the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Sep 2022
Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models
Abstract. Groundwater-dominated catchments are often critical for nationally-important water resources. Many conceptual rainfall-runoff models tend to degrade in their model performance in groundwater-dominated catchments as they are rarely designed to simulate spatial groundwater behaviours or interactions with surface waters. Intercatchment groundwater flow is one such neglected variable. Efforts have been made to incorporate this process into existing models, but there is a need for improving our perceptual models of groundwater-surface water interactions prior to any model modifications.
In this study, national meteorological, hydrological, hydrogeological, geological and artificial influence (characterising abstractions and return flows) datasets are used to develop a perceptual model of intercatchment groundwater flow (IGF) and how it varies spatially and temporally across the River Thames, United Kingdom (UK). We characterise the water balance, presence of gaining/losing river reaches and intra-annual dynamics in 80 subcatchments of the River Thames, taking advantage of its wealth of data, densely gauged river network, and geological variability.
We show the prevalence of non-conservative river reaches across the study area, with heterogeneity both between, and within, geological units giving rise to a complex distribution of recharge and discharge points along the river network. We identify where non-conservative reaches can be attributed to IGF, and where other processes (e.g. surface water abstractions) are the likely cause. Through analysis of recorded water balance data and hydrogeological perceptualisation, we conclude that outcrops of carbonate fractured aquifers (Chalk and Jurassic Limestone) show evidence of IGF both from headwater to downstream reaches, and out-of-catchment via spring lines. We found temporal as well as spatial variability across the study area, with more seasonality and variability in river catchments on Jurassic Limestone outcrops compared to Chalk and Lower Greensand outcrops. Our results demonstrate the need for local investigation and hydrogeological perceptualisation within regional analysis, which we show to be achievable given relatively simple geological interpretation and data requirements. We support the inclusion of IGF fluxes within existing models to enable calibration improvements in groundwater-dominated catchments, but with geologically-specific temporal and spatial characteristics, and (when perceptually appropriate) connectivity between catchments.
-
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
(1643 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1643 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
loss functionsmay be used in conceptual rainfall–runoff models but should be supported by perceptualisation of IGF processes and connectivities.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-529', Anonymous Referee #1, 02 Oct 2022
Review of Oldham et al: Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models.
General remarks
The study uses an extensive dataset from the Thames catchment to find proof for (variations in) intercatchment groundwater flow (IGF). IGF is hard to quantify, but it is an important process to consider in hydrological modelling. The paper is very well written and well structured. A significant amount of data collection and processing work was done to enable a useful analysis of IGF at this scale.
On several places in the paper it is stated that both spatial and temporal variations of IGF were studied (e.g. L 19, 112, 605). The spatial variations of IGF are indeed well analyzed. I did not see much about temporal variations of IGF. There is Figure 7 with seasonal patterns in groundwater levels and water balance metrics, which are shortly described in chapter 5.2 and in L476-490. However, the link between this seasonality and the temporal variability in IGF was not described. In addition, there is probably more than just seasonality: what about year to year variations in IGF losses and gains? Do these year to year variations match between losing and gaining stretches (or is there a delay)? These temporal aspects could be either better covered or left out of this paper.
Regarding the spatial analysis: would it be possible to connect losing and gaining stretches, compare the IGF fluxes and maybe combine catchments into larger scale conservative catchments? This may be possible for e.g. the Coln, Kennet, Colne and Mole catchments.
Detailed comments
L27: We found temporal as well as spatial variability of…? See above regarding temporal variability in IGF.
L110-112: Consider to rephrase into an objective statement.
L190-198: Groundwater abstractions are not mentioned here. Later, this is covered and discussed. Still, I am curious whether the volumes of groundwater abstraction are significant enough to have impact on your analysis. Maybe some regional numbers are available?
L271: in->is?
L302: is the->is in the?
L338: it could save a lot of space to only show the naturalized results in fig 4,5,6. The difference is indeed not that large.
L356-357: with the low amount of catchments (4 in LG, 11 in JL) these interquartile ranges are highly uncertain. You could also choose to plot the averages with error bars to show the (variable) uncertainty of the statistics.
Figure 5: the catchment boundaries are unclear in these maps. Would more legend colors be possible?
Citation: https://doi.org/10.5194/egusphere-2022-529-RC1 - AC1: 'Reply on RC1', Louisa Oldham, 29 Nov 2022
-
RC2: 'Comment on egusphere-2022-529', Anonymous Referee #2, 10 Oct 2022
Review of the paper: Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models.
Non-conservative reaches and catchments are popular in groundwater-dominated regions and karst areas. Perceptualizing of hydrological processes in these regions is of great importance as it enables us to recognise where intercatchment groundwater flow (IGF) may be occurring and highlights the need for local investigation. In this study, a framework is proposed to evaluate the spatial and temporal IGF and applied to the River Thames with wealth of data and densely gauged river network. It is an interesting topic in hydrology, and the manuscript is well organized. However, there are still several problems and deficiencies in the paper and further revision is needed
General remarks
The water balance is the basic metric to recognize the IGF and the term AET plays a key factor in determine the metric. However, the estimation of the AET contains great uncertainty and AET has great spatial variability in large mountainous basin. The uncertainty in estimate of AET and then water balance metric should be analyzed.
Line 393-410. The analysis of water balance at inter-annual scale should be careful, as the temporal variation of water balance metrics is more complex than that for the multi-year average condition. For example, the soil water storage is a nonnegligible term for water balance. Further, the change of groundwater level is mainly controlled by local hydrogeological conditions. That's for sure, there are significant differences in the temporal variation of hydrological factors among hydrogeological units. But I don’t catch that how these reflect or indicate the differences in IGF.
The perceptual model of the Thames is of great importance in the paper (Figure 8). But it seems confusing as too many lines and explanatory text. I suggest authors reorganize the figure 8.
A description of climate, especially the spatial and temporal variability of P and AET, is needed for the basin in the section of study area. A brief introduce of the runoff depth and its temporal varation for the basin is also needed. It will be helpful to understand the degree of losing and gaining of reaches.
Others:
Line 220. “A positive residual” is , which should be pointed out clearly.
Line 265. More explanatory text is needed for figure 2.
Line 380. In figure 5, the water balance metric is greater than 1000 mm/yr in several catchments. The value seems too large for the region. The authors may check it carefully.
Line 413. In figure 7, what are the means of shadows in the sub-figures (a)-(g) and different colors of curves in (h)-(o).
Lines 600-6015. In this section, the authors should focus on what you have found in the paper rather than suggestions.
Citation: https://doi.org/10.5194/egusphere-2022-529-RC2 - AC2: 'Reply on RC2', Louisa Oldham, 29 Nov 2022
-
RC3: 'Comment on egusphere-2022-529', Anonymous Referee #3, 11 Oct 2022
Review of “Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models” by Oldham et al.
General comments:
It is a common fact that water actively exchanges between surface divides in regions with carbonate fractured aquifer outcrops. This have challenged the ‘watertight substratum’ assumption that is the foundation of many existing catchment rainfall-runoff models for long time without appropriate solutions and model conceptulizations. The main aim of this manuscript is to improve our perceptual models of intercatchment groundwater flow. The authors took advantage of their wealth of data, densely gauged river network, and geological variability from national meteorological, hydrological, hydrogeological, geological and artificial influence datasets to develop a perceptual model of intercatchment groundwater flow (IGF) and to show how it varies spatially and temporally in 80 subcatchments of the River Thames, United Kingdom (UK). The water balance, presence of gaining/losing river reaches and intra-annual dynamics were investigated through a water balance analysis.
The study is important for hydrological predictions and water resources management in groundwater-activated catchments. However, the method adopted by the authors can only provide site-specific results about qualitative water balance, it is still difficult to represent regional inter-catchment groundwater dynamics as they could not provide some essential functions that describing how groundwater between neighbor units exchanges according to different conditions of groundwater levels, different lithology, human abstractions and so on. In order to couple IGF processes into existing hydrologic models, it is important for the authors to derive the IGF functions quantitatively describing how IGF varies with time, groundwater levels and abstractions, ….
The water balance equations (1)-(4) adopted are also not rigorous as discussed by the authors themselves in Section 6.3 that input data uncertainties can lead to large computational uncertainty. In fact, equations (1) or (3) represent multiyear water balance instead of single year water balance. So dS/dt=0 is not strictly true, and a empirical 100 mm/yr was used by the authors to help to identify the non-conservative reach water balance. As the IGF fluxes could not be measured directly in catchment scale, empirical estimation is inevitable. However, the fudge factors e.g., 100 mm/yr as well as the physical meaning of S (groundwater storage or soil water storage?) should be discussed in depth.
The quality of many figures could be further improved e.g., to fully show the reach units are subdivided and to accompany their figures tightly with the text words to upgrade the readability. The reviewer suggests that reach units can be subdivided into two categories, the headwater reach and the internal reach. The water balance of reach units from the headwater areas which is recharged singly by precipitation in conservative catchments should be highlighted in order to identify the leakage recharge from outside catchment.
Specific Comments:
P means page, and L means lines
P4L120: Here annual average precipitation for the whole basin and its spatiotemporal distribution is needed. Discharge volume of main gauges also should be provided.
P6L144-146: “reach as the catchment area between river gauging stations. The analysis undertaken in this study is developed at the river reach scale rather than at the sub-catchment scale”. However, the title of this manuscript is “…perceptualizing inter-catchment groundwater flow…”. What is the difference between reach drainage area and sub-catchment? Are the units presented in Figure 1c the reach units? I suggest that the authors provide reach units distribution map in terms of the river gauging stations.
P7L175: Provide the cells adopted for water balance computations.
P8L175: “A limiting factor of 70% of the total reach area was assigned as an indicator of reach coverage.” What is the meaning of 70% here.
P8L217-219: S represents many storage components, e.g., groundwater storage, soil water storage, vegetation water storage, etc. how do they calculate groundwater exchanges without eliminations of other terms. It is possibly due to this reason I guess that an empirical factor 100mm/yr was adopted (see in Lines 573-576), which helps to filter disturbance from other terms? In addition, equation (1) or (3) can represent multiyear water balance instead of single year water balance. So the authors should explain the limits of using these equations.
P12L315-316: “Due to the high storage (Table 1)”. In table 1 lower greensand aquifers are with the lowest average (0.005) storage coefficients? Why you claimed the high storage in the main text? Similar expressions can be seen also in P11L290, P21L479.
P13L340-341: “The three lowest main river reaches show particularly large naturalised water balance losses (>1000 mm yr-1)”. I noticed that the average annual precipitation of Thames basin is only about 710 mm (Gabriel et al., 2022). Why so much losses of water (>1000 mm yr-1) in the main reaches in the River Thames? Do you have the losses averaged over the reach units? In P14L349-357, other values about water losses or gains seems to be regular. However, I don’t understand how do you convert the water losses into water depth. I suggest that the authors may use water losses volume in m3 yr-1 instead of water depth since the reference reach unit area is quite difference and upstream inflow is also different from up to down river reaches.
P14L362: what is the raio of 622 mm yr-1 annual loss in total volume of precipitation in the Kennet headwater reaches. As we know, headwater reaches do not receive upland surface inflow, so the net loss of 622 mm is large compared to the annual precipitation 710 mm over the whole basin.
P19: I suggest that the total amount of groundwater exchange should be marked in Figure 8. And how do you judge the flux directions? From the method in Section 3, I do not find related algorithm for estimating the flux directions.
P20L454-455. “The Chalk of the Thames Basin can be locally sub-karstic, but fracture and fissure flow remain the primary groundwater flow”. It maybe true as you claimed, however, if the IGFs should occur in the relatively less passageways of karstic conduits?
P21L503. It is true that not including IGF as a model flux will result in many models overestimating river flows or actual evapotranspiration. But the key question may be to describe how groundwater between neighbor units exchanges according to different conditions of groundwater levels, different lithology, human abstractions and so on.
References:
Gabriel, R.K.; Fan, Y. Multivariate Hydrologic Risk Analysis for River Thames. Water 2022, 14, 384. https://doi.org/10.3390/w14030384.
Citation: https://doi.org/10.5194/egusphere-2022-529-RC3 - AC3: 'Reply on RC3', Louisa Oldham, 01 Dec 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-529', Anonymous Referee #1, 02 Oct 2022
Review of Oldham et al: Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models.
General remarks
The study uses an extensive dataset from the Thames catchment to find proof for (variations in) intercatchment groundwater flow (IGF). IGF is hard to quantify, but it is an important process to consider in hydrological modelling. The paper is very well written and well structured. A significant amount of data collection and processing work was done to enable a useful analysis of IGF at this scale.
On several places in the paper it is stated that both spatial and temporal variations of IGF were studied (e.g. L 19, 112, 605). The spatial variations of IGF are indeed well analyzed. I did not see much about temporal variations of IGF. There is Figure 7 with seasonal patterns in groundwater levels and water balance metrics, which are shortly described in chapter 5.2 and in L476-490. However, the link between this seasonality and the temporal variability in IGF was not described. In addition, there is probably more than just seasonality: what about year to year variations in IGF losses and gains? Do these year to year variations match between losing and gaining stretches (or is there a delay)? These temporal aspects could be either better covered or left out of this paper.
Regarding the spatial analysis: would it be possible to connect losing and gaining stretches, compare the IGF fluxes and maybe combine catchments into larger scale conservative catchments? This may be possible for e.g. the Coln, Kennet, Colne and Mole catchments.
Detailed comments
L27: We found temporal as well as spatial variability of…? See above regarding temporal variability in IGF.
L110-112: Consider to rephrase into an objective statement.
L190-198: Groundwater abstractions are not mentioned here. Later, this is covered and discussed. Still, I am curious whether the volumes of groundwater abstraction are significant enough to have impact on your analysis. Maybe some regional numbers are available?
L271: in->is?
L302: is the->is in the?
L338: it could save a lot of space to only show the naturalized results in fig 4,5,6. The difference is indeed not that large.
L356-357: with the low amount of catchments (4 in LG, 11 in JL) these interquartile ranges are highly uncertain. You could also choose to plot the averages with error bars to show the (variable) uncertainty of the statistics.
Figure 5: the catchment boundaries are unclear in these maps. Would more legend colors be possible?
Citation: https://doi.org/10.5194/egusphere-2022-529-RC1 - AC1: 'Reply on RC1', Louisa Oldham, 29 Nov 2022
-
RC2: 'Comment on egusphere-2022-529', Anonymous Referee #2, 10 Oct 2022
Review of the paper: Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models.
Non-conservative reaches and catchments are popular in groundwater-dominated regions and karst areas. Perceptualizing of hydrological processes in these regions is of great importance as it enables us to recognise where intercatchment groundwater flow (IGF) may be occurring and highlights the need for local investigation. In this study, a framework is proposed to evaluate the spatial and temporal IGF and applied to the River Thames with wealth of data and densely gauged river network. It is an interesting topic in hydrology, and the manuscript is well organized. However, there are still several problems and deficiencies in the paper and further revision is needed
General remarks
The water balance is the basic metric to recognize the IGF and the term AET plays a key factor in determine the metric. However, the estimation of the AET contains great uncertainty and AET has great spatial variability in large mountainous basin. The uncertainty in estimate of AET and then water balance metric should be analyzed.
Line 393-410. The analysis of water balance at inter-annual scale should be careful, as the temporal variation of water balance metrics is more complex than that for the multi-year average condition. For example, the soil water storage is a nonnegligible term for water balance. Further, the change of groundwater level is mainly controlled by local hydrogeological conditions. That's for sure, there are significant differences in the temporal variation of hydrological factors among hydrogeological units. But I don’t catch that how these reflect or indicate the differences in IGF.
The perceptual model of the Thames is of great importance in the paper (Figure 8). But it seems confusing as too many lines and explanatory text. I suggest authors reorganize the figure 8.
A description of climate, especially the spatial and temporal variability of P and AET, is needed for the basin in the section of study area. A brief introduce of the runoff depth and its temporal varation for the basin is also needed. It will be helpful to understand the degree of losing and gaining of reaches.
Others:
Line 220. “A positive residual” is , which should be pointed out clearly.
Line 265. More explanatory text is needed for figure 2.
Line 380. In figure 5, the water balance metric is greater than 1000 mm/yr in several catchments. The value seems too large for the region. The authors may check it carefully.
Line 413. In figure 7, what are the means of shadows in the sub-figures (a)-(g) and different colors of curves in (h)-(o).
Lines 600-6015. In this section, the authors should focus on what you have found in the paper rather than suggestions.
Citation: https://doi.org/10.5194/egusphere-2022-529-RC2 - AC2: 'Reply on RC2', Louisa Oldham, 29 Nov 2022
-
RC3: 'Comment on egusphere-2022-529', Anonymous Referee #3, 11 Oct 2022
Review of “Evidence-based requirements for perceptualising intercatchment groundwater flow in hydrological models” by Oldham et al.
General comments:
It is a common fact that water actively exchanges between surface divides in regions with carbonate fractured aquifer outcrops. This have challenged the ‘watertight substratum’ assumption that is the foundation of many existing catchment rainfall-runoff models for long time without appropriate solutions and model conceptulizations. The main aim of this manuscript is to improve our perceptual models of intercatchment groundwater flow. The authors took advantage of their wealth of data, densely gauged river network, and geological variability from national meteorological, hydrological, hydrogeological, geological and artificial influence datasets to develop a perceptual model of intercatchment groundwater flow (IGF) and to show how it varies spatially and temporally in 80 subcatchments of the River Thames, United Kingdom (UK). The water balance, presence of gaining/losing river reaches and intra-annual dynamics were investigated through a water balance analysis.
The study is important for hydrological predictions and water resources management in groundwater-activated catchments. However, the method adopted by the authors can only provide site-specific results about qualitative water balance, it is still difficult to represent regional inter-catchment groundwater dynamics as they could not provide some essential functions that describing how groundwater between neighbor units exchanges according to different conditions of groundwater levels, different lithology, human abstractions and so on. In order to couple IGF processes into existing hydrologic models, it is important for the authors to derive the IGF functions quantitatively describing how IGF varies with time, groundwater levels and abstractions, ….
The water balance equations (1)-(4) adopted are also not rigorous as discussed by the authors themselves in Section 6.3 that input data uncertainties can lead to large computational uncertainty. In fact, equations (1) or (3) represent multiyear water balance instead of single year water balance. So dS/dt=0 is not strictly true, and a empirical 100 mm/yr was used by the authors to help to identify the non-conservative reach water balance. As the IGF fluxes could not be measured directly in catchment scale, empirical estimation is inevitable. However, the fudge factors e.g., 100 mm/yr as well as the physical meaning of S (groundwater storage or soil water storage?) should be discussed in depth.
The quality of many figures could be further improved e.g., to fully show the reach units are subdivided and to accompany their figures tightly with the text words to upgrade the readability. The reviewer suggests that reach units can be subdivided into two categories, the headwater reach and the internal reach. The water balance of reach units from the headwater areas which is recharged singly by precipitation in conservative catchments should be highlighted in order to identify the leakage recharge from outside catchment.
Specific Comments:
P means page, and L means lines
P4L120: Here annual average precipitation for the whole basin and its spatiotemporal distribution is needed. Discharge volume of main gauges also should be provided.
P6L144-146: “reach as the catchment area between river gauging stations. The analysis undertaken in this study is developed at the river reach scale rather than at the sub-catchment scale”. However, the title of this manuscript is “…perceptualizing inter-catchment groundwater flow…”. What is the difference between reach drainage area and sub-catchment? Are the units presented in Figure 1c the reach units? I suggest that the authors provide reach units distribution map in terms of the river gauging stations.
P7L175: Provide the cells adopted for water balance computations.
P8L175: “A limiting factor of 70% of the total reach area was assigned as an indicator of reach coverage.” What is the meaning of 70% here.
P8L217-219: S represents many storage components, e.g., groundwater storage, soil water storage, vegetation water storage, etc. how do they calculate groundwater exchanges without eliminations of other terms. It is possibly due to this reason I guess that an empirical factor 100mm/yr was adopted (see in Lines 573-576), which helps to filter disturbance from other terms? In addition, equation (1) or (3) can represent multiyear water balance instead of single year water balance. So the authors should explain the limits of using these equations.
P12L315-316: “Due to the high storage (Table 1)”. In table 1 lower greensand aquifers are with the lowest average (0.005) storage coefficients? Why you claimed the high storage in the main text? Similar expressions can be seen also in P11L290, P21L479.
P13L340-341: “The three lowest main river reaches show particularly large naturalised water balance losses (>1000 mm yr-1)”. I noticed that the average annual precipitation of Thames basin is only about 710 mm (Gabriel et al., 2022). Why so much losses of water (>1000 mm yr-1) in the main reaches in the River Thames? Do you have the losses averaged over the reach units? In P14L349-357, other values about water losses or gains seems to be regular. However, I don’t understand how do you convert the water losses into water depth. I suggest that the authors may use water losses volume in m3 yr-1 instead of water depth since the reference reach unit area is quite difference and upstream inflow is also different from up to down river reaches.
P14L362: what is the raio of 622 mm yr-1 annual loss in total volume of precipitation in the Kennet headwater reaches. As we know, headwater reaches do not receive upland surface inflow, so the net loss of 622 mm is large compared to the annual precipitation 710 mm over the whole basin.
P19: I suggest that the total amount of groundwater exchange should be marked in Figure 8. And how do you judge the flux directions? From the method in Section 3, I do not find related algorithm for estimating the flux directions.
P20L454-455. “The Chalk of the Thames Basin can be locally sub-karstic, but fracture and fissure flow remain the primary groundwater flow”. It maybe true as you claimed, however, if the IGFs should occur in the relatively less passageways of karstic conduits?
P21L503. It is true that not including IGF as a model flux will result in many models overestimating river flows or actual evapotranspiration. But the key question may be to describe how groundwater between neighbor units exchanges according to different conditions of groundwater levels, different lithology, human abstractions and so on.
References:
Gabriel, R.K.; Fan, Y. Multivariate Hydrologic Risk Analysis for River Thames. Water 2022, 14, 384. https://doi.org/10.3390/w14030384.
Citation: https://doi.org/10.5194/egusphere-2022-529-RC3 - AC3: 'Reply on RC3', Louisa Oldham, 01 Dec 2022
Peer review completion
Journal article(s) based on this preprint
loss functionsmay be used in conceptual rainfall–runoff models but should be supported by perceptualisation of IGF processes and connectivities.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 432 | 134 | 17 | 583 | 3 | 4 |
- HTML: 432
- PDF: 134
- XML: 17
- Total: 583
- BibTeX: 3
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
Louisa D. Oldham
Jim Freer
Gemma Coxon
Nicholas Howden
John P. Bloomfield
Christopher Jackson
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1643 KB) - Metadata XML