ROCHESTER, NY (WROC) – Scientists are learning more about the relationship between carbon and temperatures going back millions of years using proxy’s and mathematics. Below is an interview with Tony Wong, assistant professor at RIT’s school of mathematical sciences.
James Gilbert is in bold text, Dr. Tony Wong is in standard text.
I KNOW WE AS HUMANS HAVE BEEN ABLE TO DIRECTLY MEASURE CARBON DIOXIDE FOR SEVERAL DECADES, AND WE CAN USE OTHER METHODS LIKE TREE RINGS AND ICE CORES TO GO BACK SEVERAL THOUSAND YEARS, BUT YOU’RE TRYING TO USE MATHEMATICS TO GO DEEPER. CAN YOU GO DEEPER INTO HOW YOU ARE DOING THIS AND WHAT THE IMPORTANCE IS?
Yes, Tree rings and ice cores are a really great tool to go back into the thousands of years range, we’re going even further back to millions of years ago, all the way to about 500 or so million years in the past, and we are using proxy’s, that’s what these tree rings and ice cores are for temperatures and CO2.
Some of the proxies that we use are things like the densities of stomata, these little pores in fossilized leaves so these ancient leaves can give us some insight on other proxies like cave speleothems, water dripping in caves, building up. You can use ancient record from there to get a sense of what the co2 and temperature would look like millions of years in the past. We’re using this modeling approach where we take our data sets for temperatures and CO2, and we’re looking at what kinds of the input parameters These parameters represent our knowledge of these geophysical processes and out of the model comes C02 and temperature estimates, and what we do is we’re looking at what value of those input parameters represent the observed through the proxy observations perceive C02 and temperature well so we can rule out any kind of parameters that don’t match those proxy records.
SO YOU RUN A MODEL AND YOU CAN JUST A MODEL ON HOW IT PERFORMS BASED ON THE PROXY AND RUN THAT MODEL BACK FURTHER THAN THE PROXY, AND ASSUME THAT IT’S GENERALLY CORRECT?
TONY: We go back to where the proxy records generally begin, that’s about as far back as we can go, about half a billion years. We have CO2 proxy data getting back to about 420 million years ago. We run the model from about that period to present.
DOES IT FILL IN THE GAPS? WHAT IS THE GOAL?
TONY: One of the goals is to get some insight into what our current understanding of the earth’s system is and how we might be able to improve that. One of the processes that’s baked into this model is the way that different types of rocks are weathered, so over the course of millions of years, you can imagine wind blowing over a rock and just gradually wearing away at it, and then these sediments coming off of these rocks. In cemeteries you can see old limestone starting to be weathered by some of these processes. All that sediment coming off these rocks is making its way into rivers, those rivers are making its way onto oceans through ocean circulation and tectonic plate movement, you might see these sediments sinking their way down to the bottom of the ocean, subducting as these continental plates are moving.
This is happening over millions of years, and eventually making its way down to the mantle, where it gets mixed up and it gets either vented out as CO2 into the atmosphere or it could also be vented out using volcanic activity.
SO YOUR TEAM MAINLY TRACKS CARBON DIOXIDE AND TEMERATURE?
There is also O2 being tracked in this model.
ARE THESE MODELS RUN OUT OF RIT?
There’s a balance with models. Especially with weather prediction are extreme4ly computationally intensive and they’re running over millions of individual CPUs using lots and lots of computer power, and a lot of that’s for resolving things down to city size weather events. For our mode3l we’re looking for global scale reservoirs, in bulk how much CO2’s in the atmosphere.
We’re not resolving cloud processes, individual tectonic plates, we’re just looking at the global average over the course of millions of years. There’s a tradeoff between the resolution of individual processes versus doing detailed statistics and uncertainty quantification, which we were focusing on in this paper.
WHAT ARE SOME OF YOUR INITIAL TAKEAWAYS FROM SOME OF THE EARLY FINDINGS?
One of the key takeaways was this bias in some of the previous work using this model during the cretaceous period, somewhere around 90 million years ago, the model seemed to be a little bit cool relative to proxy observations for temperature.
What we were able to do is run a really large set of simulations. Thousands of model simulations and figure out based on varying the input parameters, each of which represents a set of different geophysical processes, in what way did we have to modify those parameters in order to actually match the temperatures that we’ve observed through the proxy record over the past few hundred million years.
That highlights a place where our understanding of how temperatures evolved and what processes contributed to that temperate where we might be able to improve that understanding and where there might be gaps, where our record for say things like river discharge into the oceans, maybe we need to do more research in that area that can lead to improvements in how our model can match prehistoric temperatures.
IT SEEMS TEMPERATURES AND CARBON DIOXIDE ARE BOTH INTERRELATED, WITH HIGHER TEMPERATURES AS GREENHOUSE GASES GO UP. IS THAT SOMETHING YOU GUYS SEE AS WELL?
The model does not assume there is a 1 to 1 relationship. There are a couple of other factors that relate to the model temperature besides CO2, so changes in solar luminosity for example, or changes in the continental configuration of total land surface area will also contribute to model temperatures, so it’s not necessarily a 1 to 1 relationship, but there is definitely a relationship that as CO2 goes up, in general, you see a rise in global surface temperature.
Tony goes on to say this model focuses on larger timescales, thousands of years, so the anthropogenic climate change that has happened over the last 100 + years is not well tracked by this model.
IS THERE ANYTHING ELSE YOU WOULD WANT PEOPLE TO KNOW?
It’s a great approach because we’re looking at trying to get a better understanding of how the earth system works and by doing this kind of a model experiment, we are able to rule out, purely based on whether or not your model matches the data that you have available, we’re able to rule out completely implausible sets of model parameters, or ways that it’s plausible that the earth’s system maybe worked millions of years ago, but then when you go back and compare to data it just doesn’t match the data, so you get this inverse approach, where you’re working backwards to get a better understanding of these fundamental processes.
That’s where this starts to link in to the more anthropogenic climate change side of things. We just want to understand better how the earth’s system works, and this gives us an opportunity to use information from the distant past to get insight into that. When we’re thinking about 400 ppm CO2, when we’re thinking about just how quickly temperatures has increased in the past few decades, that’s something we have to go back thousands of years or millions of years to really get a good handle on, back to ice ages, back to these large swings from greenhouse states on earth to icing states on earth, and so that gives us a better analog into what we’re starting to see today in terms of really large swings in earth’s climate over fairly short time periods.