Fast Gaussian Process Emulation of Mars Global Climate Model

The NASA Ames Mars Global Climate Model (MGCM) software has been in steady use at NASA for decades and was recently released to the public. This model simulates the complex interactions of various weather cycles that exist on Mars, namely the Dust Cycle, the CO2 Cycle, and the Water cycle. Utilized by NASA, the MGCM is used to help understand their empirically observed data through the use of sensitivity studies. However, these sensitivity studies are computationally taxing, requiring weeks to run. To address this issue, we have trained a surrogate model using Gaussian processes (GP) that can emulate the output of this model with relatively small amounts of data in a reduced amount of time (on the order of minutes). We demonstrate the effectiveness of our emulator using backward error. The NASA Ames Mars Global Climate Model (MGCM) is a computer model that simulates the weather on Mars. The MGCM is used by NASA to help understand weather data collected from satellites and other sources. The model has many inputs, the correct values of which are unknown, so it is often run many times with varying input values. Each run of the MGCM takes a long time, so running enough times to gather all of the desired outputs can take weeks. To address this issue, we have developed a method to approximate the output of the MGCM through the use of a machine learning model. This model requires a relatively small amount of data to train and once trained can approximate the MGCM output accurately and very quickly. Gaussian Process emulation is effective at vastly decreasing the required runtime for sensitivity studies using the Mars Global Climate Model (MGCM)The use of Gaussian Process emulation is remarkably accurate for the MGCMGaussian Process emulation has significant potential to support more refined studies that would otherwise be computationally unfeasible

Automated summary

The NASA Ames Mars Global Climate Model (MGCM) is a computer model that simulates the weather on Mars. The MGCM is used by NASA to help understand weather data collected from satellites and other sources. To address the computational challenge of sensitivity studies, a surrogate model using Gaussian processes (GP) has been developed. This surrogate model can accurately and quickly approximate the output of the MGCM with a relatively small amount of training data.