Breakout Summary Report

This session will provide an overview of ARM's radar (scanning cloud and precipitation and vertically profiling) and lidar networks (ceilometer, MPL, Doppler, Raman, and HSRL), emphasizing their science applications and the latest advancements in atmospheric research and their contributions to understanding cloud, precipitation and aerosol properties, wind and turbulence profiling, temperature and humidity profiling, and offshore measurements.

The session will highlight value added products data quality calibration work and include the integration of airborne and space-based sensors, edge computing, agile scanning, machine learning, measurement gaps, and AI-based models to address operational challenges. The session will also address ongoing and upcoming projects giving updates on current and planned instrument deployments. This session will foster a lively exchange of ideas and feedback, ensuring that both radar and lidar systems continue to support groundbreaking science and meet evolving user needs.

The session began with an overview of ARM's radar (scanning cloud and precipitation and vertically profiling) and lidar networks (ceilometer, MPL, Doppler, Raman, and HSRL). This included brief updates on instrument deployments and operations, and examples of how the observations from these systems are contributing to improved understanding of cloud, precipitation, and aerosol properties.

Initially there were several questions about Doppler lidar operations at SGP, specifically about the operation of the two scanning Doppler lidars at E13 and C1, and the high-frequency wind profiling. Some folks expressed interest in the high-frequency data, but pointed out some of the practical concerns, such as: 1) its expensive because it requires two collocated systems, and 2) for systems with mechanical scanners this may increase the failure rate. As a result, it is probably better (and cheaper) to use profiling lidars for high-frequency winds. Jim Mather points out the Doppler lidars have been a huge success since they were first acquired (in 2010), and part of the reason for that success is that we have kept the scanning simple (vertical staring with wind scans done once every 15 minutes).

Suggestions were made how to incorporate high-frequency wind scans at all sites with a single Doppler lidar by interleaving scans. For example, stare vertically for the first 30 minutes of each hour, then do high-frequency wind profiling during the second 30-minute period. We could do that if folks want, but its important to maintain at least a 30min period in either mode for variance/covariance calculations.

Question: Are the Doppler lidar spectral data saved on a routine basis?

Response: Currently, we save the spectra upon request. But we’re looking at developing an on-site processing system that would reduce the data before sending it on to the archive.

Question: Regarding the Raman+HSRL 3 beta 2 alpha aerosol retrievals, have we have considered incorporating sun photometer data and other sources (MPL, Doppler lidar, Ceilometer)?

Response:  Participants highlighted the benefit of enhancing aerosol retrievals using existing additional instruments specifically by integrating MPL data that is available at all ARM sites. Including the ceilometer and Doppler lidars would add the benefit of additional wavelengths. SGP was suggested as a testbed for developing a machining learning training dataset using this integrated, multi-instrument approach.

Question: Can instruments be moved or configured a different way for short duration IOPs?

Response: Instruments can’t generally be moved, but you can request different scan strategies or configuration settings.

Question: Can we make VAPs from ARM publicly available?

Response: Yes, but only to a certain extent as it depends on who wrote the code. Sometimes the code is provided by an external collaborator. Also running the code outside of the ARM system would require modifying the I/O. Damao is agreeable to allowing access to his code for the boundary-layer height VAP. PyArt is open source.

Question: What do we need from ARM radars?

Response: Drizzle rate would be helpful data. One comment was that the ARM Radar documentation is very old - so please update! DSDs are easy to understand to the modeler - so please provide a VAP that provides DSDs. Some sort of Metadata guide that would provide an index or classification of radar scans would be helpful.  Sometimes the data become too much when broken down into multiple files. So more modern data techniques such as Zarr. There is a monthly Radar mentor meeting - to ask any questions. Alyssa makes tools for ARM Radar data. These are available through the DQplotbrowser or ARM data dashboard.

Question: Are there other (radar) bands being used in any field campaigns? Say G-Band.

Response: [Pavlos] Cape-K is a good example of where the technology is mature. You can build multiple band systems. Measurements at Cape-K are nearly as good as disdrometer measurements.  Slam dunk someone says! - To talk to Scott and Die.

Question: What is the scan strategy for SACR and CSAPR?

Response: [Adam] We’re working with the science team. The idea is to sync the scans as much as possible.

Question: Is there any interest in Radar/Lidar simulators?

Response: There are several radar and lidar simulators in the works, and lots of “research grade”  development efforts, but none are being developed within the context of ARM.

Question: Are we missing any key retrievals that can be implemented?  What about integrated datasets with satellite, aircraft and ground-based observations?

Responses: Consider creating data bundles. Consider incorporating NASA’s cloud retrievals. Satellite data should be integrated in near real-time because its harder to get the data if you wait too long (weeks). It would be helpful if the satellite data used ARM time stamps. Someone commented that there is a need for long-term spatial data from ground-based sensors. Education and outreach activities could also be used to streamline or evaluate the VAPs.

Question: How do we optimize radar and LIDAR configuration to maximize scientific impact? Are there synergies between radar and LIDAR?

Response: {Connor] There are lots of opportunity here. For example, Ed Eloranta has the combined lidar and radar retrievals written In Python.

Based on input from the participants, a number of common themes emerged. This included:

• Interest in Radar-Lidar merged products and scan strategies during precipitation events.

• A desire to make ARM Radar/Lidar VAP codes available to ASR members (or similar projects).

• Need to more seamlessly merge the satellite retrieved data alongside observations from ARM sites.  NASA SatCorps team is willing to support.

• Drizzle rate and drop Size distributions from KAZRs or Doppler lidars would be helpful to the community.

N/A

• Improved metadata and documentation for radar scan strategies, including development of a classification system for scan types and clearer, updated guidance in instrument handbooks to support broader community use and interpretation of radar data. 

• Asess feasibility for storing Doppler lidar spectra more routinely to support advanced research applications.  

• Since the current Doppler lidar scan pattern at sites are mostly vertical staring, it would be important to show the community (via conferences/ARM handbook updates) that the modified Doppler lidar scan pattern, i.e., continuous 6-beam VAD retrieved vertical velocity provides similar higher order vertical velocity moments as vertical stare data.  At SGP currently, two lidars are operating side by side, one with vertical only profiles and the other with continuous VAD scans.

• Show the value of continuous profiles of turbulence kinetic energy and momentum flux data from Doppler lidars and their need to validate high-resolution LES models/UAV profiles.

• Create a VAP to package drizzle rate from KAZR and or DL data, if possible.

N/A

N/A