the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 May 2026
High-Resolution Karst Spring Discharge Datasets of the Euro-Mediterranean Mountain Regions
Abstract. Karst groundwater systems exhibit heterogeneity in recharge, circulation, and discharge, occupying a unique position within groundwater systems. This complexity facilitates rapid responses via the preferential flow routes, making karst systems vulnerable to climatic and anthropogenic pressures. High-altitude karst aquifers are particularly susceptible to shifting climate patterns – specifically rising temperatures, declining snow cover, and increasingly less and inconsistent precipitation – within the Mediterranean climate hotspot. Effective sustainable management of these groundwater systems require robust hydrological modelling; however, the application of such models is often constrained by the availability of high-quality, reliable datasets. This study presents a comprehensive collection of high-resolution karst spring discharge data from major Euro-Mediterranean mountain belts, including the Atlas, Betics, Pyrenees, Jura, Alps, Carpathians, Apennines, Dinarides, Hellenides, Balkans, Taurus, Levant, and Zagros. We compiled a total of 118 discharge time series specifically curated for hydrological modelling. Geographically, the dataset is led by the Alps (approx. 42%), followed by the Dinarides (approx. 10%), with the Apennines, Carpathians, and Zagros each contributing approx. 7%. The Levant and Taurus account for approx. 5% each, while the remaining regions (Atlas, Balkans, Betics, Hellenides, Jura, and Pyrenees) represent less than 5% each. In terms of temporal resolution, 92% of the records are daily, while hourly and monthly data each comprise 4%. The average record length is 19 years, which is led by a 99-year series from Unica Spring, Slovenia (1926–2025). Regional analysis indicates that the Alps, Apennines, Balkans, Betics, Dinarides, Jura, and Levant maintain average record lengths exceeding 20 years, whereas the Atlas, Carpathians, Taurus, and Zagros range between 10 and 20 years. The shortest average records were observed in the Hellenides and Pyrenees (7 and 8 years, respectively), which is still adequate for hydrological modelling applications.
- Preprint
(3711 KB) - Metadata XML
-
Supplement
(176 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on essd-2026-281', Sylke Hilberg, 09 Jul 2026
-
AC1: 'Reply on RC1', Suleyman Selim Calli, 02 Aug 2026
We sincerely thank the reviewer for the careful evaluation of our manuscript and for the highly positive and encouraging comments. We are grateful for your appreciation of the dataset, its presentation, and its potential value. It is very rewarding to know that the objectives and structure of the manuscript were found to be clear, transparent, and accessible despite the large volume of information presented.
We also thank you for your two suggestions regarding the figures. Following your recommendations, we have revised the figures to improve their readability (please find in the attachment). Specifically, the mountain labels in Figure 2 have been enlarged and repositioned, and Figure 7A has been updated to clearly indicate the area displayed in Figure 7B.
In addition to these revisions, we would like to inform the reviewer that, during the revision process, we further expanded the dataset by incorporating four additional karst spring discharge time series from Hungary and three from Romania, increasing the total number of karst springs from 118 to 125. The newly added datasets have already been uploaded to the repository and are available for inspection. We are also in contact with the corresponding data providers and intend to include them as co-authors in the revised manuscript, subject to the editor's approval. These additions will be fully described in the revised manuscript once the editor invites us to submit the revised version.
Kind regards,
S. Selim Calli
on behalf of the authors
-
AC1: 'Reply on RC1', Suleyman Selim Calli, 02 Aug 2026
-
RC2: 'Comment on essd-2026-281', Anonymous Referee #2, 10 Jul 2026
This manuscript presents the KaMERaMAN dataset, which compiles high-resolution discharge time series from 118 karst springs distributed across 13 European–Mediterranean mountain systems. The topic addresses the pressing need for high-quality observational data for hydrological modeling in climate-sensitive regions of the Mediterranean, and is therefore of clear scientific significance and practical relevance.
The dataset demonstrates good performance in terms of adherence to the FAIR principles, metadata completeness, spatial coverage, and data volume. All 120 data files are accompanied by comprehensive descriptive metadata with unique site identifiers, and the use of a standardized CSV format substantially enhances data accessibility and reusability. The dataset contains more than one million observations, with an average record length of approximately 19 years, making it a valuable resource for regional comparative hydrological studies.
I recommend minor revision prior to acceptance, with the following comments:
1.Negative discharge values: Some stations (e.g., TAU_03_TR@Pinarbasi) contain a small number of physically implausible negative discharge values. The authors should verify the origin of these values and provide an explanation in the data documentation. If these values result from measurement or post-processing errors, appropriate correction or clear flagging should be implemented.
2.Empty data files: A small number of station files contain entirely missing discharge values. The authors are encouraged to further check data completeness, clarify the reasons for missing data, or adjust the statement regarding the total number of “118 stations” to ensure consistency between the manuscript and the dataset.
3.Justification of variable selection: The dataset currently includes discharge as the sole variable. It would be beneficial to briefly justify this design choice in the manuscript and discuss potential future extensions to additional hydrological variables, in order to better define the scope and applicability of the dataset.Citation: https://doi.org/10.5194/essd-2026-281-RC2 -
AC2: 'Reply on RC2', Suleyman Selim Calli, 03 Aug 2026
We thank the reviewer for the thorough evaluation of our manuscript and for the highly positive and encouraging comments. We are delighted that you recognize the scientific value of the KaMERaMAN dataset, metadata quality, spatial coverage, and potential for supporting hydrological research in Mediterranean karst regions. We are especially grateful for the time and effort you invested in carefully examining such a large dataset. Your detailed inspection helped us identify several issues that have now been corrected, thereby improving the overall quality and reliability of the dataset.
In addition to addressing the comments raised during the review process, we further expanded the KaMERaMAN dataset by incorporating four new karst spring discharge time series from Hungary and three from Romania, increasing the total number of monitored karst springs from 118 to 125. The newly added datasets have already been uploaded to the repository (V2) for inspection. We are also in contact with the corresponding data providers and intend to include them as co-authors in the revised manuscript, subject to the editor's approval. These additions will be fully documented in the revised manuscript when we are invited to submit the revision.
Comment 1.
Negative discharge values: Some stations (e.g., TAU_03_TR@Pinarbasi) contain a small number of physically implausible negative discharge values. The authors should verify the origin of these values and provide an explanation in the data documentation. If these values result from measurement or post-processing errors, appropriate correction or clear flagging should be implemented.
Response: We sincerely thank the reviewer for carefully identifying this issue. We investigated the negative discharge values and confirmed that they originated from the extrapolation of the stage–discharge rating curve under very low-flow conditions, rather than representing actual negative spring discharge. We have corrected these values in the dataset and updated the repository accordingly. We also included an explanation in the revised manuscript and the data documentation to clarify the origin of these values and the corrections that were applied.
Comment 2.
Empty data files: A small number of station files contain entirely missing discharge values. The authors are encouraged to further check data completeness, clarify the reasons for missing data, or adjust the statement regarding the total number of “118 stations” to ensure consistency between the manuscript and the dataset.
Response: We sincerely thank the reviewer for bringing this issue to our attention. Following your observation, we carefully rechecked all station files in the repository. We identified that the files ALP_05_AT@Doserfall, ALP_29_AT@Schreiende Brunnen, and ALP_31_AT@Schwarzbach Moosquelle contained formatting errors introduced during dataset preparation. Specifically, the date columns were missing, and the discharge values had been inadvertently shifted to the first column, resulting in the discharge records being interpreted as missing (NA). We have corrected these formatting errors and updated the repository accordingly.
We would also like to clarify that several other stations contain missing discharge observations during certain monitoring periods rather than entirely missing datasets. These gaps reflect periods during which no measurements were available and are an inherent characteristic of long-term observational records. To avoid any ambiguity, we have expanded the discussion in the revised manuscript to explicitly describe data completeness, the occurrence of missing periods within individual records, and the reasons for these gaps wherever such information is available.
Furthermore, during the revision process, the KaMERaMAN dataset was expanded from 118 to 125 karst spring discharge time series through the addition of four Hungarian and three Romanian springs. The manuscript, metadata, and repository have been updated accordingly to ensure complete consistency.
Comment 3.
Justification of variable selection: The dataset currently includes discharge as the sole variable. It would be beneficial to briefly justify this design choice in the manuscript and discuss potential future extensions to additional hydrological variables, in order to better define the scope and applicability of the dataset.
Response: We thank the reviewer for this valuable suggestion. In the revised manuscript, we have added a paragraph explaining the rationale for focusing exclusively on spring discharge. Spring discharge is the most widely and consistently monitored hydrological variable across the Euro-Mediterranean mountain karst regions, making it the only variable that currently allows the compilation of a harmonized, large-scale dataset suitable for comparative hydrological analyses and hydrological model evaluation. We also discuss the potential future expansion of the KaMERaMAN dataset to include additional hydrometeorological variables, such as precipitation, air temperature, snow observations, groundwater levels, and water-quality parameters, where these data become available through collaboration with data providers. Such extensions will further broaden the applicability of the dataset for process-based hydrological and hydrogeological studies.
Kind regards,
S. Selim Çallı
on behalf of all authors
Citation: https://doi.org/10.5194/essd-2026-281-AC2
-
AC2: 'Reply on RC2', Suleyman Selim Calli, 03 Aug 2026
Data sets
High-Resolution Karst Spring Discharge Datasets of the Euro-Mediterranean Mountain Regions Çallı SS et al. https://doi.org/10.5281/zenodo.19448791
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 438 | 142 | 26 | 606 | 45 | 23 | 19 |
- HTML: 438
- PDF: 142
- XML: 26
- Total: 606
- Supplement: 45
- BibTeX: 23
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript High Resolution Karst Spring Discharge Datasets of the Euro-Mediterranean Mountain Regions by Selim Calli et al. presents a newly compiled dataset of 118 high-resolution karst spring discharge time series from major mountain regions across the Euro-Mediterranean area.
General comments: The study is specifically designed to support hydrological modelling of vulnerable karst groundwater systems. The dataset provides long-term, discharge records, offering an important resource for studying the impacts of climate change and improving the sustainable management of high-altitude karst aquifers. The manuscript is clearly written, provides a good overview of the most important karst aquifers in Europe, and thus serves as a solid foundation for in-depth karst hydrological studies. Publication is highly recommended.
Specific comments: I don't have any specific comments, as the study has been presented in a perfectly adequate way. The data collection and presentation of the results are clear and transparent. The challenge of presenting a large volume of data and a broad scope of investigation in a way that conveys information while keeping the manuscript easy to read is effectively addressed using examples in combination with supplementary materials.
Technical comments: I just have two very minor comments: in Fig. 2, some of the mountain names are hard to read; at first, I couldn't find the Alps at all. In Figure 7A, there is no boundary indicating which section is shown in Figure 7B.