GloH2O

MSWEP

Multi-Source Weighted-Ensemble Precipitation

Overview

MSWEP is a glob­al pre­cip­i­ta­tion prod­uct with a 3‑hourly 0.1° res­o­lu­tion avail­able from 1979 to ~3 hours from real-time. The prod­uct is unique in that it merges gauge, satel­lite, and reanaly­sis data to obtain the high­est qual­i­ty pre­cip­i­ta­tion esti­mates at every location.
MSWEP incor­po­rates dai­ly gauge obser­va­tions and accounts for gauge report­ing times to reduce tem­po­ral mis­match­es between satel­lite-reanaly­sis esti­mates and gauge obser­va­tions. Near real-time esti­mates are avail­able with a laten­cy of ~3 hours. MSWEP is com­pat­i­ble with GloH2O’s oper­a­tional Mul­ti-Source Weath­er (MSWX); MSWX fore­casts can thus be used to extend MSWEP into the future. MSWEP tends to exhib­it bet­ter per­for­mance than oth­er pre­cip­i­ta­tion prod­ucts in both dense­ly gauged and ungauged regions (see the Per­for­mance sec­tion on this page, Beck et al., 2017, and Beck et al., 2019).

Accurate

Based on state-of-the-art gauge, satel­lite, and reanaly­sis data sources

Gauge-corrected

Dai­ly cor­rec­tions using obser­va­tions from over 77,000 gauges

Near real-time

Near real-time avail­abil­i­ty to sup­port time-crit­i­cal applications

Transparent

Meth­ods pub­lished in open-access peer-reviewed sci­en­tif­ic journals

Performance

Mean tem­po­ral cor­re­la­tion (R2) against two ref­er­ences for MSWEP and oth­er wide­ly used pre­cip­i­ta­tion prod­ucts, demon­strat­ing the high­er over­all accu­ra­cy of MSWEP in dense­ly gauged and ungauged regions, respectively. 

Reference: Stage-IV gauge-radar data

MSWEP V2.2 74%
IMERG Final V6 58%
CMORPH-CRT 53%
GSMaP Std V7 49%
ERA5-HRES 45%
ERA-Inter­im 42%
CHIRPS V2 40%
CPC Uni­fied 37%
TMPA 3B42 V7 34%
PERSIANN-CCS
SM2R-CCI V2

Val­ues rep­re­sent the mean dai­ly tem­po­ral R2 over the US
Adapt­ed from Beck et al. (2019)

Reference: 75,540 gauges worldwide 

MSWEP* V2.8 48%
ERA5-HRES 44%
GPM+SM2R V0.1.0 41%
NCEP GDAS 39%
IMERG Late V6 32%
GSMaP Std V7 31%
CHIRP 23%
SM2R-CCI V2
PERSIANN-CCS

Val­ues rep­re­sent the mean 3‑day tem­po­ral R2
*Ver­sion with­out gauge cor­rec­tions
Pub­li­ca­tion in preparation

Methodology

The fol­low­ing paper pro­vides a detailed descrip­tion of the MSWEP V2.2 methodology: 
MSWEP V2.8 fea­tures new data sources, improved weight maps, less peaky pre­cip­i­ta­tion esti­mates, a longer record (1979–2020), near real-time (NRT) esti­mates, and com­pat­i­bil­i­ty with Mul­ti-Source Weath­er (MSWX). See the tech­ni­cal doc­u­men­ta­tion for the full list of changes and descrip­tions of MSWEP-NRT, the file nam­ing con­ven­tion, and the data format. 
Real-time infor­ma­tion about the sta­tus of the MSWEP-NRT pro­duc­tion sys­tem is avail­able here.

Data license

MSWEP is released under the Cre­ative Com­mons Attri­bu­tion-Non­Com­mer­cial 4.0 Inter­na­tion­al (CC BY-NC 4.0) license. Please con­tact us if you are affil­i­at­ed with a com­mer­cial enti­ty and want to tri­al MSWEP. If you do not have a com­mer­cial affil­i­a­tion and you intend to use the prod­uct for non-com­mer­cial pur­pos­es, please send us a request using the fol­low­ing form. You will receive a link to the Google Dri­ve con­tain­ing MSWEP once your request has been approved. 

Frequently asked questions

Numer­ous grid­ded pre­cip­i­ta­tion prod­ucts have been devel­oped over the last decades. How­ev­er, MSWEP is the only product: 
  1. to merge gauge, satel­lite, and reanaly­sis pre­cip­i­ta­tion esti­mates to enhance the per­for­mance in dense­ly gauged, con­vec­tion-dom­i­nat­ed, and frontal-dom­i­nat­ed weath­er regimes, respectively;
  2. to incor­po­rate dai­ly gauge obser­va­tions and account for gauge report­ing times (which min­i­mizes tem­po­ral mis­match­es between the satel­lite-reanaly­sis esti­mates and the gauge observations);
  3. with com­pat­i­ble medi­um-range fore­cast ensem­bles (updat­ed dai­ly) and sea­son­al fore­cast ensem­bles (updat­ed month­ly; see MSWX).
In addi­tion, MSWEP has a high 3‑hourly 0.1° res­o­lu­tion, glob­al cov­er­age, a long record start­ing in 1979, and near real-time esti­mates. MSWEP also tends to exhib­it bet­ter per­for­mance than oth­er pre­cip­i­ta­tion prod­ucts in both dense­ly gauged and ungauged regions (see the Per­for­mance sec­tion on this page, Beck et al., 2017, and Beck et al., 2019).
Down­load­ing data from shared Google Dri­ve fold­ers is rel­a­tive­ly easy using cloud store man­ag­er soft­ware such as rclone. The fol­low­ing instruc­tions explain how to set up rclone to down­load MSWEP: 
  1. Down­load and install rclone.
  2. Link rclone to your Google account by fol­low­ing the steps in this video.
  3. Access the MSWEP shared fold­er by vis­it­ing it via your brows­er. Check if the shared fold­er is list­ed under “Shared with me” on your Google Dri­ve page.
  4. Con­firm that rclone can find the shared fold­er:
    $ rclone lsd --drive-shared-with-me GoogleDrive:
              -1 2021-02-03 10:14:35        -1 MSWEP_V280
  5. Down­load dai­ly and month­ly MSWEP data from the shared fold­er to your local dri­ve:
    $ rclone sync -v --exclude 3hourly/ --drive-shared-with-me GoogleDrive:/MSWEP_V280 c:/temp/MSWEP_V280
              2021/02/03 11:09:02 INFO  : Past/Monthly/202007.nc: Copied (new)
              2021/02/03 11:09:02 INFO  : Past/Monthly/202002.nc: Copied (new)
              2021/02/03 11:09:02 INFO  : Past/Monthly/202005.nc: Copied (new)
              ...
    If the down­load is inter­rupt­ed, the com­mand can be run again, and files that already exist will be skipped.

Medi­um-range and sea­son­al fore­cast data from GloH2O’s Mul­ti-Source Weath­er (MSWX) prod­uct is com­pat­i­ble with MSWEP. MSWX can thus be used to extend MSWEP into the future. Note, how­ev­er, that some incon­sis­ten­cies between MSWEP and MSWX may be present due to the inclu­sion of gauge and satel­lite data in MSWEP.

Pre­cip­i­ta­tion data for a par­tic­u­lar day (e.g., April 25, 2020) can be down­loaded using rclone as fol­lows:
rclone sync -v --drive-shared-with-me GoogleDrive:/MSWEP_V280/Past/Daily/2020116.nc ./
The data are read and plot using MATLAB as fol­lows:
global_precip = ncread('2020116.nc','precipitation')';
imagesc(global_precip,[0 30]);
colorbar
title('Precipitation on April 25, 2020 (mm/day)')
The same data are read and plot using Python as fol­lows:
from netCDF4 import Dataset
import matplotlib.pyplot as plt

dataset = Dataset('2020116.nc','r')
global_precip = dataset.variables['precipitation'][:]
dataset.close()

plt.plot(global_precip,vmin=0,vmax=30)
plt.colorbar()
plt.title("Precipitation on April 25, 2020 (mm/day)")
plt.show()

We did not cor­rect for gauge under-catch in the lat­est ver­sion of MSWEP (unlike in pre­vi­ous ver­sions) for two main rea­sons: (i) in our expe­ri­ence, users tend to give pref­er­ence to pre­cip­i­ta­tion esti­mates that match gauge obser­va­tions as close­ly as pos­si­ble; and (ii) under-catch cor­rec­tions are gen­er­al­ly sub­ject to con­sid­er­able uncer­tain­ty. To obtain a bias-cor­rect­ed ver­sion of MSWEP — impor­tant for hydro­log­i­cal mod­el­ing — we rec­om­mend using GloH2O’s PBCOR product.

MSWEP V1 lacks cumu­la­tive dis­tri­b­u­tion func­tion (CDF) cor­rec­tions, and there­fore exhibits sig­nif­i­cant driz­zle and less reli­able trends due to changes in the data sources through time. Con­verse­ly, MSWEP V2 match­es the CDF of each dataset com­bi­na­tion to a sin­gle ref­er­ence CDF, result­ing in less driz­zle and a more homo­ge­neous record. See the tech­ni­cal doc­u­men­ta­tion for the full list of changes.

Pre­cip­i­ta­tion prod­uct eval­u­a­tions car­ried out at the dai­ly time scale should account for the report­ing times of gauges (i.e., the end times of the dai­ly accu­mu­la­tions). Dis­re­gard­ing report­ing times in dai­ly eval­u­a­tions can result in sig­nif­i­cant tem­po­ral mis­match­es between the prod­uct and the gauge, which can affect the results of the eval­u­a­tion. Account­ing for report­ing times requires the use of sub-dai­ly data to com­pute new dai­ly accu­mu­la­tions that cor­re­spond to the report­ing times of the gauges. Alter­na­tive­ly, the eval­u­a­tion can be car­ried out at a 3‑day (instead of dai­ly) time scale, which reduces the impact of poten­tial mis­match­es. See Beck et al. (2019) for maps of report­ing times for GHCN‑D and GSOD gauges.

There is cur­rent­ly no oth­er way to down­load MSWEP.

Acknowledgements

ECMWF, NASA, and NOAA and thanked for pro­duc­ing the ERA5, IMERG, and Grid­Sat datasets, respec­tive­ly. The Water Cen­ter for Arid and Semi-Arid Zones in Latin Amer­i­ca and the Caribbean (CAZALAC) and the Cen­tro de Cien­cia del Cli­ma y la Resilien­cia (CR) 2 (FONDAP 15110009) are thanked for shar­ing Mex­i­can and Chilean gauge data, respec­tive­ly. We also acknowl­edge the gauge data providers in the Latin Amer­i­can Cli­mate Assess­ment and Dataset (LACA&D) project: IDEAM (Colom­bia), INAMEH (Venezuela), INAMHI (Ecuador), SENAMHI (Peru), SENAMHI (Bolivia), and DMC (Chile). We wish to thank Ali Ali­jan­ian, Koen Ver­bist, and Piyush Jain for pro­vid­ing addi­tion­al gauge data. 
LL