Keah Schuenemann Research Page

Send Keah an e-mail (schuenem@colorado.edu)

CIRES: Cooperative Institute for Research in Environmental Sciences

ATOC: Department of Atmospheric and Oceanic Science

Cassano Polar Climate and Meteorology Research Group

AGU Poster 2007 (PDF)

Using Self-Organizing Maps (SOMs) to Study

Precipitation and Precipitation Trends over Greenland

Studying the synoptic climatology and precipitation patterns over the North Atlantic region is necessary in order to better understand the atmospheric input to the mass balance of the Greenland ice sheet. The self-organizing map (SOM) technique is applied to ERA-40 average daily sea-level pressure (SLP) data from 1961 to 1999 to objectively identify synoptic SLP patterns over the North Atlantic region. A total of 35 different SLP patterns were identified.

(Below) Master SOM. Each of the 35 patterns in the master SOM is referred to as a node. The figure is plotted with sea level pressure anomalies (hPa) contoured every 2 hPa. Blue shades represent low sea level pressures and red shades represent high sea level pressures over the domain.

The list of days whose sea level pressure pattern fell into each node was used to create a list of all of the precipitation data that fell into each node. These ERA-40 precipitation data were averaged to create a precipitation pattern mapped to a certain node on the master SOM. For example, the precipitation maps for the 692 days that fell into node (0,0) were averaged to create the precipitation expected to happen whenever the synoptic pattern in node (0,0) takes place.

(Below) Node averaged precipitation anomaly (cm/day) (shades) and node SLP anomaly (hPa) (solid contour lines). Blue shading indicates positive precipitation anomalies and red shading indicates negative precipitation anomalies.

Analysis of precipitation patterns associated with each SLP pattern revealed the various forcings for precipitation over portions of Greenland. In most cases the largest precipitation events are associated with passing cyclones, which create onshore flow, allowing for the air to be lifted orographically by the ice sheet. The southern portion of Greenland received the largest magnitudes of precipitation due to its extreme topography and its location near open water. Each portion of Greenland has its own preferred synoptic pattern for receiving precipitation.

This analysis will be repeated for 15 climate model predictions for the reanalysis period and 2046 – 2065 and 2081 – 2100 in order to evaluate the ability of the various climate models to reproduce the ERA-40 synoptic climatology of the North Atlantic region. The models that perform the best over the reanalysis period will be used to study predicted future trends in synoptic patterns and precipitation over Greenland from 2046 – 2065 and 2081 – 2100. Will certain weather patterns become more common in a warmer atmosphere? Will some patterns drop more precipitation over Greenland in the future and make up for mass loss from melt on the Greenland ice sheet?