The ALCF was developed at the University of Canterbury with support from the New Zealand Deep South National Science Challenge and NIWA, and at Stockholm University with support from the NextGEMS project, for the purpose of evaluation of climate models based on automatic lidar and ceilometer (ALC) observations.
Peter Kuma1, Adrian J. McDonald1, Olaf Morgenstern2, Richard Querel3, Israel Silber4, Connor J. Flynn5
1 University of Canterbury, Christchurch, New Zealand
2 NIWA, Wellington, New Zealand
3 NIWA, Lauder, New Zealand
4 Department of Meteorology and Atmospheric Science, Pennsylvania State University, University Park, PA, USA
5 Pacific Northwest National Laboratory, Richland, WA, USA
Automatic lidars and ceilometers (ALCs) provide valuable information on cloud and aerosols but have not been systematically used in the evaluation of general circulation models (GCMs) and numerical weather prediction (NWP) models. Obstacles associated with the diversity of instruments, a lack of standardisation of data products and open processing tools mean that the value of large ALC networks worldwide is not being realised. We discuss a tool, called the Automatic Lidar and Ceilometer Framework (ALCF), that overcomes these problems and also includes a ground-based lidar simulator, which calculates the radiative transfer of laser radiation and allows one-to-one comparison with models. Our ground-based lidar simulator is based on the Cloud Feedback Model Intercomparison Project (CFMIP) Observation Simulator Package (COSP), which has been extensively used for spaceborne lidar intercomparisons. The ALCF implements all steps needed to transform and calibrate raw ALC data and create simulated attenuated volume backscattering coefficient profiles for one-to-one comparison and complete statistical analysis of clouds. The framework supports multiple common commercial ALCs (Vaisala CL31, CL51, Lufft CHM 15k and Droplet Measurement Technologies MiniMPL), reanalyses (JRA-55, ERA5 and MERRA-2) and models (the Unified Model and AMPS – the Antarctic Mesoscale Prediction System). To demonstrate its capabilities, we present case studies evaluating cloud in the supported reanalyses and models using CL31, CL51, CHM 15k and MiniMPL observations at three sites in New Zealand. We show that the reanalyses and models generally underestimate cloud fraction. If sufficiently high-temporal-resolution model output is available (better than 6-hourly), a direct comparison of individual clouds is also possible. We demonstrate that the ALCF can be used as a generic evaluation tool to examine cloud occurrence and cloud properties in reanalyses, NWP models, and GCMs, potentially utilising the large amounts of ALC data already available. This tool is likely to be particularly useful for the analysis and improvement of low-level cloud simulations which are not well monitored from space. This has previously been identified as a critical deficiency in contemporary models, limiting the accuracy of weather forecasts and future climate projections. While the current focus of the framework is on clouds, support for aerosol in the lidar simulator is planned in the future.