Convective Rainfall Rate from Cloud Physical Properties
(NWC/GEO  CRR-Ph,v2018)

 

Table of contents

1. Goal of CRR-Ph product  
2. CRR-Ph algorithm description  
3. List of inputs for CRR-Ph  
4. Description of CRR-Ph outputs 
5. Example of CRR-Ph visualisation 
 

Access to "Algorithm Theoretical Basis Document for the Precipitation Product Processors of the NWC/GEO" for a more detailed description of the algorithm.

 

 

1. Goal of CRR-Ph product

Convective Rainfall Rate from Cloud Physical Properties (CRR-Ph) product, developed within the NWC SAF context, is a Nowcasting tool that provides information on convective, and stratiform associated to convection, instantaneous rain rates and hourly accumulations.

 

2. CRR-Ph algorithm description

Day time

The main inputs of CRR-Ph product are the cloud top microphysical properties generated by NWC/GEO CMIC Cloud Top Phase, Effective Radius (Reff) and Cloud Optical Thickness (COT).

The first step of the processing of the product is the computation of Cloud Water Path (CWP). Then, depending on some Reff and CWP thresholds, the precipitation area is enclosed. Only in those pixels belonging to the precipitation area, the rain rate is computed.

To assign an instantaneous rain rate to each pixel, the following relationship between CWP and precipitation intensity, obtained through a calibration process that uses radar data, is applied:

 

where:

RR – Rain rates (mmh-1)

CWP – Cloud Water Path (gm-2)

 

In the following step, taking into account the instantaneous rain rates computed in the last hour time interval, hourly accumulations are computed through a trapezoidal integration.

For a better precipitation area location a parallax correction can be applied to this product. This option is chosen by the user through the product model configuration file and it is applied by default. When the Parallax Correction is working, a spatial shift is applied to every pixel with rain rate greater than 0 mm/h.

At this stage, the CRR-Ph precipitation pattern computed in the previous step is combined with a precipitation pattern derived through a lightning algorithm. This step is optional.

The lightning algorithm assigns a rain rate to every lightning depending on:

  • the time distance (delta tau) between the lightning event and scanning time of the processing region centre.
  • the location of the lightning
  • the spatial density of lightning in a time interval

Once the precipitation pattern has been computed, it is compared to the CRR-Ph precipitation pattern in order to obtain the final product. This final product contains the highest rain rate of the two.

The parameters used by this product are highly dependant on GEO solar channels. For this reason this product can only be generated during daytime.

It has been seen that this product provides erroneous rain rates for poor illumination conditions. For this reason an Illumination Conditions Quality flag, that provides information on the confidence of the estimated rain rates, is computed and delivered with the product.

Night time

For night-time, when no microphysical information can be retrieved, a Kernel Ridge Regression is used to establish a relation between rain rates and brightness temperatures of IR10.8 and (IR10.8-WV6.2)  SEVIRI channels. 

Cloud Water Path on a day time database is used to stablish thresholds used to exclude points where the product is going to be  calculated at nigth time.

Transition between day to night algorithm is produced whenever the solar zenith angle is greater than 70º. It is also possible to configure the product to only use the night algorithm by setting the sun zenith angle threshold to 0. This would avoid discontinuities in the product at the day/night transition on the cost of degrading performance during day time.

3. List of inputs for CRR-Ph

NWC/GEO CMIC microphysical parameters

  • CMIC Phase, COT and Reff parameters are mandatory inputs to CRR-Ph. 
  • CMIC Phase is mandatory input to compute the CRR-Ph night algorithm 

Satellite imagery

IR10.8 SEVIRI brightness temperature at full IR spatial resolution is a mandatory input to compute Parallax Correction. IR10.8 and WV6.2 are mandatory inputs to compute the CRR-Ph night algorithm 

Numerical model

Temperature at 1000, 925, 850, 700, 500, 400, 300, 250 and 200 hPa

Geopotential at 1000, 925, 850, 700, 500, 400, 300, 250 and 200 hPa

This information is used by default for parallax correction. In case of lack of NWP parameters parallax correction will be run using a climatological profile.

Ancillary data sets

Climatological profile is necessary as a back up for Parallax correction in case NWP is not available. This information is included in the software package.

Lightning information file for CRR-Ph product

A file with information on every lightning occurred in a time interval is mandatory to choose the option of adjusting the CRPh precipitation pattern with the lightning information.

Model configuration file for PPh

PPh model configuration file contains configurable system parameters in the generation process of both PC-Ph and CRR-Ph products. The CRR-Ph product related parameters refers to ancillary datasets, numerical model data, lightning algorithm and parallax correction. The complete list of these parameters and the explanation of the most useful ones are available in the"User Manual for the Precipitation Product Processors of the NWC/GEO".

 

4. Description of CRR-Ph outputs

CRR-Ph product is coded in NetCDF format. The available outputs are the following:

crrph_intensity

crrph_intensity output contains the rainfall rates associated to convective clouds. The rain rates can take values from 0.0 to 50.0 mm/h with a step of 0.2 mm/h.

 

DAY ALGORITHM

GEO-CMIC-INPUT PHASE

PHASE INPUT CLASS

COT OR REFF FROM CMIC

CRR-PH OUTPUT

Liquid

1

NO DATA

NO DATA

DATA AVAILABLE

crrph_intensity(mm/h)=scale_factor*counts + add_offset

Ice 

2

NO DATA

NO DATA

DATA AVAILABLE

crrph_intensity(mm/h)=scale_factor*counts + add_offset

Mixed

3

NO DATA

NO DATA

DATA AVAILABLE

crrph_intensity(mm/h)=scale_factor*counts + add_offset

Cloud-free

4

NOT APPLICABLE

0

Undefined

5

NOT APPLICABLE

NO DATA

No data or corrupted data

FillValue

NOT APPLICABLE

NO DATA

 

NIGHT ALGORITHM

GEO-CMIC-INPUT PHASE

PHASE INPUT CLASS

CRRC-PH OUTPUT

Liquid

1

crrph_intensity(mm/h)=scale_factor * counts + add_offset

Ice 

2

crrph_intensity(mm/h)=scale_factor * counts + add_offset

Mixed

3

crrph_intensity(mm/h)=scale_factor * counts + add_offset

Cloud-free

4

0

Undefined

5

NO DATA

No data or corrupted data

FillValue

NO DATA

 

where:   

scale_factor = 0.1   

add_offset = 0.0 

crrph_accum   

crrph_accum output provides hourly accumulations associated to convective clouds and computed using the rainfall rates from the images in the last hour. This output provides precipitation accumulations from 0.0 to 51.0 mm with a step of 0.2 mm

crrph_iqf

crrph_iqf is a flag that provides information on the confidence that a user can have on the estimated rain rates according to the illumination conditions and viewing angles.

crrph_status_flag

crrph_status_flag provides information on whether parallax correction and lightning algorithm have been applied, the inputs used to compute the product and some other aspects:

 

Data Availability:

Bit 0:            Reff or COT not computed (out of cloud, night time, phase not defined)

Bit 1:            Phase not computed or undefined

Bit 2:            IR band missing (used in parallax correction)

Applied Correction:

Bit 3:            Parallax correction applied

Use of optional data:

Bit 6:            Lightning data used

Other information

Bit 8:             crr_intensity was a hole because of the parallax correction, and then was filled by the median filter

Bit 9, 10, 11: Use of bands for accumulation

                                1: All required bands were available

                                2: One previous CRR band is missing

                                3: At least two previous CRR bands are missing (no consecutive)

                                4: At least two previous CRR bands are missing (some are consecutive)

Bit 12:          Accumulation quality flag. Set to 1 if:

                                not all crr values are available to perform the accumulation,

                                OR

                                any of the crr_intensity values was set to 0 due to filtering process

                                OR

                                Any of the crr_intensity values was a hole because parallax correction

 

5. Example of CRR-Ph visualisation

CRR-Ph Instantaneous rain rates

Below are shown two different images of the instantaneous rain rates CRR-Ph product, day and night algorithms. It has been obtained at full resolution.

 

 

 

Figure 1: CRR-Ph instantaneous rain rates for 10th May 2016 at 12:00 UTC  over Spain, day algorithm

 

Figure 2: CRR-Ph instantaneous rain rates for 10th February 2019 at 21:00 UTC  over Europe, night algorithm

 

CRR-Ph Hourly accumulations

Below is shown an image of the hourly accumulation CRR-Ph product. It has been obtained at full resolution.

 

Figure 3: CRR-Ph hourly accumulations for 10th February 2019 at 21:00 UTC  over Europe

 

 

 

References

  • Pilewskie, P. and Twomey, S., 1987. Discrimination of ice from water in clouds by optical remote sensing. Atmos. Res., 21:113-122
  • Rosenfeld, D., and G. Gutman, 1994. Retrieving microphysical properties near the tops of potential rain clouds by multispectral analysis of AVHRR data, Atmos. Res., 34, 259–283, doi:10.1016/0169-8095(94)90096-5.
  • Roebeling, R.A., A.J. Feijt and P. Stammes, Cloud property retrievals for climate monitoring: implications of differences between SEVIRI on METEOSAT-8 and AVHRR on NOAA-17 J. Geophys. Res., 2006, 111, 20210, doi:10.1029/2005JD006990.
  • Roebeling, R.A. and I. Holleman, 2009. SEVIRI rainfall retrieval and validation using weather radar observations. J. Geophys. Res., D2120, 114.
  • Tapia, A., Smith, J. A., Dixon, M., 1998: Estimation of Convective Rainfall from Lightning Observations, Bull. American Meteorological Society, Vol. 37, pp. 1497-1509.
  • Daniel Rosenfeld, William L. Woodley, Amit Lerner, Guy Kelman, Daniel T. Lindsey, 2008. Satellite detection of severe convective storms by their retrieved vertical profiles of cloud particle effective radius and thermodynamic phase.  J. Geophys. Res. D4, 113.
  • Gutierrez, J. M. and Aguado, F.: Quality image for the Spanish Radar National Composite, Proceedings of ERAD 2006, 318-320.
  • Algorithm Theoretical Basis Document for "Precipitation products from Cloud Physical Properties" (NWC-CDOP2-GEO-AEMET-SCI-ATBDPrecipitation_v2.1), v2018
  • Product User Manual for "Precipitation products from Cloud Physical Properties" ( NWC-CDOP3-GEO-AEMET-SCI-UM-Precipitation_v1.0 ), v2018
  • Validation Report for "Precipitation products from Cloud Physical Properties" (PC-Ph v2.1 & CRR-Ph v2.1), v2018