Classifying microwave radiometer observations over The Netherlands into dry, shallow-, and non-shallow precipitation using a random forest model

Linda Bogerd aHydrology and Environmental Hydraulics Group, Wageningen University and Research, Wageningen, The Netherlands
bR&D Observations and Data Technology, Royal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands

Search for other papers by Linda Bogerd in
Current site
Google Scholar
PubMed
Close
,
Chris Kidd cNASA Goddard Space Flight Center, Greenbelt, Maryland

Search for other papers by Chris Kidd in
Current site
Google Scholar
PubMed
Close
,
Christian Kummerow dDepartment of Atmospheric Science, Colorado State University, Fort Collins, Colorado

Search for other papers by Christian Kummerow in
Current site
Google Scholar
PubMed
Close
,
Hidde Leijnse bR&D Observations and Data Technology, Royal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands

Search for other papers by Hidde Leijnse in
Current site
Google Scholar
PubMed
Close
,
Aart Overeem bR&D Observations and Data Technology, Royal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands
eDepartment of Water Management, Faculty of Civil Engineering & Geosciences, Delft University of Technology, Netherlands

Search for other papers by Aart Overeem in
Current site
Google Scholar
PubMed
Close
,
Veljko Petkovic fEarth System Science Interdisciplinary Center, University of Maryland, College Park, College Park, Maryland

Search for other papers by Veljko Petkovic in
Current site
Google Scholar
PubMed
Close
,
Kirien Whan bR&D Observations and Data Technology, Royal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands

Search for other papers by Kirien Whan in
Current site
Google Scholar
PubMed
Close
, and
Remko Uijlenhoet eDepartment of Water Management, Faculty of Civil Engineering & Geosciences, Delft University of Technology, Netherlands

Search for other papers by Remko Uijlenhoet in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Spaceborne microwave radiometers represent an important component of the Global Precipitation Measurement (GPM) mission due to their frequent sampling of rain systems. Microwave radiometers measure microwave radiation (brightness temperatures, Tb), which can be converted into precipitation estimates with appropriate assumptions. However, detecting shallow precipitation systems using space-borne radiometers is challenging, especially over land, as their weak signals are hard to differentiate from those associated with dry conditions. This study uses a random forest model (RF) to classify microwave radiometer observations as dry, shallow, or non-shallow over the Netherlands - a region with varying surface conditions and frequent occurrence of shallow precipitation. The RF is trained on five years of data (2016-2020) and tested with two independent years (2015, 2021). The observations are classified using ground-based weather radar echo top heights. Various RF models are assessed, such as using only GPM’s Microwave Imager (GMI) Tb values as input features or including spatially aligned ERA-5 2-meter temperature and freezing level reanalysis and/or Dual Precipitation Radar (DPR) observations. Independent of the input features, the model performs best in summer and worst in winter. The model classifies observations from high-frequency channels (≥85 GHz) with lower Tb-values as non-shallow, higher values as dry, and those in between as shallow. Misclassified footprints exhibit radiometric characteristics corresponding to their assigned class. Case studies reveal dry observations misclassified as shallow are associated with lower Tb-values, likely resulting from the presence of ice particles in non-precipitating clouds. Shallow footprints misclassified as dry are likely related to the absence of ice particles.

© 2024 American Meteorological Society. This is an Author Accepted Manuscript distributed under the terms of the default AMS reuse license. For information regarding reuse and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Linda Bogerd, linda.bogerd@wur.nl

Abstract

Spaceborne microwave radiometers represent an important component of the Global Precipitation Measurement (GPM) mission due to their frequent sampling of rain systems. Microwave radiometers measure microwave radiation (brightness temperatures, Tb), which can be converted into precipitation estimates with appropriate assumptions. However, detecting shallow precipitation systems using space-borne radiometers is challenging, especially over land, as their weak signals are hard to differentiate from those associated with dry conditions. This study uses a random forest model (RF) to classify microwave radiometer observations as dry, shallow, or non-shallow over the Netherlands - a region with varying surface conditions and frequent occurrence of shallow precipitation. The RF is trained on five years of data (2016-2020) and tested with two independent years (2015, 2021). The observations are classified using ground-based weather radar echo top heights. Various RF models are assessed, such as using only GPM’s Microwave Imager (GMI) Tb values as input features or including spatially aligned ERA-5 2-meter temperature and freezing level reanalysis and/or Dual Precipitation Radar (DPR) observations. Independent of the input features, the model performs best in summer and worst in winter. The model classifies observations from high-frequency channels (≥85 GHz) with lower Tb-values as non-shallow, higher values as dry, and those in between as shallow. Misclassified footprints exhibit radiometric characteristics corresponding to their assigned class. Case studies reveal dry observations misclassified as shallow are associated with lower Tb-values, likely resulting from the presence of ice particles in non-precipitating clouds. Shallow footprints misclassified as dry are likely related to the absence of ice particles.

© 2024 American Meteorological Society. This is an Author Accepted Manuscript distributed under the terms of the default AMS reuse license. For information regarding reuse and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Linda Bogerd, linda.bogerd@wur.nl
Save