Bridging Thermal Infrared Sensing and Physically‐Based Evapotranspiration Modeling: From Theoretical Implementation to Validation Across an Aridity Gradient in Australian Ecosystems

Abstract

Thermal infrared sensing of evapotranspiration (E) through surface energy balance (SEB) models is challenging due to uncertainties in determining the aerodynamic conductance (g A ) and due to inequalities between radiometric (T R ) and aerodynamic temperatures (T 0 ). We evaluated a novel analytical model, the Surface Temperature Initiated Closure (STIC1.2), that physically integrates T R observations into a combined Penman-Monteith Shuttleworth-Wallace (PM-SW) framework for directly estimating E, and overcoming the uncertainties associated with T 0 and g A determination. An evaluation of STIC1.2 against high temporal frequency SEB flux measurements across an aridity gradient in Australia revealed a systematic error of 10-52% in E from mesic to arid ecosystem, and low systematic error in sensible heat fluxes (H) (12-25%) in all ecosystems. Uncertainty in T R versus moisture availability relationship, stationarity assumption in surface emissivity, and SEB closure corrections in E were predominantly responsible for systematic E errors in arid and semi-arid ecosystems. A discrete correlation (r) of the model errors with observed soil moisture variance (r 5 0.33-0.43), evaporative index (r 5 0.77-0.90), and climatological dryness (r 5 0.60-0.77) explained a strong association between ecohydrological extremes and T R in determining the error structure of STIC1.2 predicted fluxes. Being independent of any leaf-scale biophysical parameterization, the model might be an important value addition in working group (WG2) of the Australian Energy and Water Exchange (OzEWEX) research initiative which focuses on observations to evaluate and compare biophysical models of energy and water cycle components.Plain Language Summary Evapotranspiration modeling and mapping in arid and semi-arid ecosystems are uncertain due to empirical approximation of surface and atmospheric conductances. Here we demonstrate the performance of a fully analytical model which is independent of any leaf-scale empirical parameterization of the conductances and can be potentially used for continental scale mapping of ecosystem water use as well as water stress using thermal remote sensing satellite data.

Challenges and Future Perspectives of Multi-/Hyperspectral Thermal Infrared Remote Sensing for Crop Water-Stress Detection: A Review

“…To circumvent such problem, a non-parametric and calibration-free ET estimation approach has been developed: the STIC model (Surface Temperature Initiated Closure, [124][125][126][127][128]) for simultaneously estimating ET, H, g A and canopy-surface conductance (g S ), surface moisture status, and ET components (evaporation, E, and transpiration, T) from the data itself. The STIC formulation is based on physical integration of LST into the Penman-Monteith (PM) equation to find analytical solutions to g A , g S , and the aerodynamic temperature (T 0 ) thereby obtaining a "closure" of the SEB.…”

“…Some of the fundamental challenges in ET mapping through SEB models have evolved due to, (i) the inequality between LST and the aerodynamic temperature (T 0 ), which is essentially responsible for the exchanges of H [124,125,148,149]; (ii) a non-unique relationship between T 0 and LST due to differences between the roughness lengths (i.e., effective source/sink heights) for momentum (z 0M ) and heat (z 0H ) within vegetation canopy and substrate complex [124,[150][151][152]; (iii) the unavailability of a universally agreed model to estimate spatially explicit T 0 [125,153]; (iv) the lack of a physical-based or analytical g A model as a true representative for spatial application [124,125]; (v) complexities in g A parameterization to accommodate the differences between the scalar roughness lengths of heat (z 0H ) and momentum (z 0M ) transfer. Such parameterizations are significantly affected by land-surface heterogeneity and could cause the ET severely overestimated in the semi-arid or arid ecosystems.…”

“…A new ET model, SPARSE [130,158], is particularly designed for ET mapping in water-scarce ecosystems and tends to overcome the T 0 vs. LST inequality challenges. The non-parametric STIC ET model [125][126][127][128], has also been developed to circumvent the challenges of scalability of empirical conductance functions for satellite-based ET mapping, and is the only model that estimates spatially explicit ET without any biophysical parameterization of the conductances [124]. For successful application of SEB models in ET mapping, it is not only important to have an LST component in the model, but, another crucial importance is to have an ensemble ET from a range of LST-driven models to obtain the best ET estimate for respective crop, soil, and hydrometeorological conditions.…”

An evapotranspiration model self-calibrated from remotely sensed surface soil moisture, land surface temperature and vegetation cover fraction: application to disaggregated SMOS and MODIS data

“…5a and b), one notes that the use of in situ data generates more scatter than with satellite data. The apparent scatter in retrieved r ss could be interpreted by the impact of the daily cycle of meteorological (evaporative demand) conditions or soil property differences (Merlin et al, 2011;Merlin et al, 2016;Merlin et al, 2018). The retrieved soil parameters also vary from year to year: the standard deviation is 0.39 and 1.69 for a rss and b rss , respectively.…”

“…This can be explained by the compensation effects linking a rss and b rss parameters which prove the empirical nature of the r ss . Another major issue that can lead to these differences is the depth of SM measurements (Merlin et al, 2011). In Sellers et al (1992, the near-surface soil moisture is defined in the 0-5 cm soil layer, whereas in our field, SM measurements are made at 5 cm depth.…”