the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Feb 2025
Systematic overestimation of evapotranspiration over irrigated areas by an offline land surface model
Abstract. Offline Land Surface Models (LSMs) are essential for a wide range of applications, including water resource management and agricultural planning. A critical variable in these models is evapotranspiration, but its value is easily biased in irrigated areas. In fact, irrigation fundamentally alters local atmospheric conditions – cooling and humidifying the air and reducing wind speeds – factors that contribute to reducing evapotranspiration rates. This phenomenon is called "atmospheric feedback", but is often missing or poorly represented in offline LSM because most of the atmospheric forcings used, such as reanalyses and climate model outputs, overlook the atmospheric effect of irrigation. This leads to a tendency for offline LSM to overestimate evapotranspiration rates over irrigated areas. In this study, the atmospheric effects of irrigation are quantified using data from the LIAISE project field campaign. The various surface processes that influence the dynamics of evapotranspiration in response to the atmospheric feedback are then systematically investigated. The results confirm the importance of considering the atmospheric feedback in the ISBA LSM over irrigated areas in many configurations. For well irrigated crops, the average overestimation of evapotranspiration is about 25 %. Conversely, for water-stressed crops, this overestimation is mitigated because the timing of stomatal closure is influenced by atmospheric feedback mechanisms, providing a compensatory effect. These findings highlight the need for improved representation of irrigation-related atmospheric feedback in the atmospheric forcings used as upper boundary conditions in LSM to improve the accuracy of evapotranspiration estimates in agricultural or hydrological contexts.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(4647 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4647 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2024-3562', Nima Zafarmomen, 25 Feb 2025
- AC3: 'Reply on CC1', Tanguy Lunel, 27 Aug 2025
-
RC1: 'Comment on egusphere-2024-3562', Anonymous Referee #1, 08 May 2025
This paper describes a sensitivity study of a Land Surface Model (ISBA) under two atmospheric forcing conditions using coupled and offline LSM-BLM/RCM over two irrigated plots in the LIASE campaign: one considering the feedback of irrigation on the atmospheric temperature and humidity profiles and one ignoring it. The description of the numerical experiment to generate both forcing datasets as well as their analysis in terms of evapotranspiration overestimation and other meteorological variables when ignoring irrigation are given in QJRMS (https://doi.org/10.1002/qj.4736).
There are three main issues with the current manuscript:
- The importance of the impact of irrigation on the (changing) local/regional climate and the resulting areal evapotranspiration has been long recognized (cf. examples below); therefore, it is hard for the reader to identify the overall important message (to which it is referred lines 628-629), out of the obvious statement that irrigation affects the atmospheric conditions above the irrigated plot if the latter is large enough (wetter, cooler conditions) and should be considered when applying distributed hydrological or land surface models (statement which has been already put forward in the QJRMS paper) ?*
- The sensitivity study is not generic enough to carry substantial interest for the HESS community (a few days only, with specific conditions, forcings are all locally generated, no benchmarks from a reanalysis have been used); indeed, most LSM or hydrological model simulations can’t be coupled to regional climate models and are therefore offline simulations (that is, the vast majority of the research presented in journals like HESS); it would have been interesting to provide a solution to bypass the bias induced by ignoring the feedback effect on the atmosphere by the irrigated areas: should one use a simple approach such as the Bouchet Complementarity for the subareas where irrigation takes place ? (cf. the ample literature by Jozsef Szilagyi to develop an equilibrium Priestly-Taylor formalism with a “wet bulb” like information ?); in the last decade several authors have tried to use the Bouchet Complementarity to analyse the effect of irrigation on regional climate change; I can mention at least one work in Turkey (Ozdogan and Salvucci, 2002) but also the “Australian Pan Evapotranspiration Decrease Controversy”, for which several authors have analysed the cause of the decrease in pan ET data, first relating it to solar dimming (Roderick and Farquhar, 2002), but then realizing it is due mostly to the development of irrigation around the pans. I think that the literature review, which largely focuses on more recent works, should at least mention this line of research.
- The paper lacks clarity at multiple places: for instance, the analysis of the impact on the partitioning between evaporation E and transpiration T for various soil moisture levels focuses on a common stress index, SWI, and it is not clear how it affects E and T separately; only E(SWI) is shown, and it is not sure whether in Fig. 7 the same SWI levels impacts E and T. In that case, why not focusing on different moisture controls for E and T ? Also, what about advection from nearby drier areas in the simulation ? Is it dutifully accounted for (especially in those summer conditions) ?
In summary, in order to make this work fully compliant with the readership of HESS, I would recommend extending the paper by offering solutions to assess and correct for the ET overestimation at regional and seasonal scales, using for example larger scale simulations (ERA5Land-like) on the LIAISE area. I don’t really see what general outcomes one can draw from a few days of simulations over two plots, but maybe I missed out on an original sidelining interest of the paper.
Minor comments:
Line 33: “… is now fairly well understood and represented in LSMs”: I would dampen this optimistic statement. It is still fairly hard to go below 30% error.
Lines 138-144: This paragraph is curcial for the understanding of the paper, but lacks clarity and precision (e.g. “the surface features are not taken” etc).
Line 240: how does SWI also potentially affect transpiration ?
Lines 254-269 / 2.3: This echoes to my main comments 2 and 3: impact should be assessed at the seasonal scale, which is the scale of application of ET0; also, ET0 uses many assumptions that are only valid for a dense short grass fully irrigated, it is not meant to be considered in comparison with an output of an LSM outside of the numerous crop coefficient adjustments that are provided in the method to account for the “non standard” conditions (meteorological, moisture availability, type of the irrigation practice, plant development stage, etc), and typically not at the temporal scale of one single day.
Line 566: it seems to me that it has been already recognized in the LSM community that moisture-driven stress factor in the Jarvis formulation is less realistic than one based on leaf water potential.
Lines 570-578: here it would be helpful to have a discussion on what combinations of situations in the sensitivity are realistic in the given context (dry root zone vs wet surface soil moisture, no-irrig atmospheric conditions with high SWI etc etc).
Line 635: “a decrease in transpiration leads to an increase in bare soil evaporation”: this automatically the case if using a coupled model such as MEB because any drop in transpiration (resp. evaporation, for that matter) increases the temperature and decreases the humidity at the aerodynamic level and enhances evaporation (resp. transpiration); this is not a novelty, neither it is specific to those conditions.
Ozdogan, M., and G. D. Salvucci (2004), Irrigation-induced changes in potential evapotranspiration in southeastern Turkey: Test and application of Bouchet’s complementary hypothesis, Water Resour. Res., 40, W04301, doi:10.1029/2003WR002822.
Roderick, M. L., and G. D. Farquhar (2002), The cause of decreased pan evaporation over the past 50 years, Science, 298, 1410– 1411.
* Actually this is why a network of agrometeorological stations in standard conditions (i.e. well irrigated short grass) are usually required for computing reference evapotranspiration for monitoring plant water demand and plant water use in heavily irrigated areas, but rarely maintained in those standard conditions.
Citation: https://doi.org/10.5194/egusphere-2024-3562-RC1 - AC2: 'Reply on RC1', Tanguy Lunel, 27 Aug 2025
-
RC2: 'Comment on egusphere-2024-3562', Anonymous Referee #2, 08 Jul 2025
This is a well-written article on an important topic. The experimental design and data analysis seem adequate. I have only two very minor comments below. I couldn't find any visualization over the whole study duration. This can be beneficial to include.
Minor comments:
- Subsection numbering is not consistent. For instance, there is a “Materials and methods” section (Section #2), but no “Results” section.
- Figures 1-3: Is there any specific reason that the actual values over the entire experiment duration are not shown and only the average values are presented and compared with modeling schemes?
Citation: https://doi.org/10.5194/egusphere-2024-3562-RC2 - AC1: 'Reply on RC2', Tanguy Lunel, 27 Aug 2025
Peer review completion
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2024-3562', Nima Zafarmomen, 25 Feb 2025
This paper presents a robust and insightful analysis of the systematic overestimation of evapotranspiration over irrigated areas by offline land surface models. The authors effectively integrate field observations with advanced coupled modeling techniques to dissect the complex interplay between irrigation-induced atmospheric changes and land surface processes. Their detailed examination of various ISBA configurations, combined with a thorough validation against observational data, not only deepens our understanding of the atmospheric feedback mechanisms but also offers valuable guidance for improving model performance in both weather forecasting and water resource management. The comprehensive approach and meticulous quantification of key processes make this work a significant contribution to the field.
-
How sensitive are the results to the various ISBA configuration choices (e.g., canopy representation, stomatal conductance schemes, drought response) and to what extent might these choices limit the generalizability of the findings to other LSM frameworks?
-
Given that the atmospheric forcings (atmo_NOIRR and atmo_IRR_FC) are derived from a specific coupled model simulation over the LIAISE campaign period, how representative are these forcings for other irrigated regions or different meteorological conditions?
-
The paper relies on validation using data from two field sites—how robust is the model evaluation across diverse settings, and what uncertainties remain in the comparison between modeled and observed near-surface meteorological variables?
-
Can the authors clarify how the compensatory interactions between transpiration and soil evaporation are quantified, and what are the uncertainties associated with isolating the atmospheric feedback effects on these individual processes?
-
How might the biases associated with the offline LSM approach (due to missing irrigation-induced atmospheric feedback) impact downstream applications in water resource management and agricultural planning, and what strategies are proposed to mitigate these limitations in operational settings?
Additionally, it would strengthen the manuscript to reference recent advances in remote sensing applications in hydrological modeling. In particular, please consider citing the paper 'Assimilation of Sentinel‐based Leaf Area Index for Modeling Surface‐Groundwater Interactions in Irrigation Districts' to provide further context and support for the integration of satellite-based vegetation parameters in modeling surface–groundwater interactions in irrigated areas.
Citation: https://doi.org/10.5194/egusphere-2024-3562-CC1 - AC3: 'Reply on CC1', Tanguy Lunel, 27 Aug 2025
-
-
RC1: 'Comment on egusphere-2024-3562', Anonymous Referee #1, 08 May 2025
This paper describes a sensitivity study of a Land Surface Model (ISBA) under two atmospheric forcing conditions using coupled and offline LSM-BLM/RCM over two irrigated plots in the LIASE campaign: one considering the feedback of irrigation on the atmospheric temperature and humidity profiles and one ignoring it. The description of the numerical experiment to generate both forcing datasets as well as their analysis in terms of evapotranspiration overestimation and other meteorological variables when ignoring irrigation are given in QJRMS (https://doi.org/10.1002/qj.4736).
There are three main issues with the current manuscript:
- The importance of the impact of irrigation on the (changing) local/regional climate and the resulting areal evapotranspiration has been long recognized (cf. examples below); therefore, it is hard for the reader to identify the overall important message (to which it is referred lines 628-629), out of the obvious statement that irrigation affects the atmospheric conditions above the irrigated plot if the latter is large enough (wetter, cooler conditions) and should be considered when applying distributed hydrological or land surface models (statement which has been already put forward in the QJRMS paper) ?*
- The sensitivity study is not generic enough to carry substantial interest for the HESS community (a few days only, with specific conditions, forcings are all locally generated, no benchmarks from a reanalysis have been used); indeed, most LSM or hydrological model simulations can’t be coupled to regional climate models and are therefore offline simulations (that is, the vast majority of the research presented in journals like HESS); it would have been interesting to provide a solution to bypass the bias induced by ignoring the feedback effect on the atmosphere by the irrigated areas: should one use a simple approach such as the Bouchet Complementarity for the subareas where irrigation takes place ? (cf. the ample literature by Jozsef Szilagyi to develop an equilibrium Priestly-Taylor formalism with a “wet bulb” like information ?); in the last decade several authors have tried to use the Bouchet Complementarity to analyse the effect of irrigation on regional climate change; I can mention at least one work in Turkey (Ozdogan and Salvucci, 2002) but also the “Australian Pan Evapotranspiration Decrease Controversy”, for which several authors have analysed the cause of the decrease in pan ET data, first relating it to solar dimming (Roderick and Farquhar, 2002), but then realizing it is due mostly to the development of irrigation around the pans. I think that the literature review, which largely focuses on more recent works, should at least mention this line of research.
- The paper lacks clarity at multiple places: for instance, the analysis of the impact on the partitioning between evaporation E and transpiration T for various soil moisture levels focuses on a common stress index, SWI, and it is not clear how it affects E and T separately; only E(SWI) is shown, and it is not sure whether in Fig. 7 the same SWI levels impacts E and T. In that case, why not focusing on different moisture controls for E and T ? Also, what about advection from nearby drier areas in the simulation ? Is it dutifully accounted for (especially in those summer conditions) ?
In summary, in order to make this work fully compliant with the readership of HESS, I would recommend extending the paper by offering solutions to assess and correct for the ET overestimation at regional and seasonal scales, using for example larger scale simulations (ERA5Land-like) on the LIAISE area. I don’t really see what general outcomes one can draw from a few days of simulations over two plots, but maybe I missed out on an original sidelining interest of the paper.
Minor comments:
Line 33: “… is now fairly well understood and represented in LSMs”: I would dampen this optimistic statement. It is still fairly hard to go below 30% error.
Lines 138-144: This paragraph is curcial for the understanding of the paper, but lacks clarity and precision (e.g. “the surface features are not taken” etc).
Line 240: how does SWI also potentially affect transpiration ?
Lines 254-269 / 2.3: This echoes to my main comments 2 and 3: impact should be assessed at the seasonal scale, which is the scale of application of ET0; also, ET0 uses many assumptions that are only valid for a dense short grass fully irrigated, it is not meant to be considered in comparison with an output of an LSM outside of the numerous crop coefficient adjustments that are provided in the method to account for the “non standard” conditions (meteorological, moisture availability, type of the irrigation practice, plant development stage, etc), and typically not at the temporal scale of one single day.
Line 566: it seems to me that it has been already recognized in the LSM community that moisture-driven stress factor in the Jarvis formulation is less realistic than one based on leaf water potential.
Lines 570-578: here it would be helpful to have a discussion on what combinations of situations in the sensitivity are realistic in the given context (dry root zone vs wet surface soil moisture, no-irrig atmospheric conditions with high SWI etc etc).
Line 635: “a decrease in transpiration leads to an increase in bare soil evaporation”: this automatically the case if using a coupled model such as MEB because any drop in transpiration (resp. evaporation, for that matter) increases the temperature and decreases the humidity at the aerodynamic level and enhances evaporation (resp. transpiration); this is not a novelty, neither it is specific to those conditions.
Ozdogan, M., and G. D. Salvucci (2004), Irrigation-induced changes in potential evapotranspiration in southeastern Turkey: Test and application of Bouchet’s complementary hypothesis, Water Resour. Res., 40, W04301, doi:10.1029/2003WR002822.
Roderick, M. L., and G. D. Farquhar (2002), The cause of decreased pan evaporation over the past 50 years, Science, 298, 1410– 1411.
* Actually this is why a network of agrometeorological stations in standard conditions (i.e. well irrigated short grass) are usually required for computing reference evapotranspiration for monitoring plant water demand and plant water use in heavily irrigated areas, but rarely maintained in those standard conditions.
Citation: https://doi.org/10.5194/egusphere-2024-3562-RC1 - AC2: 'Reply on RC1', Tanguy Lunel, 27 Aug 2025
-
RC2: 'Comment on egusphere-2024-3562', Anonymous Referee #2, 08 Jul 2025
This is a well-written article on an important topic. The experimental design and data analysis seem adequate. I have only two very minor comments below. I couldn't find any visualization over the whole study duration. This can be beneficial to include.
Minor comments:
- Subsection numbering is not consistent. For instance, there is a “Materials and methods” section (Section #2), but no “Results” section.
- Figures 1-3: Is there any specific reason that the actual values over the entire experiment duration are not shown and only the average values are presented and compared with modeling schemes?
Citation: https://doi.org/10.5194/egusphere-2024-3562-RC2 - AC1: 'Reply on RC2', Tanguy Lunel, 27 Aug 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 5,956 | 1,128 | 246 | 7,330 | 201 | 334 |
- HTML: 5,956
- PDF: 1,128
- XML: 246
- Total: 7,330
- BibTeX: 201
- EndNote: 334
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Tanguy Ronan Lunel
Belen Marti
Aaron Boone
Patrick Le Moigne
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4647 KB) - Metadata XML
This paper presents a robust and insightful analysis of the systematic overestimation of evapotranspiration over irrigated areas by offline land surface models. The authors effectively integrate field observations with advanced coupled modeling techniques to dissect the complex interplay between irrigation-induced atmospheric changes and land surface processes. Their detailed examination of various ISBA configurations, combined with a thorough validation against observational data, not only deepens our understanding of the atmospheric feedback mechanisms but also offers valuable guidance for improving model performance in both weather forecasting and water resource management. The comprehensive approach and meticulous quantification of key processes make this work a significant contribution to the field.
How sensitive are the results to the various ISBA configuration choices (e.g., canopy representation, stomatal conductance schemes, drought response) and to what extent might these choices limit the generalizability of the findings to other LSM frameworks?
Given that the atmospheric forcings (atmo_NOIRR and atmo_IRR_FC) are derived from a specific coupled model simulation over the LIAISE campaign period, how representative are these forcings for other irrigated regions or different meteorological conditions?
The paper relies on validation using data from two field sites—how robust is the model evaluation across diverse settings, and what uncertainties remain in the comparison between modeled and observed near-surface meteorological variables?
Can the authors clarify how the compensatory interactions between transpiration and soil evaporation are quantified, and what are the uncertainties associated with isolating the atmospheric feedback effects on these individual processes?
How might the biases associated with the offline LSM approach (due to missing irrigation-induced atmospheric feedback) impact downstream applications in water resource management and agricultural planning, and what strategies are proposed to mitigate these limitations in operational settings?
Additionally, it would strengthen the manuscript to reference recent advances in remote sensing applications in hydrological modeling. In particular, please consider citing the paper 'Assimilation of Sentinel‐based Leaf Area Index for Modeling Surface‐Groundwater Interactions in Irrigation Districts' to provide further context and support for the integration of satellite-based vegetation parameters in modeling surface–groundwater interactions in irrigated areas.