the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Apr 2021
The application of Budyko framework to irrigation districts in China under various climatic conditions
Abstract. Budyko's framework has been widely used to study basin-scale water balance. In this study, we focus on the extended application of Fu's equation (one formulation of the Budyko-type curves) to 371 large irrigation districts in China over a period of 2010–2017. Water balance method was used to estimate actual evapotranspiration (ET) in the irrigated areas. Considering the contribution of shallow groundwater to ET, the water availability in the Budyko framework defined as equivalent precipitation (Pe) for irrigation areas is the sum of irrigation water (I), precipitation (P) and groundwater evaporation (ETgw). Results showed that the relationships between evapotranspiration (ET), water availability (Pe) and energy supply (ET0) can be accurately described by the Budyko's curves. The Fu's equation performed better in humid and semi-humid regions than arid and semi-arid regions. The comparison between δET/δPe and δET/δET0 confirmed the relative effect of water availability and energy supply on ET according to the variation of climatic conditions. The optimal values of Budyko parameter ω for each irrigation district was obtained with multi-annual data using least square method. Normalized Difference Vegetation Index (NDVI) and soil property (denoted by the proportion of clay and sand) were selected to develop empirical equation for parameter ω using multiple linear regression analysis method. This study showed that the Budyko framework can be extended to irrigation areas and provide useful information on evapotranspiration to assist in water management in irrigation areas.
- Preprint
(1146 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on hess-2021-80', Anonymous Referee #1, 27 May 2021
General Comments
This article presents a study that extends the Budyko framework to irrigated areas and applies it on several districts in China. The topic is clearly relevant for HESS and this article may guide further studies aiming at taking into account irrigated areas in hydrological modeling studies. But at this stage, the proposed study relies on many hypotheses that are not tested/mentioned clearly and consequently, the reach of the results is difficult to assess.
1. Applicability of the Budyko hypothesis in unclosed systems
The Budyko framework is usually intended to describe/model the partitioning of water fluxes at the catchment scale. The catchment scale is important since it allows to work on a closed system, where inputs and outputs can be clearly and unequivocally stated. In the proposed study, the Budyko framework is applied on irrigation districts and the different fluxes considered are, to my opinion inter-dependent. For example in Eq. 4, equivalent precipitation is proposed as the sum of Precipitation, Irrigation and groundwater consumption. This suggests implicitly that the Irrigation water and Groundwater used for evapotranspiration are water fluxes originated from other sources than Precipitation falling over the district area, which is questionable. Consequently, using Eq. 4 may lead to count Precipitation fluxes twice since water provided by irrigation (and possibly groundwater) originate from precipitation. This may be true depending on spatial and temporal scales considered but this is not discussed in the paper. Both references cited in the paper to present Eq. 4 are not relevant since Wang et al. (2011) did not consider irrigation but water storage and Chen et al. (2020) considered catchment scale modelling.
2. Lack of clear validation with observed data
The authors propose a validation of ET using MOD16 product but it should be stated that the comparison to MOD16 ET cannot be viewed as a strict validation since MOD16 ET relies heavily on modelling. The validation using streamflow time series at catchment scales is to my opinion the unique way to perform a real validation with independent data. The interpretation of Fig.3 is also complicated since all variables (ET/Pe, ET0/Pe) are derived from computation with associated uncertainties that are very difficult to quantify at this stage. Interpreting the deviation of the simulations and “observations” is thus impossible.
3. Other comments
l.52-53: Phrasing problems.
l. 151: Why considering Pan evaporation in Eq.3 instead of Penman equation?
l.197-198: Computing effective rainfall is highly uncertain. Eq. 8 is a way to estimate it but may leads to large errors. The USDA SCS method provides alternative ways to take into account soil types and land use classes. Besides, I failed to understand why ET0 is not involved in this calculus of effective rainfall. I would expect that the authors quantify the uncertainties related to this estimation, or at least provide the magnitude of effective rainfall compared to rainfall and Irrigated fluxes
l.219 : Typo in y-axis label.
l. 237: Perhaps I missed something but why Pe is replaced with (I+P) in Semi-arid areas?
l.279-281. Is there a clear justification why w is different according to the climatic settings? I would expect that w be more likely dependent on land use, soil and vegetation types, not climate.
l.379: “Effective precipitation efficiency” is not clearly defined. How is it computed and what is really shown on Fig. 9?
Citation: https://doi.org/10.5194/hess-2021-80-RC1 -
AC1: 'Reply on RC1', zailin Huo, 13 Jul 2021
The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2021-80/hess-2021-80-AC1-supplement.pdf
-
AC1: 'Reply on RC1', zailin Huo, 13 Jul 2021
-
RC2: 'Comment on hess-2021-80', Anonymous Referee #2, 11 Jun 2021
The authors present a study where they investigate the role of irrigation in the Budyko framework. This is a long standing issue in Budyko framework that is not solved yet (see e.g., Han et al 2011; Mianabaid et al 2020), mainly due to the lack of data on irrigation. This paper is fortunate to have irrigation data for 371 catchments in China, and is therefore unique in its kind. However, this data is not public as stated in the 'data availablility' section. In my opinion the given author statement is not in line with HESS's data policy (https://www.hydrology-and-earth-system-sciences.net/policies/data_policy.html). Additionally, irrigation data is also more complex in comparison to hydrometeorological data, as the data highly depends on irrigation type, associated losses, etc. I appreciate the authors effort to explain the underlying principles in the Appendix; however, this information is not enough (e.g., how is the water use efficiency calculated or measured??; is the irrigation water originating from the same catchment?). This results in the fact that I can not judge the validity of the data (which is essentially modelled irrigation), which is at the core of this study. Hence, I highly recommend to share the data so that your study can be verified.
Next, to the data issue I am not sure whether your calculations are correct. According to Eq 4 you calculate the equivalent precipitation as the sum of irrigation (I), precipitation (P) and groundwater evaporation (ETg). But how does this relate to your water balance in Eq 9? Is ETg part of ET? And more importantly: how do you define ET? I think that in Eq 9 ET equals the total actual evaporation (=sum of interception evaporation Ei, transpiration Et and soil evaporation ETg?). However in Eq 10 it seems that ET is equal to transpiration, as Peffd equals P minus interception. And how is it possible that according to Eq 10 all water entering the unsaturated zone is evaporating? This would mean that no water is percolating to the ground water reservoir? Hence, to summerize, I have some doubts on the water balance closure in relation to how you define evaporation. A schematic conceptual overview might help to clarify this.
Besides my major concerns related to the data validity and water balance calculation, the manuscript is well written. It's easy to read in good English, well structured, and the Figures are OK.
Detailed comments:
- P3 L44: "...used AT global and regional scales...."
- P4 L70-71: the unit of ET, ET0 and P is mm/y.
- P5 L103-106: are this the only irrigation methods? What about furrow or sprinkling? This would have a large impact on e.g., interception 'losses'
- P5 L111: Figure 1A doesn't show meteo data. It shows the aridity index. Similiar to my comments on data availability regarding irrigation: is meteo data avaiable?
- Eq2: I hightly recommend to use single characters in equation and not to use acronyms like ET. ET can be mathematically confused with E*T. Better use sub- and superscripts.
- P6 L120-121: I don't understnad this sentence.
- Section 2.2: How do you know that the irrigation water is originating from the same catchment. If you have transport external water into your catchment, you are violating the water balance.
- P8 L149: Why do you add ETg? What does it matter if the plants use water from the unsaturated zone, shallow or deep groundwater?
- P10 L189-190: How is the net irrigation determined? Is transpiration measured to calculate the Water Use Efficiency? How is transpiration measured?
- Eq8: this equation effectively calculates interception. Interception is highly dependent on vegetation type; however, I do not see where vegetation has a role in Eq8.
- Eq 9: I think this equation should read: Ei + ET=I+P-D
- P11 L217: unit RSME is mm/y
- Fig 2: is the water balanced ET on the y-axis calculated based on Eq 9, and thus includes I? If so, it shows to me the potential errors in the irrigation estimates, as for high MODIS-ET the data points start to deviate from the 1:1. Especially, for arid areas irrigation is important.
- Fig 3: why is the x-axix of 'semi-arid' different? And should ETg not be included here? And how is ETg determined?
- Eq 11: 'exp' should not be in italic.
- Fig 8b: Are the observed ET values from MODIS? Please note that MODIS is a model.
References:
Han, S., Hu, H., Yang, D., & Liu, Q. (2011). Irrigation impact on annual water balance of the oases in tarim basin, northwest china. Hydrological Processes, 25(2), 167-174. doi:10.1002/hyp.7830
A. Mianabadi, K. Davary, M. Pourreza-Bilondi, A. M. J. Coenders-Gerrits; (2020) Budyko framework; towards non-steady state conditions, Journal of Hydrology. https://doi.org/10.1016/j.jhydrol.2020.125089
Citation: https://doi.org/10.5194/hess-2021-80-RC2 -
AC2: 'Reply on RC2', zailin Huo, 13 Jul 2021
The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2021-80/hess-2021-80-AC2-supplement.pdf
Status: closed
-
RC1: 'Comment on hess-2021-80', Anonymous Referee #1, 27 May 2021
General Comments
This article presents a study that extends the Budyko framework to irrigated areas and applies it on several districts in China. The topic is clearly relevant for HESS and this article may guide further studies aiming at taking into account irrigated areas in hydrological modeling studies. But at this stage, the proposed study relies on many hypotheses that are not tested/mentioned clearly and consequently, the reach of the results is difficult to assess.
1. Applicability of the Budyko hypothesis in unclosed systems
The Budyko framework is usually intended to describe/model the partitioning of water fluxes at the catchment scale. The catchment scale is important since it allows to work on a closed system, where inputs and outputs can be clearly and unequivocally stated. In the proposed study, the Budyko framework is applied on irrigation districts and the different fluxes considered are, to my opinion inter-dependent. For example in Eq. 4, equivalent precipitation is proposed as the sum of Precipitation, Irrigation and groundwater consumption. This suggests implicitly that the Irrigation water and Groundwater used for evapotranspiration are water fluxes originated from other sources than Precipitation falling over the district area, which is questionable. Consequently, using Eq. 4 may lead to count Precipitation fluxes twice since water provided by irrigation (and possibly groundwater) originate from precipitation. This may be true depending on spatial and temporal scales considered but this is not discussed in the paper. Both references cited in the paper to present Eq. 4 are not relevant since Wang et al. (2011) did not consider irrigation but water storage and Chen et al. (2020) considered catchment scale modelling.
2. Lack of clear validation with observed data
The authors propose a validation of ET using MOD16 product but it should be stated that the comparison to MOD16 ET cannot be viewed as a strict validation since MOD16 ET relies heavily on modelling. The validation using streamflow time series at catchment scales is to my opinion the unique way to perform a real validation with independent data. The interpretation of Fig.3 is also complicated since all variables (ET/Pe, ET0/Pe) are derived from computation with associated uncertainties that are very difficult to quantify at this stage. Interpreting the deviation of the simulations and “observations” is thus impossible.
3. Other comments
l.52-53: Phrasing problems.
l. 151: Why considering Pan evaporation in Eq.3 instead of Penman equation?
l.197-198: Computing effective rainfall is highly uncertain. Eq. 8 is a way to estimate it but may leads to large errors. The USDA SCS method provides alternative ways to take into account soil types and land use classes. Besides, I failed to understand why ET0 is not involved in this calculus of effective rainfall. I would expect that the authors quantify the uncertainties related to this estimation, or at least provide the magnitude of effective rainfall compared to rainfall and Irrigated fluxes
l.219 : Typo in y-axis label.
l. 237: Perhaps I missed something but why Pe is replaced with (I+P) in Semi-arid areas?
l.279-281. Is there a clear justification why w is different according to the climatic settings? I would expect that w be more likely dependent on land use, soil and vegetation types, not climate.
l.379: “Effective precipitation efficiency” is not clearly defined. How is it computed and what is really shown on Fig. 9?
Citation: https://doi.org/10.5194/hess-2021-80-RC1 -
AC1: 'Reply on RC1', zailin Huo, 13 Jul 2021
The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2021-80/hess-2021-80-AC1-supplement.pdf
-
AC1: 'Reply on RC1', zailin Huo, 13 Jul 2021
-
RC2: 'Comment on hess-2021-80', Anonymous Referee #2, 11 Jun 2021
The authors present a study where they investigate the role of irrigation in the Budyko framework. This is a long standing issue in Budyko framework that is not solved yet (see e.g., Han et al 2011; Mianabaid et al 2020), mainly due to the lack of data on irrigation. This paper is fortunate to have irrigation data for 371 catchments in China, and is therefore unique in its kind. However, this data is not public as stated in the 'data availablility' section. In my opinion the given author statement is not in line with HESS's data policy (https://www.hydrology-and-earth-system-sciences.net/policies/data_policy.html). Additionally, irrigation data is also more complex in comparison to hydrometeorological data, as the data highly depends on irrigation type, associated losses, etc. I appreciate the authors effort to explain the underlying principles in the Appendix; however, this information is not enough (e.g., how is the water use efficiency calculated or measured??; is the irrigation water originating from the same catchment?). This results in the fact that I can not judge the validity of the data (which is essentially modelled irrigation), which is at the core of this study. Hence, I highly recommend to share the data so that your study can be verified.
Next, to the data issue I am not sure whether your calculations are correct. According to Eq 4 you calculate the equivalent precipitation as the sum of irrigation (I), precipitation (P) and groundwater evaporation (ETg). But how does this relate to your water balance in Eq 9? Is ETg part of ET? And more importantly: how do you define ET? I think that in Eq 9 ET equals the total actual evaporation (=sum of interception evaporation Ei, transpiration Et and soil evaporation ETg?). However in Eq 10 it seems that ET is equal to transpiration, as Peffd equals P minus interception. And how is it possible that according to Eq 10 all water entering the unsaturated zone is evaporating? This would mean that no water is percolating to the ground water reservoir? Hence, to summerize, I have some doubts on the water balance closure in relation to how you define evaporation. A schematic conceptual overview might help to clarify this.
Besides my major concerns related to the data validity and water balance calculation, the manuscript is well written. It's easy to read in good English, well structured, and the Figures are OK.
Detailed comments:
- P3 L44: "...used AT global and regional scales...."
- P4 L70-71: the unit of ET, ET0 and P is mm/y.
- P5 L103-106: are this the only irrigation methods? What about furrow or sprinkling? This would have a large impact on e.g., interception 'losses'
- P5 L111: Figure 1A doesn't show meteo data. It shows the aridity index. Similiar to my comments on data availability regarding irrigation: is meteo data avaiable?
- Eq2: I hightly recommend to use single characters in equation and not to use acronyms like ET. ET can be mathematically confused with E*T. Better use sub- and superscripts.
- P6 L120-121: I don't understnad this sentence.
- Section 2.2: How do you know that the irrigation water is originating from the same catchment. If you have transport external water into your catchment, you are violating the water balance.
- P8 L149: Why do you add ETg? What does it matter if the plants use water from the unsaturated zone, shallow or deep groundwater?
- P10 L189-190: How is the net irrigation determined? Is transpiration measured to calculate the Water Use Efficiency? How is transpiration measured?
- Eq8: this equation effectively calculates interception. Interception is highly dependent on vegetation type; however, I do not see where vegetation has a role in Eq8.
- Eq 9: I think this equation should read: Ei + ET=I+P-D
- P11 L217: unit RSME is mm/y
- Fig 2: is the water balanced ET on the y-axis calculated based on Eq 9, and thus includes I? If so, it shows to me the potential errors in the irrigation estimates, as for high MODIS-ET the data points start to deviate from the 1:1. Especially, for arid areas irrigation is important.
- Fig 3: why is the x-axix of 'semi-arid' different? And should ETg not be included here? And how is ETg determined?
- Eq 11: 'exp' should not be in italic.
- Fig 8b: Are the observed ET values from MODIS? Please note that MODIS is a model.
References:
Han, S., Hu, H., Yang, D., & Liu, Q. (2011). Irrigation impact on annual water balance of the oases in tarim basin, northwest china. Hydrological Processes, 25(2), 167-174. doi:10.1002/hyp.7830
A. Mianabadi, K. Davary, M. Pourreza-Bilondi, A. M. J. Coenders-Gerrits; (2020) Budyko framework; towards non-steady state conditions, Journal of Hydrology. https://doi.org/10.1016/j.jhydrol.2020.125089
Citation: https://doi.org/10.5194/hess-2021-80-RC2 -
AC2: 'Reply on RC2', zailin Huo, 13 Jul 2021
The comment was uploaded in the form of a supplement: https://hess.copernicus.org/preprints/hess-2021-80/hess-2021-80-AC2-supplement.pdf
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 722 | 346 | 34 | 1,102 | 31 | 27 |
- HTML: 722
- PDF: 346
- XML: 34
- Total: 1,102
- BibTeX: 31
- EndNote: 27
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
