the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
CAMELS-CH: hydro-meteorological time series and landscape attributes for 331 catchments in hydrologic Switzerland
Martina Kauzlaric
Rosi Siber
Ursula Schönenberger
Pascal Horton
Jan Schwanbeck
Marius Günter Floriancic
Daniel Viviroli
Sibylle Wilhelm
Anna E. Sikorska-Senoner
Nans Addor
Manuela Brunner
Sandra Pool
Massimiliano Zappa
Fabrizio Fenicia
Abstract. We present CAMELS-CH (Catchment Attributes and MEteorology for large-sample Studies - Switzerland), a large-sample hydro-meteorological data set for hydrological Switzerland in Central Europe. This domain covers 331 basins within Switzerland and neighboring countries. About one third of the catchments are located in Austria, France, Germany and Italy. As an Alpine country, Switzerland covers a vast diversity of landscapes, including mountainous environments, karstic regions, and several strongly cultivated regions, along with a wide range of hydrological regimes, i.e. catchments that are glacier-, snow- or rain-dominated. Similar to existing data sets, CAMELS-CH comprises dynamic hydro-meteorological variables and static catchment attributes.
CAMELS-CH (Höge et al., 2023, available at: https://doi.org/10.5281/zenodo.7957061) encompasses 40 years of data between 1st January 1981 and 31st December 2020, including daily time series of stream flow and water levels, and of meteorological data such as precipitation and air temperature. It also includes daily snow water equivalent data for each catchment starting from 2nd September 1998. Additionally, we provide annual time series of land cover change and glacier evolution per catchment. The static catchment attributes cover location and topography, climate, hydrology, soil, hydrogeology, geology, land use, human impact and glaciers. This Swiss data set complements comparable publicly accessible data sets, providing data from the "water tower of Europe".
- Preprint
(20671 KB) - Metadata XML
- BibTeX
- EndNote
Marvin Höge et al.
Status: final response (author comments only)
-
RC1: 'Comment on essd-2023-127', Anonymous Referee #1, 19 Aug 2023
The study from Marvin et al. presents CAMELS dataset for 331 basins in hydrological Switzerland. It’s pleased to see more and more publicly available data from different countries, which has laid an important foundation for large-scale hydrological analysis. It’s also great to see the efforts from authors to collect data for catchments in hydrological Switzerland including the one from neighbouring countries. Also, it is quite novel to include glacier and land cover annual data into the CAMELS datasets. Overall, I think the paper is suitable for Earth System Science Data journal. There are several places that could be improved before publication, I would recommend minor revision. My detailed comments can be found below.
(1)Figure 1a: the river name label in light blue is too small to see.
(2)Line140-145: it would be better to add the median size of all catchments.
(3)Line 180: as for the sentence: ‘The observed time series data above are complemented by simulation-based time series products to bridge the spatial discrepancy between political and hydrological Switzerland, to fill temporal gaps in observation-based time series…’, can you make it clear which variables of observed times series data are used simulated data as supplement? Since simulation-based variables might contain uncertainties, I would suggest to include one indicator (like quality control code) for observed time series data to indicate which data for specific time periods are complemented from simulated data. Because from your descriptions, the observations might use simulated data as supplement to fill the temporal gaps. The key point is to make it very clear whenever the data are complemented from simulated time series. Or I think it would be better to also have an observation only time series data and not fill the temporal gaps from simulated data. It’s ok to have missing values in observation data and you can provide simulation-based time series data as reference. The users might need to make their own decision whether they want to use simulated data to fill the gap or not.
(4)It’s nice to see authors have included annual glaciers and land cover into dataset. Especially the annual changes of land cover are useful for analysing human-nature interactions for catchments. In terms of land cover product, is there any particular reason for using CORINE Land Cover (CLC) data sets? Have you try some other land cover datasets? Because I find the temporal resolution of CLC datasets might not ideal, as the data have quite long temporal interval and lots of interpolation work need to be done. The below link includes a global land cover datasets at 30 m resolution for the period 1982-2021, which might be good an alternative to try.
https://www.tandfonline.com/doi/full/10.1080/15481603.2022.2096184
(5)Line 305-310: Is there a particular reason you compare the hydrogeology data with CAMELS-GB? Why you choose CAMELS-GB as a reference? Why not compare with other CAMELS dataset, i.e. CAMELS-US etc.?
(6)I notice that in the csv format data, all variables for one timestep (time series data) or for one catchment (static attributes) are in one Excel-Cell and separated by a semicolon at the moment. Would that be possible to provide data that already splits up into different columns, which will be more friendly for users?
(7)The link in session ‘9 Code and data availability‘ for CAMELS-CH is not working. It might be due to missing ‘doi.org’ in the link.
Citation: https://doi.org/10.5194/essd-2023-127-RC1 -
RC2: 'Comment on essd-2023-127', Rosanna Lane, 18 Sep 2023
Overview
This paper presents CAMELS-CH: a dataset of hydro-meteorological time-series and catchment attributes for catchments across hydrological Switzerland. This complements the existing CAMELS datasets, and will facilitate large sample studies.
I first want to say that I think it is fantastic that the authors have compiled this dataset, and I appreciate the large amount of effort that has gone into gathering and formatting many different data sources. The previous CAMELS datasets have proved very useful to the community, and it is excellent that Switzerland is being added. The list of time-series variables and catchment attributes presented is comprehensive, and the annual time-series of glacier attributes and land cover are valuable additions. The paper itself is well-written and logically structured, with high quality figures throughout. I particularly like that the authors have discussed and plotted the novel features of the dataset (changing glacier coverage and land use over time), and the plot showing the bio-geographic regions of Switzerland (Fig. 4) is very useful for those not familiar with Swiss hydrology. It is also helpful that the authors have used the same climatic and hydrologic attributes and corresponding descriptions as in CAMELS-GB, as this will assist future large sample studies that are using both datasets. I therefore would recommend that this manuscript is accepted into ESSD, but suggest the following minor revisions.
Minor comments:
(1) The zenodo link to the dataset provided in section 9 did not work, but the link within the abstract worked fine. Please make sure to double check all links within the manuscript.
(2) Section 4: It looks from Figure 1 that the 331 selected gauging stations have good coverage across the country, but I was curious how/why you selected these stations. In particular, 1) were the 298 river catchments and 33 lakes all of the available data?, 2) did you carry out any quality checks?, 3) were there any criteria (e.g. length of record, suitability of gauging station) that you used to filter out unsuitable gauging stations?
(3) Line 154: what was the motivation behind the selection of this time period?
(4) Line 192: not just simulation-based variables: observation based variables are also subject to uncertainties due to the interpolation method.
(5) Line 286: I agree that it makes sense to only include data for hydrological years with 95% or more records. Is there a variable somewhere that says how many years were included in the calculation of observed hydrological signatures? It would be useful to at least highlight catchments where calculations are based on relatively few years and therefore may have less robust hydrological signatures. [I’ve now seen that this is mentioned in the next paragraph – it may be worth moving the text ‘the start date, end date and number of years used for the calculations’ earlier to L286/7].
(6) Line 291: It would be helpful to add the years covered by CAMELS-GB here as well
(7) Figure 1: I found it hard to pick out the yellow dots against the grey (I had to zoom in quite a lot to this figure). A border around the dots might help make them more easily visible.
(8) Figure 3: It is difficult to directly compare the observed and simulated data in map form. It would be helpful to additionally show the data in a way that allows direct comparisons: for example, scatter plots (observed vs simulated), distribution plots such as CDFs (with separate lines for observed and simulated) or maps of the differences.
(9) The data were generally intuitively labelled and formatted. However, I noticed that that some files contained both NA and NaN values (e.g. CAMELS_CH_obs_based_6010.csv). It would be easier for users if these were kept consistently as NaN values.
(10) I noticed that the .csv files have data separated by a semi-colon, meaning that they all appear in a single column when opened in Excel. It might be better to separate data using a comma so that it is more intuitive and easy to quickly look at the data. However, this is not a major issue.
Typos:
Line 191: missing .
Line 396: extra .
Line 416: missing )
Citation: https://doi.org/10.5194/essd-2023-127-RC2
Marvin Höge et al.
Data sets
Catchment attributes and hydro-meteorological time series for large-sample studies across hydrologic Switzerland (CAMELS-CH) Marvin Höge, Martina Kauzlaric, Rosi Siber, Ursula Schönenberger, Pascal Horton, Jan Schwanbeck, Marius Günter Floriancic, Daniel Viviroli, Sibylle Wilhelm, Anna E. Sikorska-Senoner, Nans Addor, Manuela Brunner, Sandra Pool, Massimiliano Zappa, and Fabrizio Fenicia https://doi.org/10.5281/zenodo.7957061
Marvin Höge et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 608 | 244 | 20 | 872 | 14 | 12 |
- HTML: 608
- PDF: 244
- XML: 20
- Total: 872
- BibTeX: 14
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1