Data Transformation and Uncertainty in Geostatistical Combination of Radar and Rain Gauges

Abstract Geostatistics provides a popular framework for deriving high-resolution quantitative precipitation estimates (QPE) by combining radar and rain gauge data. However, the skewed and heteroscedastic nature of precipitation is in contradiction to assumptions of classical geostatistics. This study examines the potential of trans-Gaussian kriging to overcome this contradiction. Combination experiments are undertaken with kriging with external drift (KED) using several settings of the Box–Cox transformation. Hourly precipitation data in Switzerland for the year 2008 serve as test bed to compare KED with and without transformation. The impact of transformation is examined with regard to compliance with model assumptions, accuracy of the point estimate, and reliability of the probabilistic estimate. Data transformation improves the compliance with model assumptions, but some level of contradiction remains in situations with many dry gauges. Very similar point estimates are found for KED with untransformed and appropriately transformed data. However, care is needed to avoid excessive transformation (close to log) because this can introduce a positive bias. Strong benefits from transformation are found for the probabilistic estimate, which is rendered positively skewed, sensitive to precipitation amount, and quantitatively more reliable. Without transformation, 44% of all precipitation observations larger than 5 mm h−1 are considered as extremely unlikely by the probabilistic estimate in the test application. Transformation reduces this rate to 4%. Although transformation cannot fully remedy the complications for geostatistics in radar–gauge combination, it seems a useful procedure if realistic and reliable estimation uncertainties are desired, such as for the stochastic simulation of QPE ensembles.

Download Full-text

Assesment of Satellite and Radar Quantitative Precipitation Estimates for Real Time Monitoring of Meteorological Extremes Over the Southeast of Iberian Peninsula

10.20944/preprints201805.0150.v1 ◽

2018 ◽

Author(s):

Fulgencio Cánovas-García ◽

Sandra García-Galiano ◽

Francisco Alonso-Sarría

Keyword(s):

Iberian Peninsula ◽

Real Time ◽

Extreme Rainfall ◽

Rain Gauge ◽

Real Time Monitoring ◽

Rain Gauges ◽

Extreme Precipitation Events ◽

Empirical Calibration ◽

Precipitation Estimates ◽

Quantitative Precipitation Estimates

QPEs (Quantitative Precipitation Estimates) obtained from remote sensing or ground-based radars could complement or even be an alternative to rain gauge readings. However, to be used in operational applications, a validation process has to be carried out, usually by comparing their estimates with those of a rain gauges network. In this paper, the accuracy of two QPEs are evaluated for three extreme precipitation events in the last decade in the southeast of the Iberian Peninsula. The first QPE is PERSIANN-CCS, a satellite-based QPE. The second is a meteorological radar with Doppler capabilities that works in the C band. Pixel-to-point comparisons are made between the values offered by the QPEs and those obtained by two networks of rain gauges. The results obtained indicate that both QPEs were well below the rain gauge values, especially in extreme rainfall time slots. There seems to be a weak linear association between the value of the discrepancies and the precipitation value of the QPEs. It does not seem that radar is more accurate than PERSIANN-CCS, despite its larger spatial resolution and its commonly higher effectiveness. The main conclusion is that neither PERSIANN-CCS nor radar, without empirical calibration, are acceptable QPEs for the real-time monitoring of meteorological extremes in the southeast of the Iberian Peninsula.

Download Full-text

Comparative Analysis of Rain Gauge and Radar Precipitation Estimates towards Rainfall-Runoff Modelling in a Peri-Urban Basin in Attica, Greece

Hydrology ◽

10.3390/hydrology8010029 ◽

2021 ◽

Vol 8 (1) ◽

pp. 29

Author(s):

Apollon Bournas ◽

Evangelos Baltas

Keyword(s):

Research Work ◽

Weather Radar ◽

Rain Gauge ◽

Spatial And Temporal Variability ◽

Rainfall Runoff ◽

Runoff Generation ◽

Surface Precipitation ◽

Precipitation Estimates ◽

Quantitative Precipitation Estimates ◽

The Impact

In this research work, an analysis is conducted concerning the impact on rainfall-runoff simulations of utilizing rain gauge precipitation measurements against weather radar quantitative precipitation estimates. The study area is the Sarantapotamos river basin, a peri-urban basin located in the greater area of Athens, and measurements from a newly installed X-Band weather radar system, referred to as rainscanner, along with ground rain gauge stations were used. Rainscanner, in contrast to rain gauges, is able to provide with higher resolution surface precipitation datasets, but due to signal errors, uncertainty is involved, and thus proper calibration and evaluation of these estimates must be first performed. In this context, this research work evaluates the impact of adopting different precipitation datasets and interpolation methods for generating runoff, through the use of a lumped based rainfall-runoff model. Initially, the analysis focuses on the correlation between the rain gauge and the rainscanner estimations for each station, as well as for the calculated mean areal precipitation. The results of the rainfall-runoff simulations show that even though a different spatial and temporal variability of the rainfall field is calculated through the two datasets, in a lumped-based scheme, the most important factor that dictates the runoff generation is the amount of total precipitation.

Download Full-text

Assessment of Snowfall Accumulation from Satellite and Reanalysis Products Using SNOTEL Observations in Alaska

Remote Sensing ◽

10.3390/rs13152922 ◽

2021 ◽

Vol 13 (15) ◽

pp. 2922

Author(s):

Yang Song ◽

Patrick D. Broxton ◽

Mohammad Reza Ehsani ◽

Ali Behrangi

Keyword(s):

Snow Water Equivalent ◽

Stage Iv ◽

Microwave Remote Sensing ◽

Rain Gauge ◽

Snow Accumulation ◽

Correction Factors ◽

Rainfall Analysis ◽

Rain Gauges ◽

Snow Properties ◽

Precipitation Estimates

The combination of snowfall, snow water equivalent (SWE), and precipitation rate measurements from 39 snow telemetry (SNOTEL) sites in Alaska were used to assess the performance of various precipitation products from satellites, reanalysis, and rain gauges. Observation of precipitation from two water years (2018–2019) of a high-resolution radar/rain gauge data (Stage IV) product was also utilized to give insights into the scaling differences between various products. The outcomes were used to assess two popular methods for rain gauge undercatch correction. It was found that SWE and precipitation measurements at SNOTELs, as well as precipitation estimates based on Stage IV data, are generally consistent and can provide a range within which other products can be assessed. The time-series of snowfall and SWE accumulation suggests that most of the products can capture snowfall events; however, differences exist in their accumulation. Reanalysis products tended to overestimate snow accumulation in the study area, while the current combined passive microwave remote sensing products (i.e., IMERG-HQ) underestimate snowfall accumulation. We found that correction factors applied to rain gauges are effective for improving their undercatch, especially for snowfall. However, no improvement in correlation is seen when correction factors are applied, and rainfall is still estimated better than snowfall. Even though IMERG-HQ has less skill for capturing snowfall than rainfall, analysis using Taylor plots showed that the combined microwave product does have skill for capturing the geographical distribution of snowfall and precipitation accumulation; therefore, bias adjustment might lead to reasonable precipitation estimates. This study demonstrates that other snow properties (e.g., SWE accumulation at the SNOTEL sites) can complement precipitation data to estimate snowfall. In the future, gridded SWE and snow depth data from GlobSnow and Sentinel-1 can be used to assess snowfall and its distribution over broader regions.

Download Full-text

Improving Multisensor Precipitation Estimation via Adaptive Conditional Bias–Penalized Merging of Rain Gauge Data and Remotely Sensed Quantitative Precipitation Estimates

Journal of Hydrometeorology ◽

10.1175/jhm-d-19-0129.1 ◽

2019 ◽

Vol 20 (12) ◽

pp. 2347-2365 ◽

Cited By ~ 1

Author(s):

Ali Jozaghi ◽

Mohammad Nabatian ◽

Seongjin Noh ◽

Dong-Jun Seo ◽

Lin Tang ◽

...

Keyword(s):

Computational Cost ◽

Mean Field ◽

Remotely Sensed ◽

Rain Gauge ◽

Rain Gauge Data ◽

Gauge Data ◽

Precipitation Estimation ◽

Precipitation Estimates ◽

Radar Satellite ◽

Quantitative Precipitation Estimates

Abstract We describe and evaluate adaptive conditional bias–penalized cokriging (CBPCK) for improved multisensor precipitation estimation using rain gauge data and remotely sensed quantitative precipitation estimates (QPE). The remotely sensed QPEs used are radar-only and radar–satellite-fused estimates. For comparative evaluation, true validation is carried out over the continental United States (CONUS) for 13–30 September 2015 and 7–9 October 2016. The hourly gauge data, radar-only QPE, and satellite QPE used are from the Hydrometeorological Automated Data System, Multi-Radar Multi-Sensor System, and Self-Calibrating Multivariate Precipitation Retrieval (SCaMPR), respectively. For radar–satellite fusion, conditional bias–penalized Fisher estimation is used. The reference merging technique compared is ordinary cokriging (OCK) used in the National Weather Service Multisensor Precipitation Estimator. It is shown that, beyond the reduction due to mean field bias (MFB) correction, both OCK and adaptive CBPCK additionally reduce the unconditional root-mean-square error (RMSE) of radar-only QPE by 9%–16% over the CONUS for the two periods, and that adaptive CBPCK is superior to OCK for estimation of hourly amounts exceeding 1 mm. When fused with the MFB-corrected radar QPE, the MFB-corrected SCaMPR QPE for September 2015 reduces the unconditional RMSE of the MFB-corrected radar by 4% and 6% over the entire and western half of the CONUS, respectively, but is inferior to the MFB-corrected radar for estimation of hourly amounts exceeding 7 mm. Adaptive CBPCK should hence be favored over OCK for estimation of significant amounts of precipitation despite larger computational cost, and the SCaMPR QPE should be used selectively in multisensor QPE.

Download Full-text

Commercial microwave links instead of rain gauges: fiction or reality?

Water Science & Technology ◽

10.2166/wst.2014.466 ◽

2014 ◽

Vol 71 (1) ◽

pp. 31-37 ◽

Cited By ~ 27

Author(s):

Martin Fencl ◽

Jörg Rieckermann ◽

Petr Sýkora ◽

David Stránský ◽

Vojtěch Bareš

Keyword(s):

Rainfall Variability ◽

A Priori ◽

Rain Gauges ◽

Precipitation Estimates ◽

Spatio Temporal ◽

Areal Rainfall ◽

Quantitative Precipitation Estimates ◽

Commercial Microwave Links ◽

Unique Dataset

Commercial microwave links (MWLs) were suggested about a decade ago as a new source for quantitative precipitation estimates (QPEs). Meanwhile, the theory is well understood and rainfall monitoring with MWLs is on its way to being a mature technology, with several well-documented case studies, which investigate QPEs from multiple MWLs on the mesoscale. However, the potential of MWLs to observe microscale rainfall variability, which is important for urban hydrology, has not been investigated yet. In this paper, we assess the potential of MWLs to capture the spatio-temporal rainfall dynamics over small catchments of a few square kilometres. Specifically, we investigate the influence of different MWL topologies on areal rainfall estimation, which is important for experimental design or to a priori check the feasibility of using MWLs. In a dedicated case study in Prague, Czech Republic, we collected a unique dataset of 14 MWL signals with a temporal resolution of a few seconds and compared the QPEs from the MWLs to reference rainfall from multiple rain gauges. Our results show that, although QPEs from most MWLs are probably positively biased, they capture spatio-temporal rainfall variability on the microscale very well. Thus, they have great potential to improve runoff predictions. This is especially beneficial for heavy rainfall, which is usually decisive for urban drainage design.

Download Full-text

The Contribution of Rain Gauges in the Calibration of the IMERG Product: Results from the First Validation over Spain

Journal of Hydrometeorology ◽

10.1175/jhm-d-19-0116.1 ◽

2020 ◽

Vol 21 (2) ◽

pp. 161-182 ◽

Cited By ~ 4

Author(s):

Francisco J. Tapiador ◽

Andrés Navarro ◽

Eduardo García-Ortega ◽

Andrés Merino ◽

José Luis Sánchez ◽

...

Keyword(s):

Rain Gauge ◽

Surface Precipitation ◽

Rain Gauges ◽

Precipitation Measurement ◽

Precipitation Estimates ◽

Level 3 ◽

Hydrologic Impacts ◽

A Minor ◽

Global Precipitation Measurement ◽

Global Precipitation

AbstractAfter 5 years in orbit, the Global Precipitation Measurement (GPM) mission has produced enough quality-controlled data to allow the first validation of their precipitation estimates over Spain. High-quality gauge data from the meteorological network of the Spanish Meteorological Agency (AEMET) are used here to validate Integrated Multisatellite Retrievals for GPM (IMERG) level 3 estimates of surface precipitation. While aggregated values compare notably well, some differences are found in specific locations. The research investigates the sources of these discrepancies, which are found to be primarily related to the underestimation of orographic precipitation in the IMERG satellite products, as well as to the number of available gauges in the GPCC gauges used for calibrating IMERG. It is shown that IMERG provides suboptimal performance in poorly instrumented areas but that the estimate improves greatly when at least one rain gauge is available for the calibration process. A main, generally applicable conclusion from this research is that the IMERG satellite-derived estimates of precipitation are more useful (r2 > 0.80) for hydrology than interpolated fields of rain gauge measurements when at least one gauge is available for calibrating the satellite product. If no rain gauges were used, the results are still useful but with decreased mean performance (r2 ≈ 0.65). Such figures, however, are greatly improved if no coastal areas are included in the comparison. Removing them is a minor issue in terms of hydrologic impacts, as most rivers in Spain have their sources far from the coast.

Download Full-text

Precipitation Analysis over the French Alps Using a Variational Approach and Study of Potential Added Value of Ground-Based Radar Observations

Journal of Hydrometeorology ◽

10.1175/jhm-d-16-0144.1 ◽

2017 ◽

Vol 18 (5) ◽

pp. 1425-1451 ◽

Cited By ~ 4

Author(s):

Camille Birman ◽

Fatima Karbou ◽

Jean-François Mahfouf ◽

Matthieu Lafaysse ◽

Yves Durand ◽

...

Keyword(s):

Snow Depth ◽

Positive Impact ◽

Weather Prediction ◽

Rain Gauge ◽

Added Value ◽

Background Information ◽

Time Step ◽

Rain Gauges ◽

Weather Radars ◽

Precipitation Estimates

Abstract A one-dimensional variational data assimilation (1DVar) method to retrieve profiles of precipitation in mountainous terrain is described. The method combines observations from the French Alpine region rain gauges and precipitation estimates from weather radars with background information from short-range numerical weather prediction forecasts in an optimal way. The performance of this technique is evaluated using measurements of precipitation and of snow depth during two years (2012/13 and 2013/14). It is shown that the 1DVar model allows an effective assimilation of measurements of different types, including rain gauge and radar-derived precipitation. The use of radar-derived precipitation rates over mountains to force the numerical snowpack model Crocus significantly reduces the bias and standard deviation with respect to independent snow depth observations. The improvement is particularly significant for large rainfall or snowfall events, which are decisive for avalanche hazard forecasting. The use of radar-derived precipitation rates at an hourly time step improves the time series of precipitation analyses and has a positive impact on simulated snow depths.

Download Full-text

Assessing satellite-based precipitation estimates in the Southern Appalachian mountains using rain gauges and TRMM PR

Advances in Geosciences ◽

10.5194/adgeo-25-143-2010 ◽

2010 ◽

Vol 25 ◽

pp. 143-153 ◽

Cited By ~ 39

Author(s):

O. P. Prat ◽

A. P. Barros

Keyword(s):

Appalachian Mountains ◽

Rain Gauge ◽

Tropical Storm ◽

Southern Appalachian Mountains ◽

Rain Gauges ◽

Orographic Rainfall ◽

Southern Appalachian ◽

Trmm Pr ◽

Precipitation Estimates ◽

Order Of Magnitude

Abstract. A study was performed using the first full year of rain gauge records from a newly deployed network in the Southern Appalachian mountains. This is a region characterized by complex topography with orographic rainfall enhancement up to 300% over small distances (<8 km). Rain gauge observations were used to assess precipitation estimates from the Precipitation Radar (PR) on board of the TRMM satellite, specifically the TRMM PR 2A25 precipitation product. Results show substantial differences between annual records and isolated events (e.g. tropical storm Fay). An overall bias of −27% was found between TRMM PR 2A25 rain rate and rain gauge rain rates for the complete one year of study (−59% for tropical storm Fay). Besides differences observed for concurrent observations by the satellite and the rain gauges, a large number of rainfall events is detected independently by either one of the observing systems alone (rain gauges: 50% of events are missed by TRMM PR; TRMM PR: 20% of events are not detected by the rain gauges), especially for light rainfall conditions (0.1–2mm/h) that account for more than 80% of all the missed satellite events. An exploratory investigation using a microphysical model along with TRMM reflectivity factors at selected heights was conducted to determine the shape of the drop size distribution (DSD) that can be applied to reduce the difference between TRMM estimates and rain gauge observations. The results suggest that the critical DSD parameter is the number concentration of very small drops. For tropical storm Fay an increase of one order of magnitude in the number of small drops is apparently needed to capture the observed rainfall rate regardless of the value of the measured reflectivity. This is consistent with DSD observations that report high concentrations of small and/or midsize drops in the case of tropical storms.

Download Full-text

Evaluation of radar-gauge merging methods for quantitative precipitation estimates

Hydrology and Earth System Sciences Discussions ◽

10.5194/hessd-5-2975-2008 ◽

2008 ◽

Vol 5 (5) ◽

pp. 2975-3003 ◽

Cited By ~ 9

Author(s):

E. Goudenhoofdt ◽

L. Delobbe

Keyword(s):

Bias Correction ◽

Mean Field ◽

Daily Rainfall ◽

Absolute Error ◽

Rain Gauge ◽

Radar Observations ◽

Geostatistical Methods ◽

Precipitation Estimates ◽

Field Bias ◽

Quantitative Precipitation Estimates

Abstract. Accurate quantitative precipitation estimates are of crucial importance for hydrological studies and applications. When spatial precipitation fields are required, rain gauge measurements are often combined with weather radar observations. In this paper, we evaluate several radar-gauge merging methods with various degrees of complexity: from mean field bias correction to geostatical merging techniques. The study area is the Walloon region of Belgium, which is mostly located in the Meuse catchment. Observations from a C-band Doppler radar and a dense rain gauge network are used to retrieve daily rainfall accumulations over this area. The relative performance of the different merging methods are assessed through a comparison against daily measurements from an independent gauge network. A 3-year verification is performed using several statistical quality parameters. It appears that the geostatistical merging methods perform best with the mean absolute error decreasing by 40% with respect to the original data. A mean field bias correction still achieves a reduction of 25%. A seasonal analysis shows that the benefit of using radar observations is particularly significant during summer. The effect of the network density on the performance of the methods is also investigated. For this purpose, a simple approach to remove gauges from a network is proposed. The analysis reveals that the sensitivity is relatively high for the geostatistical methods but rather small for the simple methods. The geostatistical methods give the best results for all network densities except for a very low density of 1 gauge per 500 km2 where a range-dependent adjustment complemented with a static local bias correction performs best.

Download Full-text

Precipitation Estimates from Commercial Microwave Links: Practical Approaches to Wet-antenna Correction

10.31224/osf.io/nt5wg ◽

2021 ◽

Author(s):

Jaroslav Pastorek ◽

Martin Fencl ◽

Jörg Rieckermann ◽

Vojtěch Bareš

Keyword(s):

Rainfall Intensity ◽

Rain Gauge ◽

Model Parameters ◽

Rainfall Events ◽

Precipitation Estimates ◽

Attenuation Models ◽

Using Data ◽

Quantitative Precipitation Estimates ◽

Rain Gauge Network ◽

Commercial Microwave Links

An inadequate correction for wet antenna attenuation (WAA) often causes a notable bias in quantitative precipitation estimates (QPEs) from commercial microwave links (CMLs) limiting the usability of these rainfall data in hydrological applications. This paper analyzes how WAA can be corrected without dedicated rainfall monitoring for a set of 16 CMLs. Using data collected over 53 rainfall events, the performance of six empirical WAA models was studied, both when calibrated to rainfall observations from a permanent municipal rain gauge network and when using model parameters from the literature. The transferability of WAA model parameters among CMLs of various characteristics has also been addressed. The results show that high-quality QPEs with a bias below 5% and RMSE of 1 mm/h in the median could be retrieved, even from sub-kilometer CMLs where WAA is relatively large compared to raindrop attenuation. Models in which WAA is proportional to rainfall intensity provide better WAA estimates than constant and time-dependent models. It is also shown that the parameters of models deriving WAA explicitly from rainfall intensity are independent of CML frequency and path length and, thus, transferable to other locations with CMLs of similar antenna properties.

Download Full-text