This Section describes the attempt to join together the most useful characteristics of complementary methods. Several case-studies are presented, where EC was used to calibrate the data obtained by Granier sap flow and soil lysimeters (Es). The EC system measured sensible heat flux density (H) and latent heat flux density (LE), the latter corresponding to stand ET. In all cases, corrections for air density variation (according to Webb et al., 1980) and for UV radiation absorption by oxygen (according to Tanner et al., 1993) were performed. EC data were always selected according to fetch and footprint analysis (Schuepp et al., 1990) and validated through spectral analysis (Anderson et al., 1986) and energy balance equation closure (Brunet et al., 1995). The closure error of the energy balance, was used in its simplest form (Rn - G vs. H + LE). LE obtained is ETec. In the following, TEC = ETEC - Es. Tec was compared to uncorrected SF data for the periods with both measurements. The relationship obtained allowed the estimation of long term T from SF data. Es, if any, was added to T to obtain long term ET. The experiments described in this Section were carried on for several woody crops and different locations in Portugal, between 1996 and 2004.
Two experimental plots were used. The first experiment (plot 1) took place at Tagus Valley (Central Portugal) near Santarem (latitude 39° 10' N, longitude 8° 43' W, elevation 5 m), Portugal. The climate, according Reis and Gon?alves (1981) is humid mesothermic with dry hot summer (Csa, Koppen classification), with a mean annual temperature of 16.2°C, mean annual rainfall of 707 mm and ETo (Piche) of 1436 mm year-1. The vineyard was planted in 1984 and the cultivar was "Trincadeira", grafted in SO4, with a plant density of 3030 vines per ha (3 m x 1.1 m). The area was 24 ha, the fetch around 600 m. The soil was a deep clay sandy loam soil (Flm, according to FAO classification). There was no irrigation. The measurements were taken between floraison and maturity, during 1996 and 1997. Leaf area index, measured at veraison, was 2.39 and 1.96 in 1996 and 1997, respectively. Ground cover, based on shadowed areas near solar noon, was 28 %.
The second plot was located at Setubal Peninsula (latitude 38° 35' N, longitude 8° 49' W, elevation 25 m), Portugal. The climate was similar to the first, with a mean annual temperature of 16,0°C, mean annual rainfall of 746 mm and ETo 1407 mm year-1. The vineyard, planted in 1990, was drip irrigated and the cultivar under study was "Syrah" grafted in 1103P, with a plant density of 2975 vines per ha (2.8 m x 1.2 m). The area was 216 ha and the fetch above 1000 m. The soil was sandy (ARh, according to FAO classification). The measurement period occurred between floraison and the end of the vegetative cycle. Leaf area index, measured only in 2002, was 1.32. Ground cover, based on shadowed areas near solar noon, was 18 %.
Eddy covariance data for selected periods were collected using a 1-D sonic anemometer with a fine wire thermocouple and a krypton hygrometer (respectively, models CA27, 127
and KH20 from, Campbell Scientific, Inc. Logan, UT, EUA) in 1996 and 1997 and with a 3D sonic anemometer and a krypton hygrometer (respectively, models CSAT3 and KH20 from, Campbell Scientific, Inc. Logan, UT, EUA) in 2001 and 2002. Rn and G were measured as described in 4.2.2.
SF was measured with Granier method (see Section 3.). The results shown, for the main plots, were obtained with sensors (UP GmbH, Landshut, Germany) installed in representative plants: 4 vines in 1996, 1997 and 2001 and in 6 vines in 2002. The probes (0.002 m diameter, 0.01 length) were inserted 0.1 m apart. Sap flow was calculated assuming that all sectional area was effective. This was verified with destructive measurements and also with the study of the radial flow profile, using the heat field deformation method. Corrections for the influence of natural thermal gradients in the trunk were made. Since there was a strong relationship between vine leaf area and SF for individual plants, vineyard SF was estimated using a four step procedure: (1) measurement of SF of individual vines, (2) determination of leaf area of these vines and calculation of SF on a leaf area basis, (3) determination of leaf area index (LAI) and (4) multiplication of stand LAI by SF per unit of leaf area. ETo was calculated according to Allen et al. (1998).
Surface energy balance showed that the sum of measured latent and sensible heat accounted for more than 95% of the available energy (Figure 4) and therefore the data gathered with the EC method were considered valid for the purpose of the study. Evapotranspiration measured by the EC method was in the range 1.0 and 4.1 mm/day in plot 1 (1996 and 1997) and between 1.6 and 2.4 mm/day in 2001 and 2002 (plot 2).
Figure 5 shows the relationship between ET = T (EC data) and sap flow for half-hour fluxes obtained during periods of negligible soil evaporation (1996). Similar relationships were found in 1997 and in plot 2 (2001 and 2002). A strong underestimation for the highest flow rates was verified. Also, an evidence of some capacitance effects in the late afternoon was found. For higher fluxes, strong variations in TEC occurred for small variations in SF. Therefore, it was difficult to adjust an equation to correct half-hourly fluxes.
Was this article helpful?