AMIP-II Experimental Subproject 3.
Acknowledgements: Laura Fowler from the Department of Atmospheric Science at Colorado state University and Peter Gleckler and Mike Wehner from the Program for Climate Model Diagnosis and Intercomparison at the Lawrence Livermore National Laboratory are acknowledged for their help with the sampling experiments.
The recommended sampling strategy has been revised following the results of several studies. See below for details.
Overview
The objective of this subproject is to simulate selected TOVS/HIRS and TOVS/MSU channels using GCM model input with a sampling strategy that approximates idealized (no drift) satellite trajectories. The proposal is for interested modeling groups to run specified routines during AMIP II integrations that generate brightness temperatures of selected channels. These can then be compared to radiance data now being developed under NASA/NOAA TOVS pathfinder archives. The channels proposed are HIRS channels 2, 4, 6, 8 (sensitive to temperature), HIRS channels 10, 11, 12 (sensitive to water vapor), and MSU channels 1, 2, 3, 4 (sensitive to temperature).Objectives
In the line of our AMIP-II experimental subproject "TOVS-HIRS/MSU Model validation", we anticipate three kinds of analyses which will largely be carried out at CSU. We also plan to carry out the analyses with some collaboration with selected modelling groups. For example, we are currently working with both the UKMO and CSIRO groups on topics related to this project. We intend to do the following analysis:- We plan to compare the radiance fields calculated by the models to radiances measured by the TOVS instrument on board of the NOAA satellites. Dr. J. Bates of NOAA-CDC leads a NASA/NOAA TOVS pathfinder activity which compiles these satellite data, carries out satellite-to-satellite checks, and maintains careful calibration. We plan to carry out the comparisons in collaboration with Dr. Bates, who has been involved in the analysis of AMIP-I data. We hope to identify and understand the nature of any differences between the model and observed radiance fields. Feedback from the different modelling groups will be essential for this analysis. Because of the non-linear aspects of the radiative transfer and in order to be able to focus on short-term hydrological processes, it is essential to calculate the radiances on a daily basis. Details of the comparison approach will be provided at a later date on our web page.
- The second analysis seeks to compare observed water vapor fields which are derived from the TOVS radiances (e.g., NVAP, TOVS Pathfinder) with the water vapor fields from the AMIP-II models. The aim of this study is to help identify differences between models and observations and relate these to the radiance differences noted in 1. above.
- The third analysis will be based on developing diagnostic relationships between tropospheric water vapor as deduced from brightness temperatures at the different wavelengths and other parameters of the climate system (such as temperature, precipitation, cloudiness).
Background
This page outlines procedures to simulate selected TOVS/HIRS and TOVS/MSU channels using GCM model input data. The proposal is for interested modeling groups to run these procedures during AMIP II and output the brightness temperatures of the selected channels. These can then be compared to radiance data now being developed under NASA/NOAA TOVS pathfinder archives.Example
Preliminary comparisons between observed TOVS brightness temperatures and temperatures derived from the CSU GCM data are shown in Figure 1. The figure shows a comparison of HIRS channel 11 brightness temperatures for January. January 1989 data were used for TOVS and a non specific January month was used for the GCM. A preliminary comparison for the MSU channels can be found at the MSU page.Proposed Sampling Procedure
Our initial proposed sampling procedure was based on a local time sampling to account for sampling biases when observations of one satellite are used. Because most of the time we have observations from two polar orbiting satellites at the same time, we decided to change this and to ask modeling groups to call the radiation routine 4 times per day at 0, 6, 12, and 18 GMT. This is significant simpler to implement in the climate model runs. More details about the orbits of the NOAA satellites can be found in the NOAA Polar Orbiter Data User's Guide. The proposed procedure and some background experiments performed with the Lawrence Livermore National Laboratory GCM and with the Colorado State University GCM to justify the proposed sampling procedure can be found on separate web pages.
Figure 1: Comparison of observed January
TOVS/HIRS channel 11 brightness temperatures with brightness temperatures
calculated form the CSU GCM.
Richard J. Engelen
Richard.Engelen@atmos.colostate.edu