ALBERT

Notes: Reliable models are required to assess the impacts of climate change on forest ecosystems. Precise and independent data are essential to assess this accuracy. The flux measurements collected by the EUROFLUX project over a wide range of forest types and climatic regions in Europe allow a critical testing of the process-based models which were developed in the LTEEF project. The ECOCRAFT project complements this with a wealth of independent plant physiological measurements. Thus, it was aimed in this study to test six process-based forest growth models against the flux measurements of six European forest types, taking advantage of a large database with plant physiological parameters.The reliability of both the flux data and parameter values itself was not under discussion in this study. The data provided by the researchers of the EUROFLUX sites, possibly with local corrections, were used with a minor gap-filling procedure to avoid the loss of many days with observations.The model performance is discussed based on their accuracy, generality and realism. Accuracy was evaluated based on the goodness-of-fit with observed values of daily net ecosystem exchange, gross primary production and ecosystem respiration (gC m−2 d−1), and transpiration (kg H2O m−2 d−1). Moreover, accuracy was also evaluated based on systematic and unsystematic errors. Generality was characterized by the applicability of the models to different European forest ecosystems. Reality was evaluated by comparing the modelled and observed responses of gross primary production, ecosystem respiration to radiation and temperature. The results indicated that: Accuracy. All models showed similar high correlation with the measured carbon flux data, and also low systematic and unsystematic prediction errors at one or more sites of flux measurements. The results were similar in the case of several models when the water fluxes were considered. Most models fulfilled the criteria of sufficient accuracy for the ability to predict the carbon and water exchange between forests and the atmosphere. Generality. Three models of six could be applied for both deciduous and coniferous forests. Furthermore, four models were applied both for boreal and temperate conditions. However, no severe water-limited conditions were encountered, and no year-to-year variability could be tested. Realism. Most models fulfil the criterion of realism that the relationships between the modelled phenomena (carbon and water exchange) and environment are described causally. Again several of the models were able to reproduce the responses of measurable variables such as gross primary production (GPP), ecosystem respiration and transpiration to environmental driving factors such as radiation and temperature. Stomatalconductance appears to be the most critical process causing differences in predicted fluxes of carbon and water between those models that accurately describe the annual totals of GPP, ecosystem respiration and transpiration.As a conclusion, several process-based models are available that produce accurate estimates of carbon and water fluxes at several forest sites of Europe. This considerable accuracy fulfils one requirement of models to be able to predict the impacts of climate change on the carbon balance of European forests. However, the generality of the models should be further evaluated by expanding the range of testing over both time and space. In addition, differences in behaviour between models at the process level indicate requirement of further model testing, with special emphasis on modelling stomatal conductance realistically.

Notes: Observations of ecosystem net carbon dioxide exchange obtained with eddy covariance techniques over a 4-year period at spruce, beech and pine forest sites were used to derive time series data for gross primary production (GPP) and ecosystem respiration (Reco). A detailed canopy gas exchange model (PROXELNEE) was inverted at half-hour time step to estimate seasonal changes in carboxylation capacity and light utilization efficiency of the forest canopies. The parameter estimates were then used further to develop a time-dependent modifier of physiological activity in the daily time step gas exchange model of Chen et al. (1999), previously used for regional simulations in BOREAS. The daily model was run under a variety of assumptions and the results emphasize the need in future analyses: (1) to focus on time-dependent seasonal changes in canopy physiology as well as in leaf area index, (2) to compare time courses of physiological change in different habitats in terms of recognizable cardinal points in the seasonal course, and (3) to develop methods for utilizing information on seasonal changes in physiology in regional and continental carbon budget simulations. The results suggest that the daily model with appropriate seasonal adjustments for physiological process regulation should be an efficient tool for use in conjunction with remote sensing for regional evaluation of global change scenarios.

Notes: Summary The surface energy exchange of 12m high Scots pine plantation at Hartheim, Germany, was measured with a variety of methods during a 11-day period of fine weather in mid-May 1992. Net radiation and rate of thermal storage were measured with conventional net radiometers, soil heat flux discs and temperature-based storage models. The turbulent fluxes discussed in this report were obtained with an interchanging Bowen ratio energy budget system (BREB, at 14 m), two one-propeller eddy correlation systems (OPEC systems 1 and 2 at 17m), a 1-dimensional sonic eddy correlation system (SEC system 3) at 15 m, all on one “low” tower, and a 3-dimensional sonic eddy correlation system (SEC system 22) at 22 m on the “high” tower that was about 46 m distant. All systems measured sensible and latent heat (H and LE) directly, except for OPEC systems 1 and 2 which estimated LE as a residual term in the surface energy balance. Closure of turbulent fluxes from the two SEC systems was around 80% for daytime and 30% for night, with closure of 1-dimensional SEC system 3 exceeding that of 3-dimensional SEC system 22. The night measurements of turbulent fluxes contained considerable uncertainty, especially with the BREB system where measured gradients often yielded erroneous fluxes due to problems inherent in the method (i.e., computational instability as Bowen's ratio approaches −1). Also, both eddy correlation system designs (OPEC and SEC) appeared to underestimate |H| during stable conditions at night. In addition, both sonic systems (1- and 3-dimensional) underestimated |LE| during stable conditions. The underestimate of |H| at night generated residual estimates of OPEC LE containing a “phantom dew” error that erroneously decreased daily LE totals by about 10 percent. These special night problems are circumvented here by comparing results for daytime periods only, rather than for full days. To summarize, turbulent fluxes on the low tower from OPEC system 2 and the adjacent SEC system 3 were in reasonable agreement, while the BREB system appeared to overestimate H and underestimate LE; H and LE measured by SEC system 22 on the high tower were lower than from OPEC and SEC3 on the low tower. The turbulent flux measurements tended to converge, but the data exhibit unexplained differences between days, between systems, and between locations.

Notes: Summary A single layer (Penman-Monteith) and a two layer (modified Shuttleworth-Wallace) evapotranspiration (ET) model are used alternatively to derive conductances related to the dominant fluxes of water vapor from a semi-closed Scots pine plantation. The derivations are based on micrometeorological measurements of above canopy energy flux densities and a simple resistance network. For a period of consecutive fine weather days, below canopy net radiation and below canopy ET were about 20 percent of the corresponding above canopy values. Resulting conductances for latent heat flux agreed well with porometric measurements of pines and understory scaled to canopy level. The shift from single to two layer modelling reduced the canopy conductance to pine conductance by the fraction of understory ET. However, characteristics of porometer results and micrometeorologically derived conductances were quite different: The porometer estimates of conductance were highly variable due to stomatal response to local environmental conditions or “natural” variability within the tree canopy and vegetation patches which characterized the forest understory. Micrometeorologically derived conductances integrate spatially resulting in relatively smooth and repetitive daily patterns that lack the information of small scale variability. This is seen as a favorable feature of micrometeorological derived conductances when used for the parameterization of atmospheric models for climate research as long as small scale bio-diversity is irrelevant.

Notes: Summary During the Hartheim Experiment (HartX) 1992 conducted in the Upper Rhine Valley, Germany, we estimated water vapor flux from the understory and the forest floor by several methods. At the vegetation “patch” level, direct estimates were made with small weighing lysimeters, and water loss was scaled-up to the stand level based on vegetation “patchtype” distribution. At the leaf level, transpiration flux was determined with a CO2/H2O porometer for the dominant understory plant species,Brachypodium pinnatum, Carex alba, andCarex flacca. Measured leaf transpiration was scaled-up to patch level with a canopy light interception and leaf gas exchange model, and then to stand level as in the case of lysimeter data, but with further consideration of patchtype leaf area index (LAI). On two days, total understory latent heat flux was estimated by eddy correlation methods below the tree canopy. The understory vegetation was subdivided into five major patch-types which covered 62% of the ground area and resulted in a cumulative LAI of approx. 1.54 when averaged over total stand ground area and compared to the average tree canopy LAI of 2.8. The remaining 38% of ground area was unvegetated bare soil and/or covered by moss (mainly byScleropodium purum) or litter. The evapotranspiration from the understory and unvegetated areas equaled approx. 20% of total forest stand transpiration during the HartX period. The understory vegetation transpired about 0.4 mm d−1 (13%) estimated over the period of May 13 to 21, whereas evaporation from moss and soil patches amounted 0.23 mm d−1 (7.0%). On dry, sunny days, total water vapor flux below the tree canopy exceeded 0.66 mm d−1. Using the transpiration rates derived from the GAS-FLUX model together with estimates of evaporation from moss and soil areas and a modified application of the Penman-Monteith equation, the average daily maximum conductance of the understory and the forest floor was 1.7 mm s−1 as compared to 5.5 mm s−1 for the tree canopy.

Notes: Summary Soybean evapotranspiration (ET) measured by the Bowen ratio energy balance method is used as the basis for the comparison of a commonly used semi-empirical (combination equation) and a pure statistical approach. Surface conductance within the combination equation was fitted to measured conductance by the Bowen ratio energy balance method. A multiple linear regression to vapor pressure deficit, solar radiation and wind speed was found. The same parameters were used to predict soybean evapotranspiration through a direct multiple linear regression. Despite a statistically rather poor forecast of surface conductance during 1987 (64% of variance explained) both evapotranspiration estimates agreed well with measured evapotranspiration (r 2 = 0.94). However, for a second year (1991) both, the combination equation and the pure statistical approach based on 1987 statistical relationships yielded only 70% of actual evapotranspiration. ActualET was 70% and 80% of potentialET in 1987 and 1991, respectively. These similar relationships to potentialET and the statistically weak prediction of surface conductance indicate poor physiological control of soybeanET or inadequate parametrization.

Notes: Summary In May 1992 during the interdisciplinary measurement campaign HartX (Hartheim eXperiment), several independent estimates of stand water vapor flux were compared at a 12-m high Scots pine (Pinus silvestris) plantation on a flat fluvial terrace of the Rhine close to Freiburg, Germany. Weather during the HartX period was characterized by ten consecutive clear days with exceptionally high input of available energy for this time of year and with a slowly shifting diurnal pattern in atmospheric variables like vapor pressure deficit. Methods utilized to quantify components of stand water flux included porometry measurements on understory graminoid leaves and on pine needles and three different techniques for determining individual tree xylem sap flow. Micrometeorological methods included eddy covariance and eddy covariance energy balance techniques with six independent systems on two towers separated by 40 m. Additionally, Bowen ratio energy balance estimates of water flux were conducted and measurements of the gradients in water vapor, CO2, and trace gases within and above the stand were carried out with an additional, portable 30 m high telescoping mast. Biologically-based estimates of overstory transpiration were obtained by up-scaling tree sap flow rates to stand level via cumulative sapwood area. Tree transpiration contributed between 2.2 and 2.6 mm/day to ET for a tree leaf area index (LAI) of 2.8. The pine stand had an understory dominated by sedge and grass species with overall average LAI of 1.5. Mechanistic canopy gas exchange models that quantify both water vapor and CO2 exchange were applied to both understory and tree needle ecosystem compartments. Thus, the transpiration by graminoid species was estimated at approximately 20% of total stand ET. The modelled estimates for understory contribution to stand water flux compared well with micrometeorologically-based determinations. Maximum carbon gain was estimated from the canopy models at approximately 425 mmol/(m2day) for the tree needles and at 100 mmol/(m2day) for the understory. Carbon gain was suggested by the modelling analysis to remain relatively constant during the HartX period, while water use efficiency in carbon fixation increased with decreasing vapor pressure deficit. Biologically- and micrometeorologically-based estimates of stand water flux showed good general agreement with variation of up to 20% that reflects both errors due to the inherent assumptions associated with different methods as well as natural spatial variability in fluxes. The various methods support a reliable estimate of average ET from this homogeneous canopy during HartX of about 2.6 mm/day (a maximum of about 3.1 mm/day) with an insignificant decreasing trend in correlation with decreasing vapor pressure deficit and possibly soil moisture. Findings during HartX were embedded in local scale heterogeneity with greater roughness over the forest and much higher ET over the surrounding agricultural fields which results in weak but clearly existant circulation patterns. A variety of measurements were continued after the HartX campaign. They allow us to extend our findings for six months with changing environmental conditions, including shortage of soil moisture. Hydrological estimates of soil water extractions and micrometeorological estimates of ET by the one-propeller eddy covariance (OPEC) system were in very good agreement, supporting the use of this robust eddy covariance energy balance technique for long-term monitoring.