Aart Overeem; Ruben Imhoff (2020): Archived 5-min rainfall accumulations from a radar dataset for the Netherlands. Version 1. 4TU.ResearchData. dataset. https://doi.org/10.4121/uuid:05a7abc4-8f74-43f4-b8b1-7ed7f5629a01

Dataset

• Other Environmental Sciences
• Environmental Science and Management
• Atmospheric Sciences
• Atmosphere and Weather

Precipitation, Radar, Rainfall, Remote Sensing

2008/2018

CC BY 4.0

RefWorks, BibTeX, Reference Manager, Endnote, DataCite, NLM, DC, CFF

by Aart Overeem , Ruben Imhoff

Quantitative precipitation estimates from two Dutch C-band weather radars on a 5-min temporal and 1 km2 spatial resolution. The weather radars are operated by the Royal Netherlands Meteorological Institute (KNMI). Before 2016, the two radars were located in De Bilt and Den Helder. In the period September 2016 – January 2017, the operational radars were replaced by two dual-polarized weather radars which are located in Herwijnen and Den Helder. The time in the file name is the time of the end of the observation. Data are available within a radius of 200 km from the radar in De Bilt, which covers the entire land surface of the Netherlands. The data has undergone Doppler filtering and (since 2013) also a cloud-mask from satellite data is employed to remove non-meteorological echoes. From the volumetric reflectivity data, a reflectivity composite is constructed per radar at a constant altitude of 1500 meters. The reflectivities of both radars are then combined with a range-weighted compositing procedure. Finally, rainfall is estimated from these reflectivities with a Z-R relationship of Zh = 200 R1.6 (Marshall et al., 1955). A brief description of the dataset is provided by Overeem et al. (2011), which follows almost the same procedure as in Overeem et al. (2009a, 2009b) to obtain composites rainfall depths, the main difference being the spatial resolution, which was increased from 2.4 to 1 km. Besides, the processing started with the radar reflectivity factor composite from both radars, instead of the composite from each radar separately. Another difference is that no additional correction for occultation (beam blockage due to tall buildings in the vicinity of the radars) has been applied. An advantage is that the data availability of the dataset is very high (higher than the 2.4-km dataset). The adjustment with gauge data, described in Overeem et al. (2009a, 2009b, 2011), has been omitted. As a consequence, on average a large underestimation in radar rainfall estimates is present. It used to be quite common that radar data are not adjusted by gauge data in real-time. Hence, this dataset is representative of a radar dataset which would have been available in real-time. Having such a dataset is relevant to mimic real-time applications, such as nowcasting, for many events (Imhoff et al., 2020). More background information on the KNMI radars can be found in Beekhuis & Holleman (2008) and Beekhuis and Mathijssen (2018). For more details, see the provided publication links below.

• 2020-03-23 first online, published, posted

4TU.Centre for Research Data

media types: application/vnd.google-earth.kml+xml, application/x-hdf5, application/zip, text/plain

• https://doi.org/10.1016/S0065-2687(08)60310-6
• https://doi.org/10.1029/2009WR007869
• https://doi.org/10.1029/2010WR010350
• https://doi.org/10.1175/2009JAMC1954.1
• https://doi.org/10.18174/454537
• https://doi.org/10.1029/2019WR026723
• https://doi.org/10.1029/2020GL089365