Identifying Cloud Types for More Effective Use of ARM Observations in Model Validation and Statistical Analysis

ARM has a wealth of atmospheric observations, with a particular strength being the existence of simultaneous measurements of many relevant quantities. However, to fully take advantage of this wealth of measurements, some context is needed to interpret these measurements statistically. Here we present new data products that can aid in that interpretation. A cloud type data product has been produced from cloud boundaries determined by vertically pointing radar and lidars. Results will be shown from a long-term climatology showing seasonal and diurnal variations. This product can be used as an index to identify how other measurements of the atmospheric state (such as radiation, moisture, aerosol, etc) vary with cloud state. This allows for statistical aggregation of observational data in order to better understand atmospheric processes and relationships. Using indices of atmospheric state also facilitates more useful comparisons with model output in order to gain understanding of the conditions when models do well or need improvement. Second, a fair weather shallow cumulus identification data product has been produced to automatically determine time periods of interest to high-resolution modeling. This product merges information from multiple instruments, and validates the results against manually selected shallow cumulus time periods. This shallow cumulus product is one example of how multiple ARM data sets can be used to give more targeted classification of observations for specific science questions.