Spatial Spin-Up

Spatial Spin-Up (sometimes just referred to as “spin-up) is the process by which a model reaches equilibrium. Complicated climate models must reach an equilibrium before they can be used. This means they must maintain a consistent climate on a long-term scale if no additional forcings are applied to the simulated climate system. If a model cannot do this than it is impossible to determine what values predicted from the model are due to the independent variable of the climate experiment and what changes are due to natural variability in the models simulated climate. Model spin-up time can range from instant to tens of thousands of years, depending on what the researcher is attempting to model. For example, if a researcher is trying to model the climate of the from the birth of Jesus Christ (0 AD.) to today (2023 AD.) many more systems in the model would need to reach equilibrium in order to isolate changes over that period. In this example, long term climate systems such as shift in global currents, biome changes, and El Nino would significantly impact the simulated data, whilst if a simulation was being done only over the course of a year these forcings would have relatively little impact. In regional climate modeling, spin-up is much more important since the LBCs is far more likely to produce an unstable simulation or bias results.

Drawbacks & Advantages

In many cases where RCMs are used there is not really an alternative method for performing the same simulation, so in many ways comparing RCMs to GCMs is akin to comparing apples to oranges. However, in the cases when a scientist is deciding between running a GCM to test their hypothesis and an RCM they may consider that an RCM is generally better at predicting extremes in the focused area. GCMs have a tendency to capture general global trends correctly but fail to predict measured precipitation extremes in areas such as the tropics or the poles. This allows them to correctly predict that global temperatures will rise by X°C following a rise in CO2 concentration, but stops them from making useful predictions as to how those higher temperatures will drastically impact quality of life and weather for people living near the equator or poles. In some cases, though, it would not be correct to compare the performance of an RCM with that of a state of the art GCM, especially when the RCM is running with a tenth, hundredth, or thousandth of the computing power. In these cases, RCMs allow researchers to achieve GCM level results with resources that would have never permitted them to run GCMs.