HR: 0800h
AN: A31A-06 [Abstracts]
TI: Impact of Variable SST on Simulated Warm Season Precipitation
AU: Saleeby, S M
EM: smsaleeb@atmos.colostate.edu
AF: Colorado State University, Dept. of Atmospheric Science, Fort Collins, CO 80523, United
States
AU: * Cotton, W R
EM: cotton@atmos.colostate.edu
AF: Colorado State University, Dept. of Atmospheric Science, Fort Collins, CO 80523, United
States
AB:
The Colorado State University - Regional Atmospheric Modeling System (CSU-RAMS) is being used to examine
the variability in monsoon-related warm season precipitation over Mexico and the United States due to variability
in SST. Given recent improvements and increased resolution in satellite derived SSTs it is pertinent to examine
the sensitivity of the RAMS model to the variety of SST data sources that are available. In particular, we are
examining this dependence across continental scales over the full warm season, as well as across the regional
scale centered around the Gulf of California on time scales of individual surge events. In this study we performed
an ensemble of simulations that include the 2002, 2003, and 2004 warm seasons with use of the Climatology,
Reynold's, AVHRR, and MODIS SSTs. From the seasonal 90-day simulations with 30km grid spacing, it was
found that variations in surface latent heat flux are directly linked to differences in SST. Regions with cooler
(warmer) SST have decreased (increased) moisture flux from the ocean which is in proportion to the magnitude
of the SST difference. Over the eastern Pacific, differences in low-level horizontal moisture flux show a general
trend toward reduced fluxes over cooler waters and very little inland impact. Over the Gulf of Mexico, however,
there is substantial variability for each dataset comparison, despite having only limited variability among the SST
data. Causes of this unexpected variability are not straight-forward. Precipitation impacts are greatest near the
southern coast of Mexico and along the Sierra Madres. Precipitation variability over the CONUS is rather chaotic
and is limited to areas impacted by the Gulf of Mexico or monsoon convection. Another unexpected outcome is the
lack of variability in areas near the northern Gulf of California where SST and latent heat flux variability is a
maximum. From the 7-day surge period simulations at 7km grid spacing, we found that SST differences on the
higher resolution nested grid reveal fine scale variability that is otherwise smoothed out or unapparent on the
coarser grid. Unlike the coarse grid, the latent heat flux, temperature, and moisture transport differences on the
fine grid reveal an inland impact. This is likely due to fine scale variability in onshore moisture transport and sea-
breeze circulations which may alter monsoonal convection and precipitation. However, only the largest SST
differences (spatially and in magnitude) tend to invoke large, coherent responses in moisture flux. The SST
variability at high resolution produces relatively large differences in precipitation that are focused along the slopes
of the SMO, with a tendency toward greater variability along the western slope adjacent to the coast. The
precipitation differences are of fine resolution, with variability of +/- 30 mm (over 5 days) along the length of the
SMO. Variability on the fine grid also invokes precipitation changes over AZ/NM that are not resolved on the coarse
grid. Vertical cross-sections examined along the GoC during the surge episode revealed variations in the
moisture and temperature structure of the surge. The cooler SSTs in the climatological dataset produced the
greatest variability compared to the other datasets. The surge produced from climatology SSTs was nearly 5g/kg
drier and up to 4°C cooler compared to surges influenced by the SST datasets. The overall northward
propagation of the surge appeared unaffected by the SSTs.
DE: 3339 Ocean/atmosphere interactions (0312, 4504)
SC: Atmospheric Sciences [A]
MN: 2007 Joint Assembly