Error: Failed to load processor EdGCMToc

No macro or processor named 'EdGCMToc' found

Interpreting Results, Part 1: Analyze Output

The most important aspect of any climate modeling project is the analysis of simulation results. For most global climate models, preparing model output for analysis can be as involved as the initial simulation setup. This preparation of model output for analysis is referred to as “post-processing,” and it involves several steps. Variables of interest must first be extracted from the large binary files that are a GCM’s raw output, and then averaged over meaningful time intervals or geographic areas (e.g., “summer temperature” or “precipitation over land areas”). Finally, the values are scaled to standard meteorological units (e.g., degrees Celsius) and converted to formats that can be read by desktop software programs.

Since post-processing tasks can be so time-consuming, they are typically performed by professional programmers. Thus, while post-processing may be a “resource limitation” for research labs, it is probably more of an “insurmountable obstacle” for middle schools, high schools and many undergraduate institutions faced with the same needs.

In order to tackle this problem, EdGCM automates a number of the most-used post-processing programs, and contains a user-friendly interface – the Analyze Output window – to perform any steps that can’t be entirely hidden. Tasks handled by Analyze Output include:

• The creation of global map files for approximately 80 different climate variables, which can be used to analyze geographic patterns of climate change
• The creation of zonal average and vertical atmospheric slices for approximately 80 climate variables.
• The production of time series plots for approximately 80 climate variables, which can be used to track climate changes that occur during the course of a simulation
• The production of summary tables for all diagnostic variables produced by the global climate model (over 400 variables in all!), averaged over months, seasons, and years for any portion of a simulation, as specified by the user

For each of the tasks available within the Analyze Output window, there are three basic steps to selecting the data to be analyzed and viewed. On the left side of the window, the output years available for a simulation selected from the toolbar are visible in the upper section. Since it is common practice for all tasks except time series analysis to average the last five or ten years of run output, the 5 and 10 buttons beneath the year listing will automatically select those years for processing. (Other year ranges can be selected manually from the year list by holding the shift key while clicking on the first and last years desired.) Year ranges that have already been averaged for analysis are then shown at the bottom left corner of the Analyze Output window.

The center portion of the Analyze Output window displays a list of variables available for analysis, and the time intervals at which the data can be compiled (monthly through annually). Simply check the boxes for those variables you wish to analyze, and then click on the Average button at the base of the window. This action will launch a Fortran program in a new window (Figure 3-14) which displays the averaging function in progress for the range of years selected.

Since it is likely that a user will want to examine only selected climate variables and not the full list available, the Analyze Output toolbar has a special feature that allows the creation of “sets.” When a Variable Set is created only those variables in the set are shown in the interface, which vastly reduces interface complexity when it is not desired. The sets are fully customizable, and multiple sets can be created to emphasize different aspects of a simulation’s results (e.g. Hydrological Variables, Energy Balance Variables, etc.).

Once the Average program has run to completion for a given range of years, the Extract button will become available. Clicking on this button literally extracts only those variables checked in the center list, and generates a new file in netCDF format (a popular file format for spatial data). This file name then appears on the right side of the Analyze Output window, and is ready for the next step in post-processing – visualization. Clicking on the View button in the lower right corner of the Analyze Output window launches that next step.