WORKING PACKAGES (WP’s)

The detailed project description is broken into 9 workpackages (WP’s).
In each of these are presented the objectives, methodology and the main participants involved.

WP1 (SOIL) DEVELOPMENT OF HIGH-RESOLUTION GEOPHYSICAL AND HYDROLOGICAL TECHNIQUES FOR MONITORING WATER CONTENT AND WATER FLOW IN SOILS
WP2 (SAPFLOW) INNOVATIVE ASPECTS OF SAP FLOW MEASUREMENTS BY HFD METHOD
WP3 (ROOTS/STEM) IMPROVEMENTS OF NON DESTRUCTIVE TECHNIQUES (NDT) FOR ANALYSING STEM AND ROOT ZONE STRUCTURE AND WATER CONTENT
WP4 (TREES) INTEGRATED WHOLE TREES STUDIES: FUNCTIONAL AND ARCHITECTURAL RELATIONSHIPS
WP5 (IRT- REMOTE) SCALING-UP OF INFRARED THERMOGRAPHY FOR ESTIMATION OF EVAPOTRANSPIRATION
WP6 (FIELD) ESTABLISH INTEGRATED MULTIDISCIPLINARY FIELD SITES IN PORTUGAL AND ITALY
WP7 (LAND-WATERUSE) USE OF INTEGRATED MULTIDISCIPLINARY FIELD TRIALS TO STUDY/DESCRIBE PARTIONING OF WATER FLUXES IN HETEROGENEOUS AGRICULTURAL AND SEMI-NATURAL LAND USE SYSTEMS
WP8 (MODEL FRAMEWORK) DEVELOPMENT OF A FRAMEWORK FOR THE MODELLING OF WATER FLUXES IN HETEROGENEOUS LANDSCAPES
WP9 (USERS) TRANSFER TECHNOLOGY TO USERS AND PREPARATION OF REPORTS AND PUBLICATIONS

There was a need to develop new approaches and to improve the accuracy of existing techniques for monitoring soil physical properties, soil water content and soil water flows. In addition to theoretical advances, calibration and testing of new techniques for data acquisition, processing and interpretation requires a fully controlled and instrumented physical model system. The development and testing of new instruments and techniques demands, in addition to theoretical investigations and field applications, controlled measurements on a tank analog which simulates the subsurface geological situation of the problem under study. A first version of a full scale (open air) model, which is unique in its construction and functions, was constructed some years ago, but it suffered from some limitations so there was a need to construct a second, greatly improved, version. This model could then be used for infiltration experiments with various boundary conditions designed for critical testing of the developed arrays and theoretical techniques and for controlling data quality. Applied methods include the use of GPR, resistivity, tracers, time domain reflectometry (TDR) and tensiometer studies. This involved the development of hardware and software and modelling techniques, concentrating on opportunities for 3-D studies at high resolution. All instrument tests (also for other working groups, especially 3, 7) have been carried out searching for the optimum combination and resolution. We considered the erection of the tank model to be an essential prerequisite for optimising instrumentation for the field part of the project. The outcome has been an enhanced availability of Integrated techniques for measuring hydrological variables in soils and trunks. Availability of these integrated techniques allows standardisation and application for monitoring and controlling sustainable conditions for a wide range of applications.

It was proposed to concentrate on heat field deformation (HFD) sap-flux technique as recent research has shown that this has the most potential among those available (Nadezhdina et al. 1998) for accurately quantifiable measurements of sap-flow and is applicable to a wide range of plants from small shrubs to the largest trees, allowing good replication. This method does not require any calibration and in addition to measurement of high flows (as do other methods as THB, HD, HPV), is sensitive enough to measure very low resaturating night flows (very important for tree survival) and also reverse flows (e.g. during fog). HFD-method allows study of the pattern of sap flow which is a crucial parameter for integration of flow from measuring points to whole trees (cross section) and for further upscaling to whole stand.
The work involved (a) Evaluation of the theoretical basis of the technique, (b) Development and validation of the approach in a range of laboratory and field situations including: treatment of highly asymmetric crosssections, especially roots, small diameter stems (branches, roots and shrubs), and measurement and calculation of radial and vector flows (relevant for water storage and pathway analysis); (c) Development of methodology of recalculation of sap flow from single measuring point to the whole tree using knowledge of radial pattern of flow, important for up-scaling purposes, (d) Development of improved software for data manipulation and visualization of data in 3D and 4D images in case of sap flow measurements in different xylem depth of tree stems.
Testing has been conducted in the laboratory and in the field close to participant’s 5 lab. In addition techniques for determining the necessary biometric parameters (e.g. Cermak 1989; Cermak et al. 1998) for scaling up individual tree measurements to stands have been refined (see also WP3).
The technique and its modifications have been subsequently applied in the field in the multiuser experiments and compared against other flow measurements (link to WP 7).

WP3 (ROOTS/STEM) IMPROVEMENTS OF NON DESTRUCTIVE TECHNIQUES (NDT) FOR ANALYSING STEM AND ROOT ZONE STRUCTURE AND WATER CONTENT

Soil: Complementary studies have been conducted on the use of these techniques for 3-D and 2-D tomography (eg Hruska et al 1999) of the root zone to identify the root distribution and its variation in different tree and shrub systems. For DC-resistivity a new arrangement of spherical electrodes inserted into the soil using slant boring, combined with surface electrodes. For GPR (100 MHz - 1.5 GHz) different tomographic measurements were carried out along similar linear and circular arrays, with buried central electrodes used as reference reflectors, and the orientation of transmitter and receiver antennae adjusted to optimise resolution (Daniels and Bower 1998). The use of bore-hole GPR to aid 3-D structural investigations of the deeper parts roots zone has been investigated.
The main outcome of this WP was therefore enhanced availability of advanced techniques for study of water in the root zone and in trunks, which were used in subsequent parts of the project.

Methodology:

Techniques to study water flow and water contents in stems and in the root zone (WP2 and WP3) need integrating to provide an accurate picture of the water flow through whole plants and hence to provide accurate estimates of actual ET (Evapotranspiration). Particular problems are raised by the integration of daily fluxes using sap-flow technology, because of the flows into and out of storage and also to the changes in radial flow. The integration over 24-h periods has only rarely been attempted so far.
This workpackage therefore brought together studies aimed at describing the architecture of individual trees (leaf area and disposition, stem and branch hydraulic architecture, and root distribution) and the contributions of flows through each of these segments. Techniques to be combined are likely to include the integration of sap-flow measurement with anatomical description, root and shoot architecture studies [photography, GPR, root excavation (including the use of the novel ‘air-spade’ technique) and radiative transfer analysis).
It has been specified which part of crown (foliage) and root system (i.e. soil layer) is connected with certain xylem layer by long-term measuring sap flow along xylem radius and carrying out destructive (cutting branches and roots) and non-destructive (shading or mist-spraying of specific parts of crown, or the use of partial irrigation from different sides of stem and at different distances from stem) treatments. Flows have been compared with soil water content, branch illumination studies and soil, stem and leaf water potentials.
This way it has been also possible to estimate the impact of different pruning on sap flow pattern (which could be applied for evaluation of pruning practices).
Integration of information on tree (canopy, stem, root) structure with corresponding flows allows to estimate absorption of water from different soil layers, dynamics of this process and importance for tree survival.
To contribute to the overall objectives of the project, preliminary development of scaling-up of sap-flow and related tree-level methods to estimation of stands (Cermak & Kucera 1990; Cermak et al 1998) was considered (see also WP7).

A key requirement for hydrological and water balance studies is the availability of robust and easily applied techniques that can be used to estimate evaporation at a range of scales from the individual plant to 100s of m, through to km.
Many high-resolution techniques are limited in their applicability to larger scales. The recent innovation of using reference wet and dry surfaces to enable infra-red measurements of canopy temperature to be used to estimate the surface resistance (Jones 1999a) has extended the power of infra-red techniques. In principle, an extension of this theory can be used to estimate evaporation rates (Jones 1999a). The first step has been to complete the analysis and a sensitivity analysis of its application to field-collected thermal images has been conducted. Results were compared with alternative measurements of ET obtained in the field experiments (WP7, WP8) using small scale techniques and eddy correlation. The possibility of extending to the estimation of evaporation using Fuchs’ (1990) method for infra-red detection of plant stress from the variance of leaf temperatures has also been investigated in this phase.
In parallel with the ground-based studies, current approaches to the estimation of ET from thermal data from satellites were reviewed. A new approach based on the combination of thermal data with NDVI (Normalized Difference Vegetation Index) to correct thermal averages for the proportion of soil and vegetation in any pixel has been suggested by Moran and co-workers (1994, 1996) and the use of within image references (e.g. water bodies, Jones 1999a) has been developed to estimate ET. Satellites were chosen on the basis of availability of both thermal and red/NIR channels. High resolution images (30 m pixels) are available from the Landsat TM but even in the absence of cloud are at only c.16 day intervals, so these have been complemented by the more frequent NOAH AVHRR data at lower resolution (1.1 km). Images were analysed to estimate ET by these new and by more conventional algorithms, for the experimental sites in Italy and Portugal and compared with the surface estimates of ET. Particular effort was aimed at scaling up the field-scale thermal approaches and comparison with available techniques for treatment of satellite data. In synthesising these contrasting approaches there was a critical need and effort, for the lower resolution satellite data especially, on treatment of the problems of sub-pixel variability in strongly heterogeneous systems and the consequent problem of biased averages (e.g. Campbell 1996).

The project aimed to provide information relating to the water management in water-limited regions of Europe. We therefore considered it essential to obtain data for a range of contrasting agricultural or semi-natural agricultural ecosystems typical of these areas. We therefore studied contrasting crops including: intensive orchards (peach in Portugal), and less intensive systems of olive (Italy) and montado (cork - Portugal). These cover a range of vegetation structures and also compare similar systems in two different water-limited areas.
Sites have been chosen on the basis of extensive areas of relatively homogeneous vegetation, ready availability of necessary infrastructural support (field laboratories, power, access). Another important criterion was the interest of the landowners, since an important objective was the transfer of technology developed to farmers and agencies of the region. Initial work has been done on the two orchard sites, with the more challenging, olive/montado site being set up from the second year.
All these situations present important challenges to many of the presently available techniques for monitoring water use, due to their generally discontinuous and often heterogenous nature. Fields have been instrumented to gather basic meteorological information throughout most of the project lifetime, in order to provide WP8 and WP9 with continuous records of environmental inputs. These have included: air temperature and humidity, wind speed and direction, global solar radiation, soil temperature and precipitation.

Towers of proper height have been installed to mount eddy covariance instrumentation during the intensive observation periods (WP7). Satellite images of the areas have been obtained over the period for remote ET estimation.

WP7 (LAND-WATERUSE) USE OF INTEGRATED MULTIDISCIPLINARY FIELD TRIALS TO STUDY/DESCRIBE PARTIONING OF WATER FLUXES IN HETEROGENEOUS AGRICULTURAL AND SEMINATURAL LAND USE SYSTEMS

The underlying objective of the whole project is to provide information of use to water resource managers on water use by heterogeneous land use systems. This integrating study therefore aims to provide this key information on physiological processes including diurnal and seasonal dynamics of transpiration (sap flow rate) in relation to potential transpiration (estimated from potential evapotranspiration for experimental trees and stands) and their relationship to stand and environmental variables. Therefore specific objectives of WP7 are:

1. Field comparison of multiple techniques for measurement of water use
2. Provision of case-history data on water use in relation to land use and management as climatic factors
3. Provision of comparative data on water balance of contrasting types of land use (orchard, forest) from viewpoint of tree and waterrelations, level of drought stress and enlargement of tree vitality and functional stability.

This Workpackage brings together all participants in the central co-ordinated multidisciplinary series of experiments. Measurement of energy/water budgets in discontinuous and/or heterogeneous canopies is an ambitious task which the project addressed at a range of scales with different techniques both for technique intercomparison and for quantification of fluxes in the case study systems. During Intensive Observation Periods (IOP), a full set of instruments has been deployed in the fields prepared in WP6.
All techniques developed in WPs 1-5 soil and flux measurement have been deployed. In addition to continuous standard meteorological data, other instrumentation  included Kipp&Z CNR1 net radiometers, with particular attention to horizontal variability of radiative properties of canopies and energy fluxes, by positioning instruments at a proper height. Soil heat flux has been sampled at several locations to account for heterogeneity. Turbulent fluxes of sensible and latent heat have been measured by eddy covariance technique, using triaxial sonic anemometers (Campbell CSAT3 and Gill R3) and IR gas analyzers (LI-COR 6262 and 7500). Ancillary micrometerological measurements included air temperature profiles in the canopy layer and above at 10-20 Hz, synchronous with turbulence data, in order to assess aerodynamic coupling to the boundary layer. Bulk fluxes of evapotranspiration have been split into transpiration from plants and evaporation from soil surface, using measurements of whole-plant sap-flow transpiration data on the whole tree level for evaluation of conductance at the tree and stand levels. Leaf water potential, stomatal and boundary layer conductances (steady-state porometers and heat balance of facsimile leaves - using infra-red) have been addressd. These data has been used in WP8 to feed detailed, multidimensional models of water flow in the SPA continuum.

WP8 (MODEL FRAMEWORK) DEVELOPMENT OF A FRAMEWORK FOR THE MODELLING OF WATER FLUXES IN HETEROGENEOUS LANDSCAPES

An appropriate modeling approach for the water use of complex agricultural and agroforestry systems in water-limited areas of southern Europe needs to follow a consistent, step by step, scaling-up from simple process analyses procedure, to integrated areal analyses over large areas (usually catchments).
Therefore in this workpackage we have focused first on the understanding and modeling of key partial processes such as soil water movement, the water uptake by the root systems, transpiration and the evaporation. The modeling of such process is yet far from being satisfactorily solved for ecosystems showing a significant spatial heterogeneity (horizontally and vertically) in the vegetation distribution. Water fluxes were described using soil water storage capacity models as well as models based on Richards equation, which differentiate between different horizontal soil layers on the basis of soil and root distribution characteristics. Typical soil water extraction and redistribution patterns have been described. Possibilities of simplified modeling of ET have also been evaluated and tested against datasets obtained in WP7. Based on results, experience and state of the art (literature) shortcomings have been defined.
In a second step up-scaling to establish the water balance for a whole stand/orchard was involved. The database allows us to identify adequate approaches to simulate the water fluxes without loosing critical hydrological information identified at the single tree level.
WP8 therefore provides information relevant to the practical decisionmaking process in water resources management, contributing to more rational decisions, and more predictable and economic means of achieving. Input parameters had to be determined independently from the dataset used for validation - here the parallel studies in Italy and Portugal havel been used. In addition part of the results were used for calibration and a second part served as reference during validation. The degree of simplicity of the used models in relation to the reliability and accuracy of the forecasts required were considered. Results allow the definition of shortcomings in our knowledge and the identification of critical functional relationships to improve prediction robustness.