WATERUSE
Intensive Observation Period 1 - IOP1

Andria, Italy
July 22nd to August 3rd, 2003

olive grove in Southern Italy


The IOP1 took place in the olive grove at the test site of Andria (Southern Italy, Apulia, 41°12'N, 16°10'E, 175 m a.s.l.) from July 22nd to August 3rd, 2003, on approximately level terrain (2° inclined towards NNE). The average stand age was 100 years. The distance between the rows was 8.3 m and the tree density was 132 trees per ha, with an average tree height of 5 m. Between the trees was bare soil and the canopy was relatively open. Fetch in the main wind directions (W and N) was 200 to 300 m. It has been the first official experiment where most of the project partners were involved in a common field study, bringing in the field instruments and techniques specifically developed in the first year of the project, in order to quantify, characterize, and especially partition water budget components of a typical agricultural land use system (an irrigated olive grove in Southern Italy).


The main objective of the campaign was the concurrent deployment and use of various, integrated techniques to study water fluxes in the different segments of the Soil-Plant-Atmosphere System in a sparse yet regular tree canopy system, and understand regulative properties of the plants in partitioning water fluxes. Thus, extensive measurements of soil geo-pedological, physical and especially hydrological properties were planned at plot scale, and intensive monitoring of water content at significant points, too. Plant canopy structure was also to be investigated by several destructive and non-destructive techniques. Water flow in the trees was planned to be investigated by several sap-flow methods and by remote-sense techniques, integrated by direct stomatal conductance measurements. General validation of evapotraspiration loss to the atmosphere and eventually partitioning between transpiration and soil evaporation were planned to be gathered through analytical micrometeorological measurements, based on the eddy covariance technique and the measurements of the energy balance components.

A very extensive survey of soil electromagnetical properties was performed by Partner 8 in an area approximately 1 ha large positioned in the normal daytime footprint of the ET fluxes, using the Ground Penetrating Radar technique. It showed interesting features of the soil structure, especially related to the vertical profile, and spatial (horizontal) variability, which demonstrated to be very useful in addressing the sampling strategy of other soil measurements, especially geoelectrical and time-domain reflectometry. The vertical structure of the soil profile often showed thick hard plates of carbonates at 0.2-0.5 m, which could be crucial in determining the movement of water in the soil and the root distribution. Two trenches 1 m deep and several meters long have been dug out in areas where GPR had showed anomalous behaviour or strong gradients of soil electromagnetical properties, in order to better understand the resolution capacity of this technique.

Geolectrical methods based of measurement of electrical resistivity of soil between complex electrodes array configuration have been applied by Partner 6 in specific areas of the plot and around trees which were monitored also by other techniques (sap flow) . The inversion of these readings showed to be able to detect with good resolution the vertical stratification of the soil and even the presence of the bigger tree roots. Also, the parallel measurements of volumetric water content by time-domain reflectometry (TDR) probes, showed a good correlation between electrical resistivity values and local water content. Further validation of these relationships will be possible after the analysis of the extensive soil sampling made by Partner 9 at the same locations where geoelectrical methods have been applied. Partner 6 also studied infiltration properties of the soil, yielding interesting information on proper irrigation management.

Similar (geo)electrical techniques were applied to the study of the resistivity of trunk sections, demonstrating quite well the diameter of the heart wood. These measurements have been carried out in strict cooperation with Partner 5, on trees where sap flow measurements were also run at the same time.

Several techniques of sap-flow measurements were applied by Partners 1, 5, and 7, ranging from the simpler Granier-type sensor, to the heat-balance gauges, and the more complex heat field deformation technique developed by Partner 5. On some trees, all the techniques were applied concurrently, in order to better understand the relative performance of the methods in the very complex application under study. Indeed, the olive tree trunks showed to be very heterogeneous and variable, as demonstrated also by independent techniques (reflection of elastic waves), developed by Partner 8. However, a remarkable similarity between probes installed on different trees has been noticed, and this result could be encouraging for the application of these methods in the real world of agricultural water management.

       
                        Sap flow measurements with Granier method

Analytical information on the leaf and canopy energy balance were collected by Partner 4, who deployed instruments for monitoring leaf boundary layer conductance, soil heat flux, global radiation under the canopy and in open areas, air temperature and humidity and windspeed. Thermal images taken by ground camera of foliage kept at different transpiration rates (suppressed with vaseline, not altered, thoroughly wet), in connection with measurements of stomatal resistance provided by Partner 1, should allow the development of remote-sensing techniques for estimating the (evapo)transpiration fluxes. Unfortunately, it was not possible to acquire good Landsat images for the IOP because of poor weather at the time of the satellite pass (see below). These would have been needed for a full testing of upscaling.

Strong effort has been put on the geometrical and biometrical characterization of the olive grove. This activity, lead in the field by Partner 5, has been carried out with the help of Partners 1, 4, and 7. It provided detailed informations on canopy structure and average foliage density, both crucial parameters in interpreting and upscaling some of the sap-flow measurements.

General micrometeorological measurements were provided by Partner 7, together with his subcontractor MetInform and the team of the MCR-Lab of the University of Basel. These included: downward and upward fluxes of short and longwave radiation above and below the canopy, air temperature and humidity, atmospheric pressure, soil temperature at different depths and heat flux on several transects, precipitation. During the IOP these variables were recorded at 1 min interval. Special attention has been put on the measurement of turbulence statistics in the canopy and in the surface layers, deploying a vertical array of 7 sonic anemometers, from 0.5 to 12.0 m (i.e. more than 2 times the canopy height) above the ground, synchronously logged at 20 Hz. The top level was also equipped with a fast IR Gas Analyzer for water vapour and carbon dioxide to calculate eddy fluxes, while two Krypton fast hygrometers were installed at lower levels. The characterization of momentum transfer and turbulence in relation to canopy structure being the main purpose of such installation, it was also intended to yield informations on the build-up of sensible and latent heat fluxes with height and , therefore, to attack the problem of partioning between soil and plant fluxes.

 

Unfortunately, during the IOP the meteorological conditions got rapidly bad, because of an unusual low pressure area that persisted South, South-East of Apulia for several days, conveying humid air to the region and bringing several showers. Rainfall during July 25th, 26th and 27th severely compromised many of the planned experiments, especially those related to the soil physics and physiology of trees under water stress. Measurements and trials went on anyway as planned, despite the adverse weather, but on the 31st a further, very intense rainfall event (24.4 mm in the day, with peak rate of about 3 mm min-1) really flooded the field, provoking a very peculiar hydrological situation. First analysis of latent and sensible heat fluxes (Fig. 7. 3) shows a clear and steadily changing partitioning of sensible/latent heat fluxes, with a Bowen ratio increasing from about 0.5 (just after the heavy rainfall) to 7.3 (few days later). Partner 7 tried to extend as possible the measurements, with the hope of catching the steady evolution of fluxes in relation to soil water content, but instruments have to be dismantled on August the 6th.