Gavin Schmidt Dr. Gavin Schimdt NASA Goddard Institute for Space Studies USA If you would like to interview this scientist, please contact Leah Christen. > biographical information > abstract > paper

Biographical Information:

Gavin Schmidt is a climate modeller at the NASA Goddard Institute for Space Studies in New York and is interested in modeling past, present and future climate. He works on developing and improving coupled climate models and, in particular, is interested in how their results can be compared to paleoclimatic proxy data. He also works on assessing the climate response to multiple forcings, such as solar irradiance, atmospheric chemistry, aerosols, and greenhouse gases.

He received a BA (Hons) in Mathematics from Oxford University, a PhD in Applied Mathematics from University College London and was a NOAA Postdoctoral Fellow in Climate and Global Change Research. He serves on the CLIVAR/PAGES Intersection and the Earth System Modeling Framework Advisory Panels. He was recently cited by Scientific American as one of the 50 Research leaders of 2004.

Abstract:

The Thermohaline Circulation in Past, Present and Future Climate

There is strong evidence that thermohaline circulation changes have influenced climate on decadal-to-millennial timescales in the past. However, direct measures of such variability are sparse, both in the paleo-record and in the modern oceans. Models of this circulation produce qualitatively similar results, but in projections of future climate, the variation among different models is very large. This is partly due to the difficulty in evaluating the models' response to various forcings. I will discuss the utility of using the last clear abrupt climate change in the Greenland ice core record (the 8.2kyr event) as a test case for the models. Results from multi-proxy simulations of the event will be assessed to examine whether this can help reduce uncertainty in the future projections.

Paper:

The thermohaline circulation in past, present and future climate

Gavin A. Schmidt and Allegra N. LeGrande

There is strong evidence that ocean circulation changes in the North Atlantic (the ’thermohaline circulation’ - THC) have influenced climate on timescales of tens to hundreds of years. However, direct measures of such variability are extremely sparse in the present-day ocean. Climate models produce qualitatively similar results, but show very varied results in pro jections of future climate. For instance, as greenhouse gases increase, the models used in the IPCC Third Assessment Report show a range of results from a strong weakening to a mild increase in the thermohaline circulation by 2100. In part, this uncertainty stems from modellers making sure that they have a realisitc and stable North Atlantic circulation, but not being able to evaluate its sensitivity for lack of appropriate data. Over the 20th century, there is simply not enough documented variability to serve a useful role in validating the models. These complex coupled ocean-atmosphere models therefore need to be tested against the larger changes that have occurred in the record of past climate. Fortunately, examples of THC variability abound in the record of past climate, and one example in particular stands out as an excellent case study for the models. This example occurred around 8200 years ago, and hence is referred to as the 8200 yr event. For this relatively short event there are well-dated and widespread data, a background climate similar to today’s, a potentially important ocean response, and crucially, a quantifiable hypothesis for a cause - the catastrophic draining of Lakes Ojibway and Agassiz. These lakes are known to have drained extremely quickly and at around the same time as the 8200 yr event appears in the record, This can be described as a ’Goldilocks’ event for the climate modellers. It is not so small or short term that there is no impact in the record, and it isn’t so large that anything that causes the THC to collapse will do. Instead, it is just about the right size and duration. This event is an abrupt climate change forced (it appears) by an even more abrupt forcing. No thresholds appear to have been crossed. While this event therefore cannot be used to fully address the potential bifurcation structure of the North Atlantic circulation, it may tell us much about the processes that generally keep the THC stable.

Appropriately, these processes will be highly relevant for the climate of the 21st century.
We have modelled this event with a state-of-the-art climate model that is also being used in the latest IPCC assessment. Over areas related to the Lake Agassiz drainage, we put in a range of freshwater amounts to investigate the response of the model to the hypotheised forcing. To properly compare the model output to the climate record, we also incorporate the physics of water isotopes and the impacts on atmospheric chemistry and aerosols directly within the models since these are the processes that are most relevant for actual past climate records.

The response of the model temperature and precipitation are shown in Figure 1.
Basically there is a cooling over the whole Northern Hemisphere (peaking at around 4 degrees in the North Atlantic), and a reduction in rainfall over Eurasia. This is associated with short period of decrease in the THC to between 30 and 60% of its present day strength. In addition, the mutiple modelled proxies are consistent with the observed response.

We will discuss the implications of these results for assessing the projections of future circulation changes.

Figure 1. The annual mean modelled climate response (equivalent to a 40% reduction in the THC) to the 8200 yr event forcing. (a) Surface air temperature, (b) Precipitation.

Gavin A. Schmidt and Allegra N. LeGrande
NASA Goddard Institute for Space Studies and Center for Climate Systems Research, Columbia University, USA