OceanRAIN: Ocean Rainfall And Ice-phase precipitation measurement Network


RESTRICTED only accessable in CEN net or via CliSAP login  What does that mean?

Data access via file system: /data/icdc/atmosphere/oceanrain/


OceanRAIN—the Ocean Rainfall And Ice-phase precipitation measurement Network—provides in-situ along-track shipboard data of precipitation, evaporation and the resulting freshwater flux in 1-min resolution over the global oceans. All routinely measured atmospheric and oceanographic state variables along with those required to derive the turbulent heat fluxes are included.

The precipitation parameter is based on measurements from the optical disdrometer ODM470 that is specifically designed for all-weather shipboard operation. The rain, snow and mixed-phase precipitation occurrence, intensity and accumulation are derived from particle size distributions (PSD). Additionally, microphysical parameters and radar-related parameters are provided.

Three OceanRAIN version 1.0 products are available in NetCDF and Ascii formats:

  • OceanRAIN-W: Water cycle components, continuous in time, 73 parameters, > 6.83 million minutes of data, 1-minute resolution.
  • OceanRAIN-M: Precipitation microphysical parameters, discontinuous in time, 37 parameters plus 128 size-bin number concentration particle size distributions, > 692.000 minutes, 1-minute resolution.
  • OceanRAIN-R: Optical disdrometer raw data and precipitation microphysical parameters, discontinuous in time, 37 parameters plus 128 size-bin raw particle count size distributions, > 692.000 minutes,  1-minute resolution.

OceanRAIN Version 1.0 contains 73 parameters plus PSD data in 128 size bins. The time period from 06/2010 to 04/2017 comprises more than 6.83 million minutes of data from eight ships with precipitation observed in about 10% of the time. The research vessels sail the global oceans during all seasons, avoiding the fair-weather bias and thus covering the entire spectrum of weather events.

OceanRAIN provides in-situ water cycle surface reference data for satellite product validation and retrieval calibration of the GPM (Global Precipitation Measurement) era, to analyze the point-to-area representativeness of precipitation and to improve our understanding of water cycle processes over the global oceans. Moreover, the data can be applied to evaluate re-analysis and climate model data.

The data set is funded by Initiative Pro Klima (www.initiativeproklima.de) and the CliSAP excellence cluster at the University of Hamburg. More information on OceanRAIN data, instrumentation and the processing chain are available via www.oceanrain.org.

Last update of this data set at ICDC: Nov 14, 2017.

To top


Tables showing all parameters can be found in the README file.

To top

Coverage, temporal and spatial resolution

Period and temporal resolution: 

  • June 2010 - April 2017

  • 1-minute resolution

Coverage and spatial resolution: 

  • Global oceans

  • Spatial resolution: 1-minute resolution, along-track ship data for eight ships, global oceans

  • Geographic latitude: -90.0°N to 90.0°N

  • Geographic longitude: -180.0° E to 180.0°W

  • Dimension: along-track point ship data for eight ships

  • Altitude: 0.0 m

To top

Data quality

Quality control is of major importance because the raw data ingested into OceanRAIN originates from four different shipboard data streams. These are collocated into a 1-minute resolution match-up data base before deriving additional parameters. Therefore, precise temporal synchronization of the individual data streams is crucial and achieved through the use of a joint IP time server protocol onboard the ships. Where this is not possible, manual synchronization of clocks is performed in regular time intervals. The navigation data (NAV) is inspected for the existence of date and time recorded in UT, latitude, longitude and ship heading. Non-sequential, missing or corrupt data is corrected where possible or flagged missing otherwise. Duplicate time steps are removed. The meteorological and oceanographic data stream (MET) is inspected for out of range and missing values (Klepp, 2015). Suspect and erroneous data are rigorously excluded from the data records using cascading automated quality control and manual visualization screening procedures. This ensures that derived parameters (e.g. evaporation) are only calculated once the input parameters successfully passed the quality control. The optical disdrometer precipitation data (ODM) is inspected and corrected for unrealistically high single-minute spikes. Such occurrences are however not entirely deleted from the database as they can be identified by their non-zero number of particles and bins and their theoretical rain and snowfall rate that got rejected in the final precipitation rate value. The automatically derived precipitation phase probability is further inspected to be in agreement with the derived rainfall and snowfall rate to avoid unrealistic minute-to-minute variations in the precipitation record when mixed-phase precipitation is likely to be present. Changes made can be tracked by the records of the theoretical rainfall and snowfall rate and the precipitation phase flag.

Each minute in the data records is additionally labelled with two precipitation flags. Flag1 assigns the precipitation phase or instruments condition while flag2 distinguishes the precipitation events (Tables 1 and 2). Stratifying the data by these flags is recommended to either identify common precipitation phases or precipitation rate regimes. This is especially important for very light precipitation because two sources for 0.00 mm/h precipitation rates exist. A precipitation minute can be zero if precipitation particles were falling but result in no accumulation (flag1=0,1,2) or the minute contains no precipitation (true-zero) with flag1 = 3. Furthermore, the disdrometer is able to detect precipitation as low as 0.01 mm/h while gauges usually measure from 0.1 mm/h onward. Therefore, a spurious precipitation to unknown source for the signal is introduced (flag2=11) where it is left to the user to consider these insignificant precipitation rates as real precipitation. Harbor times (flag1=5) are filled with missing values because either the NAV and MET data streams are unavailable or maintenance of the instruments are producing data outages or erroneous data. Therefore, also the ODM data stream is set to missing values.

Optical disdrometer calibration is performed before and after shipboard operation whenever the ships are accessible during port or maintenance periods. The twofold procedure at first comprises hardware calibration of the optical axis, the reference voltage and dropping test spheres of increasing size. This is followed by an outdoor test site calibration using real rainfall events wind speed conditions below 5 m/s and a reference ship rain gauge with side collectors for accumulation comparison. The bias in terms of percentage deviation between both instruments is in the order of 2% (Klepp, 2015). The calibration drift after shipboard operation is negligible in most cases because the reference voltage is continuously checked and adjusted during the cruises if necessary. Underway lens cleaning is required in two months intervals on average as indicated by a reference voltage drop towards a quality check threshold of 3 Volt.

To top


Name: Dr. Christian Klepp
Institute: CliSAP / CEN
Email: christian.klepp (at) uni-hamburg.de

Name: Remon Sadikni
Institute: CliSAP / CEN / ICDC
Email: remon.sadikni (at) uni-hamburg.de

To top


Data citation

When using the data please cite:

  • Klepp, C., 2015: The Oceanic Shipboard Precipitation Measurement Network for Surface Validation – OceanRAIN. Atmos. Res., Special issue of the International Precipitation Working Group (IPWG), 163, 74-90, doi: 10.1016/j.atmosres.2014.12.014.
  • OceanRAIN-W DOI, submitted
  • OceanRAIN-M DOI, submitted
  • OceanRAIN-R DOI, submitted

until Sci. Data paper is published

and acknowledge as follows:

OceanRAIN version 1.0 is designed, developed and created by Christian Klepp at the University of Hamburg, Germany. Contributions to the data set OceanRAIN are provided by Simon Michel (University Hamburg), Nicole Albern and Jörg Burdanowitz (University Hamburg and Max Planck Institute for Meteorology, Hamburg, Germany, Alain Protat and Valentin Louf (Bureau of Meteorology, Melbourne Australia). Technical support is provided by Andrea Dahl (Eigenbrodt, Königsmoor, Germany). Support and is further provided by Tanja Thiele (Petronord, Initiative Pro Klima, Hamburg, Germany) and Stephan Bakan (Max Planck Institute for Meteorology, Hamburg, Germany). The data set is funded by Initiative Pro Klima (www.initiativeproklima.de), the CliSAP excellence cluster at the University of Hamburg and the Max Planck Institute for Meteorology in Hamburg, Germany. More information on OceanRAIN data, instrumentation and the processing chain are available via www.oceanrain.org.

To top

###BACKLINK###     ###TOPLINK###