J.A., 1994. Transpiration measurements on apple trees with an improved stem heat-balance method. Plant Soil 166 (2), 203–219.