A Correction to this paper has been published: https://doi.org/10.1038/s41597-021-00851-9.

Although global variation in actual evapotranspiration has been widely investigated, it remains unclear how its two major components, transpiration and soil evaporation, are driven by climate drivers across global land surface. This paper uses a well‐validated, process‐based model that estimates transpiration and soil evaporation, and for the first time investigates and quantifies how the main global drivers, associated to vegetation process and the water and energy cycle, drive the spatiotemporal variation of the two components. The results show that transpiration and soil evaporation dominate the variance of actual evapotranspiration in wet and dry regions, respectively. Dry southern hemisphere from 13°S to 27°S is highlighted since it contributes to 21% global soil evaporation variance, with only 11% global land area. In wet regions, particularly in the humid tropics, there are strong correlations between transpiration, actual evapotranspiration, and potential evapotranspiration, with precipitation playing a relatively minor role, and available radiative energy is the major contributor to the interannual variability in transpiration and actual evapotranspiration in Amazonia. Conversely in dry regions, there are strong correlations between soil evaporation, actual evapotranspiration, and precipitation. Our findings highlight that ecohydrological links are highly related to climate regimes, and the small region such as Australia has important contribution to interannual variation in global soil evaporation and evapotranspiration, and anthropogenic activities strongly influence the variances in irrigation regions.

OzFlux is the regional Australian and New Zealand flux tower network that aims to provide a continental-scale national research facility to monitor and assess trends, and improve predictions, of Australia’s terrestrial biosphere and climate. This paper describes the evolution, design, and current status of OzFlux as well as provides an overview of data processing. We analyse measurements from all sites within the Australian portion of the OzFlux network and two sites from New Zealand. The response of the Australian biomes to climate was largely consistent with global studies except that Australian systems had a lower ecosystem water-use efficiency. Australian semi-arid/arid ecosystems are important because of their huge extent (70 %) and they have evolved with common moisture limitations. We also found that Australian ecosystems had a similar radiationuse efficiency per unit leaf area compared to global values that indicates a convergence toward a similar biochemical efficiency. The two New Zealand sites represented extremes in productivity for a moist temperate climate zone, with the grazed dairy farm site having the highest GPP of any OzFlux site (2620 gC m−2 yr−1) and the natural raised peat bog site having a very low GPP (820 gC m−2 yr−1). The paper discusses the utility of the flux data and the synergies between flux, remote sensing, and modelling. Lastly, the paper looks ahead at the future direction of the network and concludes that there has been a substantial contribution by OzFlux, and considerable opportunities remain to further advance our understanding of ecosystem response to disturbances, including drought, fire, land-use and land-cover change, land management, and climate change, which are relevant both nationally and internationally. It is suggested that a synergistic approach is required to address all of the spatial, ecological, human, and cultural challenges of managing the delicately balanced ecosystems in Australasia.

Appendix A lists example of Earth observation applications in World Bank projects. Appendix B lists and briefly describes a selective selection of water information systems that prove notable examples of the integration of ground observations, EO data, and models.

Climate and physiological controls of vegetation gross primary production (GPP) vary in space and time. In many ecosystems, GPP is primary limited by absorbed photosynthetically-active radiation; in others by canopy conductance. These controls further vary in importance over daily to seasonal time scales. We propose a simple but effective conceptual model that estimates GPP as the lesser of a conductance-limited (Fc) and radiation-limited (Fr) assimilation rate. Fc is estimated from canopy conductance while Fr is estimated using a light use efficiency model. Both can be related to vegetation properties observed by optical remote sensing. The model has only two fitting parameters: maximum light use efficiency, and the minimum achieved ratio of internal to external CO2 concentration. The two parameters were estimated using data from 16 eddy covariance flux towers for six major biomes including both energy- and water-limited ecosystems. Evaluation of model estimates with flux tower-derived GPP compared favourably to that of more complex models, for fluxes averaged; per day (r2=0.72, root mean square error, RMSE=2.48μmolCm2s−1, relative percentage error, RPE=−11%), over 8-day periods (r2=0.78 RMSE=2.09μmolCm2s−1,RPE=−10%), over months (r2=0.79, RMSE=1.93μmolCm2s−1, RPE=−9%) and over years (r2=0.54, RMSE=1.62μmolCm2s−1, RPE=−9%). Using the model we estimated global GPP of 107PgCy−1 for 2000–2011. This value is within the range reported by other GPP models and the spatial and inter-annual patterns compared favourably. The main advantages of the proposed model are its simplicity, avoiding the use of uncertain biome- or land-cover class mapping, and inclusion of explicit coupling between GPP and plant transpiration.

The Cooperative Research Centre for Greenhouse Gas Technologies (CO2CRC) Otway Project is Australia's first demonstration of the geological storage of carbon dioxide (CO2), where about 65,000 metric tons of fluid consisting of 92% CO2 and 8% methane (CH4) by mass have been injected underground. As part of the project objective of developing methodologies to detect, locate, and quantify potential leakage of the stored fluid into the atmosphere, we formulate an inverse atmospheric model based on a Bayesian probabilistic framework coupled to a state-of-the-art backward Lagrangian particle dispersion model. A Markov chain Monte Carlo method is used for efficiently sampling the posterior probability distribution of the source parameters. Controlled experiments used to test the model involved releases of the injected fluid from one of the nearby wells and were staggered over 1 month. Atmospheric measurements of CO2 and CH4 concentrations were taken at two stations installed in an upwind-downwind configuration. Modeling both the emission rate and the source location using the concentration measurements from only two stations is difficult, but the fact that the emission rate was constant, which is not an unrealistic scenario for potential geological leakage, allows us to compute both parameters. The modeled source parameters compare reasonably well with the actual values, with the CH4 tracer constraining the source better than CO2, largely as a result of its 6 times higher signal-to-noise ratio. The results lend confidence in the ability of atmospheric techniques to quantify potential leakage from CO2 storage as well as other source types.

We compared estimates of actual evapotranspiration (ET) produced with six different vegetation measures derived from the MODerate resolution Imaging Spectroradiometer (MODIS) and three contrasting estimation approaches using measurements from eddy covariance flux towers at 16 FLUXNET sites located over six different land cover types. The aim was to assess optimal approaches in using optical remote sensing to estimate ET. The first two approaches directly regressed various MODIS vegetation indices (VIs) and products such as leaf area index (LAI) and fraction of photosynthetically active radiation (fPAR) with ET and evaporative fraction (EF). In the third approach, the Penman–Monteith (PM) equation was inverted to obtain surface conductance (Gs), for dry plant canopies. The Gs values were then regressed against the MODIS data products and used to parameterize the PM equation for retrievals of ET. Jack-Knife cross-validation was used to evaluate the various regression models against observed ET. The PM-Gs approach provided the lowest root mean square error (RMSE), and highest determination coefficients (R2) across all sites, with an average RMSE=38Wm−2 and R2=0.72. Direct regressions of observed ET against the VIs resulted in an average RMSE=60Wm−2 and R2=0.22, while the EF regressions an average RMSE=42Wm−2 and R2=0.64. The MODIS LAI and fPAR product produced the poorest estimates of ET (RMSE>44Wm−2 and R2<0.6); while the VIs each performed best for some of the land cover types. The enhanced vegetation index (EVI) produced the best ET estimates for evergreen needleleaf forest (RMSE=28.4Wm−2, R2=0.66). The normalized difference vegetation index (NDVI) best estimated ET in grassland (RMSE=23.8Wm−2 and R2=0.68), cropland (RMSE=29.2Wm−2 and R2=0.86) and woody savannas (RMSE=25.4Wm−2 and R2=0.82), while the VI-based crop coefficient (Kc) yielded the best estimates for evergreen and deciduous broadleaf forests (RMSE=27Wm−2 and R2=0.7 in both cases). Using the ensemble-average of ET as estimated using NDVI, EVI and Kc we computed global grids of dry canopy conductance (Gc) from which annual statistics were extracted to characterise different functional types. The resulting Gc values can be used to parameterize land surface models.

To understand the dynamics of ecosystem carbon cycling more than 10 years of eddy covariance data, measured over an evergreen, temperate, wet sclerophyll forest, were analysed and related to climate drivers on time scales ranging from hours to years. On hourly timescales we find that incoming shortwave radiation is the major meteorological driver of net ecosystem carbon exchange (NEE). Light use efficiency is higher under diffuse light conditions and carbon uptake is further modulated by the effects of variable and suboptimal temperatures (optimal temperature T-opt = 18 degrees C) as well as by water demand (critical vapour pressure deficit VPDcrit = 12 hPa). Incoming shortwave radiation is also the major driver on daily time scales. Effects of increased light use efficiency under diffuse conditions, however, are over-compensated by the increased carbon uptake with larger amounts of total incoming shortwave radiation under clear sky conditions. On synoptic time scales a low ratio of actual to potential incoming shortwave radiation is also related to a reduced carbon uptake, or carbon release, and associated with precipitation events. Overcast conditions during an extended wet period (2010-2011) led to lower than average carbon uptake as did extended dry periods during 2003 and 2006. The drought in 2003 triggered an insect attack which turned the ecosystem into a net source of carbon for almost one year. The annual average normalised difference vegetation index (NDVI) is highly correlated with NEE at this site and multiple linear regression shows that NDVI, incoming solar radiation and air temperature explain most of the variance in NEE (r(2) = 0.87, p< 0.001). Replacing air temperature with average spring air temperatures further increases the correlation (r(2) = 0.91, p < 0.001). Results demonstrate that carbon uptake in this ecosystem is highly dynamic, that wavelet analysis is a suitable tool to analyse the coherence between the carbon exchange and drivers seamlessly, and that long time series are needed to capture the variability. (C) 2013 Elsevier B.V. All rights reserved.

An adaptation of a simple model for evapotranspiration (E) estimations in drylands based on remotely sensed leaf area index and the Penman-Monteith equation (PML model) (Leuning et al., 2008) is presented. Three methods for improving the consideration of soil evaporation influence in total evapotranspiration estimates for these ecosystems are proposed. The original PML model considered evaporation as a constant fraction (f) of soil equilibrium evaporation. We propose an adaptation that considers f as a variable primarily related to soil water availability. In order to estimate daily f values, the first proposed method (f(SWC)) uses rescaled soil water content measurements, the second (f(Zhang)) uses the ratio of 16 days antecedent precipitation and soil equilibrium evaporation, and the third (f(drying)), includes a soil drying simulation factor for periods after a rainfall event. E estimates were validated using E measurements from eddy covariance systems located in two functionally different sparsely vegetated drylands sites: a littoral Mediterranean semiarid steppe and a dry-subhumid Mediterranean montane site. The method providing the best results in both areas was f(drying) (mean absolute error of 0.17 mm day(-1)) which was capable of reproducing the pulse-behavior characteristic of soil evaporation in drylands strongly linked to water availability. This proposed model adaptation, f(drying), improved the PML model performance in sparsely vegetated drylands where a more accurate consideration of soil evaporation is necessary.

A Bayesian inversion technique to determine the location and strength of trace gas emissions from a point source in open air is presented. It was tested using atmospheric measurements of N2O and CO2 released at known rates from a source located within an array of eight evenly spaced sampling points on a 20-m radius circle. The analysis requires knowledge of concentration enhancement downwind of the source and the normalized, three-dimensional distribution (shape) of concentration in the dispersion plume. The influence of varying background concentrations of ∼1% for N2O and ∼10% for CO2 was removed by subtracting upwind concentrations from those downwind of the source to yield only concentration enhancements. Continuous measurements of turbulent wind and temperature statistics were used to model the dispersion plume. The analysis localized the source to within 0.8 m of the true position and the emission rates were determined to better than 3% accuracy. This technique will be useful in assurance monitoring for geological storage of CO2 and for applications requiring knowledge of the location and rate of fugitive emissions.

A fundamental equation of eddy covariance (FQEC) is derived that allows the net ecosystem exchange (NEE) Ns¯ of a specified atmospheric constituent s to be measured with the constraint of conservation of any other atmospheric constituent (e.g. N2, argon, or dry air). It is shown that if the condition Ns¯≫χs¯NCO2¯ is true, the conservation of mass can be applied with the assumption of no net ecosystem source or sink of dry air and the FQEC is reduced to the following equation and its approximation for horizontally homogeneous mass fluxes:Ns¯=cd¯w′χ′s¯h+∫0hcd¯(z)∂χs∂t¯dz+∫0h[χs¯(z)−χs¯(h)]∂cd∂t¯dz≈cd¯(h)w′χ′s¯h+∫0h∂χs∂t¯dz.Here w is vertical velocity, c molar density, t time, h eddy flux measurement height, z vertical distance and χs = cs/cd molar mixing ratio relative to dry air. Subscripts s, d and CO2 are for the specified constituent, dry air and carbon dioxide, respectively. Primes and overbars refer to turbulent fluctuations and time averages, respectively. This equation and its approximation are derived for non-steady state conditions that build on the steady-state theory of Webb, Pearman and Leuning (WPL; Webb et al., 1980. Quart. J. R. Meteorol. Soc. 106, 85–100), theory that is widely used to calculate the eddy fluxes of CO2 and other trace gases. The original WPL constraint of no vertical flux of dry air across the EC measurement plane, which is valid only for steady-state conditions, is replaced with the requirement of no net ecosystem source or sink of dry air for non-steady state conditions. This replacement does not affect the ‘eddy flux’ term cd¯w′χ′s¯ but requires the change in storage to be calculated as the ‘effective change in storage’ as follows:∫0h∂cs∂t¯dz−χs¯(h)∫0h∂cd∂t¯dz=∫0hcd¯(z)∂χs∂t¯dz+∫0h[χs¯(z)−χs¯(h)]∂cd∂t¯dz≈cd¯(h)∫0h∂χs∂t¯dz.Without doing so, significant diurnal and seasonal biases may occur. We demonstrate that the effective change in storage can be estimated accurately with a properly designed profile of mixing ratio measurements made at multiple heights. However further simplification by using a single measurement at the EC instrumentation height is shown to produce substantial biases. It is emphasized that an adequately designed profile system for measuring the effective change in storage in proper units is as important as the eddy flux term for determining NEE. When the EC instrumentation measures densities rather than mixing ratios, it is necessary to use:Ns¯≈w′c′s¯h+χs¯w′c′v¯+c¯w′T′¯T¯h+cd¯(h)∫0h∂χs∂t¯dz.Here T is temperature and cv and c are the molar densities of water vapor and moist air, respectively. For some atmospheric gas species such as N2 and O2, the condition Ns¯≫χs¯NCO2¯ is not satisfied and additional information is needed in order to apply the EC technique with the constraint of conservation of dry air.

The 'energy imbalance problem' in micrometeorology arises because at most flux measurement sites the sum of eddy fluxes of sensible and latent heat (H + lambda E) is less than the available energy (A). Either eddy fluxes are underestimated or A is overestimated. Reasons for the imbalance are: (1) a failure to satisfy the fundamental assumption of one-dimensional transport that is necessary for measurements on a single tower to represent spatially-averaged fluxes to/from the underlying surface, and (2) measurement errors in eddy fluxes, net radiation and changes in energy storage in soils, air and biomass below the measurement height. Radiometer errors are unlikely to overestimate A significantly, but phase lags caused by incorrect estimates of the energy storage terms can explain why H + lambda E systematically underestimates A at half-hourly time scales. Energy closure is observed at only 8% of flux sites in the La Thuile dataset (http://www.fluxdata.org/DataInfo/default.aspx) with half-hourly averages but this increases to 45% of sites using 24 h averages because energy entering the soil, air and biomass in the morning is returned in the afternoon and evening. Unrealistically large and positive horizontal gradients in temperature and humidity are needed for advective flux divergences to explain the energy imbalance at half-hourly time scales. Imbalances between H + lambda E and A still occur in daily averages but the small residual energy imbalances are explicable by horizontal and vertical advective flux divergences. Systematic underestimates of the vertical heat flux also occur if horizontal u'T' covariances contaminate the vertical w'T' signal due to incorrect coordinate rotations. Closure of the energy balance is possible at half-hourly time scales by careful attention to all sources of measurement and data processing errors in the eddy covariance system and by accurate measurement of net radiation and every energy storage term needed to calculate available energy. (C) 2011 Elsevier B.V. All rights reserved.

Satellite and gridded meteorological data can be used to estimate evaporation (E) from land surfaces using simple diagnostic models. Two satellite datasets indicate a positive trend (first time derivative) in global available energy from 1983 to 2006, suggesting that positive trends in evaporation may occur in "wet" regions where energy supply limits evaporation. However, decadal trends in evaporation estimated from water balances of 110 wet catchments ((E) over bar (wb)) do not match trends in evaporation estimated using three alternative methods: 1) (E) over bar (MTE), a model-tree ensemble approach that uses statistical relationships between E measured across the global network of flux stations, meteorological drivers, and remotely sensed fraction of absorbed photosynthetically active radiation; 2) (E) over bar (Fu), a Budyko-style hydrometeorological model; and 3) (E) over bar (PML), the Penman-Monteith energy-balance equation coupled with a simple biophysical model for surface conductance. Key model inputs for the estimation of (E) over bar (Fu) and (E) over bar (PML) are remotely sensed radiation and gridded meteorological fields and it is concluded that these data are, as yet, not sufficiently accurate to explain trends in E for wet regions. This provides a significant challenge for satellite-based energy-balance methods. Trends in (E) over bar (wb) for 87 "dry" catchments are strongly correlated to trends in precipitation (R-2 = 0.85). These trends were best captured by (E) over bar (Fu), which explicitly includes precipitation and available energy as model inputs.

A fundamental equation of eddy covariance (FQEC) is derived that allows the net ecosystem exchange (NEE) (N-s) over bar of a specified atmospheric constituent s to be measured with the constraint of conservation of any other atmospheric constituent (e.g. N-2, argon, or dry air). It is shown that if the condition |(N-s) over bar| >> |(chi(s)) over bar||(N-CO2) over bar| is true, the conservation of mass can be applied with the assumption of no net ecosystem source or sink of dry air and the FQEC is reduced to the following equation and its approximation for horizontally homogeneous mass fluxes: (N-s) over bar = (C-d) over bar(w'chi(s)') over bar|(h) + integral(h)(0)(C-d) over bar (z)(partial derivative chi(s)) over bar/partial derivative tdz + integral(h)(0)[(chi(s)) over bar (z) - (chi(s)) over bar (h)](partial derivative C-d) over bar/partial derivative d dz approximate to (C-d) over bar (h) {(w'chi(s)') over bar|(h) + integral(h)(0) (partial derivative chi(s)) over bar/partial derivative t dz}. Here w is vertical velocity, c molar density, t time, h eddy flux measurement height, z vertical distance and chi(s) C-s/C-d molar mixing ratio relative to dry air. Subscripts s, d and CO2 are for the specified constituent, dry air and carbon dioxide, respectively. Primes and overbars refer to turbulent fluctuations and time averages, respectively. This equation and its approximation are derived for non-steady state conditions that build on the steady-state theory of Webb, Pearman and Leuning (WPI.; Webb et al., 1980. Quart. J. R. Meteorol. Soc. 106,85-100), theory that is widely used to calculate the eddy fluxes of CO2 and other trace gases. The original WPL constraint of no vertical flux of dry air across the EC measurement plane, which is valid only for steady-state conditions, is replaced with the requirement of no net ecosystem source or sink of dry air for non-steady state conditions. This replacement does not affect the 'eddy flux' term (C-d) over bar(w'chi(s)') over bar but requires the change in storage to be calculated as the 'effective change in storage' as follows: integral 0h (partial derivative C-s) over bar/partial derivative t dz - (chi(s)) over bar (h) integral(h)(0) (partial derivative C-d) over bar/partial derivative t dz = integral(h)(0) (C-d) over bar (z)(partial derivative chi(s)) over bar/partial derivative t dz + integral 0h [(chi(s)) over bar (h)](partial derivative C-d) over bar/partial derivative t dz approximate to (C-d) over bar (h) integral(h)(0)(partial derivative chi(s)) over bar/partial derivative t dz. Without doing so, significant diurnal and seasonal biases may occur. We demonstrate that the effective change in storage can be estimated accurately with a properly designed profile of mixing ratio measurements made at multiple heights. However further simplification by using a single measurement at the EC instrumentation height is shown to produce substantial biases. It is emphasized that an adequately designed profile system for measuring the effective change in storage in proper units is as important as the eddy flux term for determining NEE. (C) 2011 Elsevier B.V. All rights reserved. When the EC instrumentation measures densities rather than mixing ratios, it is necessary to use: (N-s) over bar approximate to (w'C-s') over bar|(h) + (chi(s)) over bar[(w'C-v') over bar+(c) over bar(w'T') over bar/(T) over bar](h) + (C-d) over bar (h) integral(h)(0)(partial derivative chi(s)) over bar/partial derivative t dz. Here T is temperature and C-v and c are the molar densities of water vapor and moist air, respectively. For some atmospheric gas species such as N-2 and O-2, the condition |(N-s) over bar >> |(chi(s)) over bar||(N-CO2) over bar is not satisfied and additional information is needed in order to apply the EC technique with the constraint of conservation of dry air. (C) 2011 Elsevier B.V. All rights reserved.