the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Technical Note: Space-Time Statistical Quality Control of Extreme Precipitation Observations
- 1Institute for Modelling Hydraulic and Environmental Systems, University of Stuttgart, D-70569 Stuttgart, Germany
- 2Deutscher Wetterdienst, Offenbach, Germany
- 1Institute for Modelling Hydraulic and Environmental Systems, University of Stuttgart, D-70569 Stuttgart, Germany
- 2Deutscher Wetterdienst, Offenbach, Germany
Abstract. Precipitation extremes form the basis of many engineering design decisions. Extremes are rare events which may differ strongly from “normal” observations. Unfortunately, some of the observed extremes may be inaccurate or false. The purpose of this investigation is to present a quality check of observed extremes using space-time statistical methods. As a first step, the biggest values for each observation location and event duration are selected. For each of these the observed values of all other stations corresponding to the same time steps are collected and transformed using a Box-Cox transformation factor derived from a fitted truncated normal distribution. The value at the extreme location is estimated using the surrounding stations and the calculated spatial variogram, and this estimated value is compared to the observed extreme. If the difference exceeds the critical value of the test, the extreme is flagged as possible outlier. The same procedure is repeated for different aggregations in order to avoid singularities caused by convection. The flagged extremes are then compared to the extremes of the surrounding stations using the same procedure – interpolation and subsequent comparison of the interpolated and the observed values. Flagged extremes are subsequently compared to the corresponding radar and discharge observations and finally, implausible extremes are removed. The procedure is demonstrated using observations of sub-daily and daily temporal resolution in Germany.
Abbas El Hachem et al.
Status: closed
-
RC1: 'Comment on hess-2022-177', Anonymous Referee #1, 15 Jun 2022
Review HESS
Title: Technical Note: Space-Time Statistical Quality Control of Extreme Precipitation Observations
Author(s): Abbas El Hachem et al.
MS No.: hess-2022-177
MS type: Technical note
OverviewThe contribution is presenting a methodology to identify outlyers in precipitation measurements and to verify if the values aree plausible. The methodology is based on neighbouring stations for comparison, checking several time aggregations and utilizing external measurements from radar or water level gauges. The basic idea is that precipitation values are similar for neighbouring stations with short distances and getting more different with larger distance. This is being modeled with geostatstical methods.
General RemarksThe authors give a good explanation of the statistical methodology, but there are some shortcomings in the applied section:
- the ancillary data are introduced "as is" without acknowledging their quality
- there are several figures which are not cited but part of the contribution: here clearly textual explanation is missing I assume that the authors have a clear idea about the value of the figures, and they should share these ideas more explicitly with the readers
For this reason, a careful major revision should be performed in order to increase the quality of the description for the practical section up to the quality level of the theoretical section.
Specific Remarks
line 93: RW data have two properties which need to be discussed before use:
- they are hourly data
- the quality is varying over the years
line 99: the term "kernel density estimation with a Quartic shape" deserves more explanation for a better understanding
Figure 1: an explanation for the colouring with a scale would be useful
line 147: the quantitative criteria for comparison with radar data and with discharge data should somewhere be explained
line 152ff: a flow chart would be more helpful for the reader to follow the steps of the algorithm. In particular, the limit of loops is not obvious in the current list.
Section 4.1: a discussion of figure 4 and table 2 is missing here - or move the figures back to section 3.1
line 238: why are minute data here directly comparable to hourly sums? What are your criteria and where do you expect them to have limitations?
Tables 3 and 4 have never been cited or explained! Please add appropriate sections for explanation and discussion!
Figures 9 to 12 have never been cited or explained! Please add appropriate sections for explanation and discussion!
lines 266+267: please be more precise in your description.
line 275: the statement should be supported by at least a figure
line 280-281: what length of time interval? Please expand on this!Detailed corrections
line 56: should read additionally instead of additional
line 57: "is done" should be deleted
line 62: should read "Germany-wide"
line 85: citation of reference incorrect - should be "Kaspar et al. (2013)"
line 90: I assume you mean: radar derived rainfall images
line 124: better: "... is used, a method ..."
line 150: "were" should read "Where"
line 169: "is" should read "are"
line 171: spelling: hypothesis
line 176: delete "are"
line 211: "reflects" should read "reflect"
line 213: "an" should read "a"
Figure 3: spelling of "skewness" is wrong
line 241: spelling of "catchment"
line 244: replace "is" by "this" or reformulate
line 257: replace "have" by "has"
line 262: replace "were" by "where"
line 272: please add "s" to "investigate"
line 276: should read "convective rainfall processes" or should be reformulated
line 288: "this" should read "these"
line 320: reference incomplete: please add all co-authors!-
AC1: 'Reply on RC1', Abbas El Hachem, 19 Jul 2022
We thank the reviewer for taking the time to review our manuscript. Our reply to
the reviewer's comments are as follows:
The authors give a good explanation of the statistical methodology, but
there are some shortcomings in the applied section: - the ancillary data
are introduced ”as is” without acknowledging their quality
Ancillary data is referring to the radar and discharge data. Regarding
the discharge data, our approach is to check whether a precipitation event
has led to a reaction in runoff or not. Therefore, the accuracy of the runoff
values is of minor importance. Hence, we take the discharge data of the
provider(s) without performing any further plausibility checks. The case
with the radar data is similar, we are aware that precipitation estimates
form radars are prone to errors but the case here is similar do the one
with the discharge data. We primarily use this data to check whether
precipitation has occurred at a specific site or not.there are several figures which are not cited but part of the contribution:
here clearly textual explanation is missing I assume that the authors have
a clear idea about the value of the figures, and they should share these
ideas more explicitly with the readers
We will extend the manuscript and provide explanations for all the
figures. Furthermore, we will implement the specific remarks and detailed
corrections accordingly.
-
AC1: 'Reply on RC1', Abbas El Hachem, 19 Jul 2022
-
RC2: 'Comment on hess-2022-177', Anonymous Referee #2, 07 Jul 2022
The authors present a method for quality control/assessment of precipitation measurements especially in view a large values at different aggregation times between hours and days partly based on subhourly observations from DWD. As a new and innovative check they introduce radar precipitation estimates (DWD Radolan data set) as an independent set of observatins and discharge measurements at selected catchments which should show a hydrological response in case the observed rain gauge extreme rainfall event is not a fake observation.
Overall this could be a very useful contribution to HESS if the authors would indicate how the proposed procedure may generalize to other observing systems (e.g synop temperature/wind/humidity observations). Furthermore the text is written almost completely from the hydrolocical-engineering point of view: the authors should remember the E in the journals title, namely that other Earth System scientists should also be addressed and not to be scared off. Finally the text is largely written in a funding - interim report style, eg about one third of the figures is just added to the text without any discussion or even referencing in the text. All these points (and those commented in the annotated pdf document) should be revised before publication in HESS can be considered.-
AC2: 'Reply on RC2', Abbas El Hachem, 19 Jul 2022
We thank the reviewer for taking the time to review our manuscript. Our reply to
the reviewer's comments are as follows:Overall this could be a very useful contribution to HESS if the authors
would indicate how the proposed procedure may generalize to other observ-
ing systems (e.g synop temperature, wind and humidity observations).Our method is specifically designed for data with skewed distribution
and a high spatio-temporal variability. Outlier checks for other parameters
such a temperature are much more straight forward to implement. Since
the title specifically mentions precipitation we don’t see the necessity to
discuss how this could be used for other parameters
Our method is specifically designed for data with skewed distribution
and a high spatio-temporal variability. Outlier checks for other parameters
such a temperature are much more straight forward to implement. Since
the title specifically mentions precipitation we don’t see the necessity to
discuss how this could be used for other parameters.Furthermore the text is written almost completely from the hydrolocical-
engineering point of view: the authors should remember the E in the jour-
nals title, namely that other Earth System scientists should also be ad-
dressed and not to be scared off.See reply above, we explicitly present a quality check for precipitation
extremes.Finally the text is largely written in a funding - interim report style,
eg about one third of the figures is just added to the text without any dis-
cussion or even referencing in the text. All these points (and those com-
mented in the annotated pdf document) should be revised before publication
in HESS can be considered.We will provide and discuss all the figures in the manuscript, this was
also pointed out by referee 1. Regarding the “interim report style” we
submitted this as a technical note and not a full-fledged scientific pa-
per, therefore we consider a shorter, more report like text to be justified.
Regarding the comments in the manuscript (supplement), we will imple-
ment/correct/address them accordingly. There are however some remarks
which we would like to comment:Line 21 ff. This remark is correct, but since the DWD gauges assumed
to be installed according to the WMO guidelines, we can rule out such
scenarios to a large degree.
Line 74 ff. Of course, there are other weighing gauges, but we specifically refer to those used by the DWD.
Line 153: A flow chart is already presented in Figure 3.
Line 195: We fitted a spherical or an exponential variogram model
with no nugget to the experimental one. Both models guarantee that
the kriging covariance matrix is positive definite. Moreover we will add a
section showing that the choice of the variogram model has little influence
on the results
-
AC2: 'Reply on RC2', Abbas El Hachem, 19 Jul 2022
Status: closed
-
RC1: 'Comment on hess-2022-177', Anonymous Referee #1, 15 Jun 2022
Review HESS
Title: Technical Note: Space-Time Statistical Quality Control of Extreme Precipitation Observations
Author(s): Abbas El Hachem et al.
MS No.: hess-2022-177
MS type: Technical note
OverviewThe contribution is presenting a methodology to identify outlyers in precipitation measurements and to verify if the values aree plausible. The methodology is based on neighbouring stations for comparison, checking several time aggregations and utilizing external measurements from radar or water level gauges. The basic idea is that precipitation values are similar for neighbouring stations with short distances and getting more different with larger distance. This is being modeled with geostatstical methods.
General RemarksThe authors give a good explanation of the statistical methodology, but there are some shortcomings in the applied section:
- the ancillary data are introduced "as is" without acknowledging their quality
- there are several figures which are not cited but part of the contribution: here clearly textual explanation is missing I assume that the authors have a clear idea about the value of the figures, and they should share these ideas more explicitly with the readers
For this reason, a careful major revision should be performed in order to increase the quality of the description for the practical section up to the quality level of the theoretical section.
Specific Remarks
line 93: RW data have two properties which need to be discussed before use:
- they are hourly data
- the quality is varying over the years
line 99: the term "kernel density estimation with a Quartic shape" deserves more explanation for a better understanding
Figure 1: an explanation for the colouring with a scale would be useful
line 147: the quantitative criteria for comparison with radar data and with discharge data should somewhere be explained
line 152ff: a flow chart would be more helpful for the reader to follow the steps of the algorithm. In particular, the limit of loops is not obvious in the current list.
Section 4.1: a discussion of figure 4 and table 2 is missing here - or move the figures back to section 3.1
line 238: why are minute data here directly comparable to hourly sums? What are your criteria and where do you expect them to have limitations?
Tables 3 and 4 have never been cited or explained! Please add appropriate sections for explanation and discussion!
Figures 9 to 12 have never been cited or explained! Please add appropriate sections for explanation and discussion!
lines 266+267: please be more precise in your description.
line 275: the statement should be supported by at least a figure
line 280-281: what length of time interval? Please expand on this!Detailed corrections
line 56: should read additionally instead of additional
line 57: "is done" should be deleted
line 62: should read "Germany-wide"
line 85: citation of reference incorrect - should be "Kaspar et al. (2013)"
line 90: I assume you mean: radar derived rainfall images
line 124: better: "... is used, a method ..."
line 150: "were" should read "Where"
line 169: "is" should read "are"
line 171: spelling: hypothesis
line 176: delete "are"
line 211: "reflects" should read "reflect"
line 213: "an" should read "a"
Figure 3: spelling of "skewness" is wrong
line 241: spelling of "catchment"
line 244: replace "is" by "this" or reformulate
line 257: replace "have" by "has"
line 262: replace "were" by "where"
line 272: please add "s" to "investigate"
line 276: should read "convective rainfall processes" or should be reformulated
line 288: "this" should read "these"
line 320: reference incomplete: please add all co-authors!-
AC1: 'Reply on RC1', Abbas El Hachem, 19 Jul 2022
We thank the reviewer for taking the time to review our manuscript. Our reply to
the reviewer's comments are as follows:
The authors give a good explanation of the statistical methodology, but
there are some shortcomings in the applied section: - the ancillary data
are introduced ”as is” without acknowledging their quality
Ancillary data is referring to the radar and discharge data. Regarding
the discharge data, our approach is to check whether a precipitation event
has led to a reaction in runoff or not. Therefore, the accuracy of the runoff
values is of minor importance. Hence, we take the discharge data of the
provider(s) without performing any further plausibility checks. The case
with the radar data is similar, we are aware that precipitation estimates
form radars are prone to errors but the case here is similar do the one
with the discharge data. We primarily use this data to check whether
precipitation has occurred at a specific site or not.there are several figures which are not cited but part of the contribution:
here clearly textual explanation is missing I assume that the authors have
a clear idea about the value of the figures, and they should share these
ideas more explicitly with the readers
We will extend the manuscript and provide explanations for all the
figures. Furthermore, we will implement the specific remarks and detailed
corrections accordingly.
-
AC1: 'Reply on RC1', Abbas El Hachem, 19 Jul 2022
-
RC2: 'Comment on hess-2022-177', Anonymous Referee #2, 07 Jul 2022
The authors present a method for quality control/assessment of precipitation measurements especially in view a large values at different aggregation times between hours and days partly based on subhourly observations from DWD. As a new and innovative check they introduce radar precipitation estimates (DWD Radolan data set) as an independent set of observatins and discharge measurements at selected catchments which should show a hydrological response in case the observed rain gauge extreme rainfall event is not a fake observation.
Overall this could be a very useful contribution to HESS if the authors would indicate how the proposed procedure may generalize to other observing systems (e.g synop temperature/wind/humidity observations). Furthermore the text is written almost completely from the hydrolocical-engineering point of view: the authors should remember the E in the journals title, namely that other Earth System scientists should also be addressed and not to be scared off. Finally the text is largely written in a funding - interim report style, eg about one third of the figures is just added to the text without any discussion or even referencing in the text. All these points (and those commented in the annotated pdf document) should be revised before publication in HESS can be considered.-
AC2: 'Reply on RC2', Abbas El Hachem, 19 Jul 2022
We thank the reviewer for taking the time to review our manuscript. Our reply to
the reviewer's comments are as follows:Overall this could be a very useful contribution to HESS if the authors
would indicate how the proposed procedure may generalize to other observ-
ing systems (e.g synop temperature, wind and humidity observations).Our method is specifically designed for data with skewed distribution
and a high spatio-temporal variability. Outlier checks for other parameters
such a temperature are much more straight forward to implement. Since
the title specifically mentions precipitation we don’t see the necessity to
discuss how this could be used for other parameters
Our method is specifically designed for data with skewed distribution
and a high spatio-temporal variability. Outlier checks for other parameters
such a temperature are much more straight forward to implement. Since
the title specifically mentions precipitation we don’t see the necessity to
discuss how this could be used for other parameters.Furthermore the text is written almost completely from the hydrolocical-
engineering point of view: the authors should remember the E in the jour-
nals title, namely that other Earth System scientists should also be ad-
dressed and not to be scared off.See reply above, we explicitly present a quality check for precipitation
extremes.Finally the text is largely written in a funding - interim report style,
eg about one third of the figures is just added to the text without any dis-
cussion or even referencing in the text. All these points (and those com-
mented in the annotated pdf document) should be revised before publication
in HESS can be considered.We will provide and discuss all the figures in the manuscript, this was
also pointed out by referee 1. Regarding the “interim report style” we
submitted this as a technical note and not a full-fledged scientific pa-
per, therefore we consider a shorter, more report like text to be justified.
Regarding the comments in the manuscript (supplement), we will imple-
ment/correct/address them accordingly. There are however some remarks
which we would like to comment:Line 21 ff. This remark is correct, but since the DWD gauges assumed
to be installed according to the WMO guidelines, we can rule out such
scenarios to a large degree.
Line 74 ff. Of course, there are other weighing gauges, but we specifically refer to those used by the DWD.
Line 153: A flow chart is already presented in Figure 3.
Line 195: We fitted a spherical or an exponential variogram model
with no nugget to the experimental one. Both models guarantee that
the kriging covariance matrix is positive definite. Moreover we will add a
section showing that the choice of the variogram model has little influence
on the results
-
AC2: 'Reply on RC2', Abbas El Hachem, 19 Jul 2022
Abbas El Hachem et al.
Abbas El Hachem et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 491 | 138 | 14 | 643 | 6 | 10 |
- HTML: 491
- PDF: 138
- XML: 14
- Total: 643
- BibTeX: 6
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1