Climate Models (Week 5) - Post 2

Climate Models

          Climate change is a growing concern for the future of the natural world, and climate models are the reason scientists are so sure of that. Everything scientists know about climate change can be split up into three categories: what is known about the past, what is known about the present, and what is known about the future. Present data is simple, because one must simply walk outside to see the affect climate change has on the present. Past data is slightly more complicated as it is data scientists have either passed down through generations, or preserved data that current scientists uncover in rocks or ice sheets. All types of climate data are important to climate models, however, future data is the most important. Obviously it is not possible to see what affect climate change has on the environment in the future, but it is very possible to predict the future. Climate models are computer generated simulations, based on current knowledge of physics and earth processes, grounded in real world observations, that predict future climate behavior. Scientists have been modeling the climate since the invention of the computer fifty years ago. Each year as technological advances are made and more real world data is collected climate models get more and more accurate. Scientists make climate models to represent Earth's climate system so it is important that the models are grounded in real world observations, of physics, biology, and chemistry. Climate models are can range from the most basic energy balance models (EBMs), to more indepth models called intermediate complexity models (EMICs), to the most complex models named general circulation models (GCMs). Energy balance models estimate the changes in the climate from analysis of the Earth's energy budget. In their simplest form they provide global averages based on different variables like cloud cover or carbon in the atmosphere. They are often referred to as zero dimensional models as they usually do not include any spatial (land/air) dimensions. Spatial dimensions essentially define the space around us like a cube; length by width on the ground, and then height for how far into the atmosphere they go. Intermediate complexity models involve some simplifications like EBMs, however they also include Earth's geography. With an EMIC model a scientist can predict more complex climate behavior, like the behavior of deep ocean currents or mountain ranges. General climate models provide the most precise and complex description of the climate system. They can model large amounts of space with the most specific grid resolution This means that instead of a grid defined by squares the size of entire regions of the earth, they can split their grid into tiny meter-by-meter spaces. With GCMs climate modelers can take more components into account, creating sophisticated models or things like sea ice extent, the carbon cycle, or atmospheric circulation. Each model is made up of thousands of lines of code written in fortran, the scientific computer programming language. The average climate model today has about five hundred thousand lines of code! Each line of code is determined when climate modelers make choices about what variables to include, what to actually model, and what to parametrize; they make these decisions based what questions the modeler wants to answer. Parametrization happens when modelers choose to approximate some processes using reasonable assumptions and relates variables. An example of parametrization is when a modeler decides to define water vapor as a whole unit, instead of writing code for each individual droplet of water. Parametrizations are reasonable and useful approximations. Modelers also make decisions about defining factors in the model like space and time. Modelers decide how much space they will model cover, ranging from a few meters to the whole earth. In addition to the size of area the model will encompass, modelers must also think about time scales. Space and time scales share a direct relationship. The smaller the scale, the shorter the time step. If a scientist is modeling the affect of carbon emissions for an entire city they might use a time scale, or increment, of one year, but if they are modeling the sunlight one small section of ground receives they might use a scale of days or months. Because climate models are made up of code lines programmed into a computer, the models are limited by computer power. Each calculation the model makes must then be multiplied by the  amount of space, and then multiplied again by the time scale; so the time a model takes to output results can increase very quickly. Every decision a climate modeler makes determines the computer power the model will take up. As complexity increases computer power can limit model resolution, model complexity, and length of virtual time that can be modeled. Climate models can be extremely complex and take years to complete, but they can also be very simple containing only a few hundred lines of code. however, even the most basic climate models help scientists understand how the world works and what could happen in the future. Knowing what might happen in the future is very important as it can affect how society chooses to adapt or mitigate to the problem of climate change.
          Scientists use climate models to predict the future of climate change and global warming. Since no one knows the future, scientists cannot test their models' accuracy based on future data; thus they must rely on past and present data to test for accuracy. They use models to test if perturbations by humans have an actual affect on climate by running models with human activity as a factor, and then not as a factor, and then seeing which models results align more closely with actual climate observations. Another way to check the accuracy of a model it to use it to predict past climate and then comparing the results to actual recordings of the past climate. Overall climate models have done very well modeling accurate global temperature recordings. Unfortunately there are still skeptics who believe that climate model data should not be taken so seriously. In response to the skeptics, meteorologist Robert Tuleya said it best "If we had observations of the future, we obviously would trust them more than models, but unfortunately... observations of the future are not available at this time". Whichever way it is put, the information climate models give scientists is a vital resource society must make use of. Nobel prize winner Sherwood Rowland affirmed this importance when he said "What is the use of having developed a science well enough to make predictions if, in the end, all we're willing to do is stand around and wait for them to come true". If society can use these models to adapt its ways and mitigate the affects of climate change, the world just might have a chance of survival.