Sea Surface Temperature and Precipitation in the West African Monsoon Climate

Sponsors:National Aeronautics and Space Administration (NASA)NASA Goddard Space Flight Center (GSFC)NASA Goddard Institute for Space Studies (GISS)NASA New York City Research Initiative (NYCRI)

Contributors Dr. Leonard Druyan Dr. Matthew Fulakeza Ruben Worrell Egbuta Oji

Crae Sosa Sarah Dapul-Weberman

Abstract

Regional Model

Monsoons

Future Work

References

Sea Surface Temperatures

Model Correlations

TRMM

Winds

Crae Sosa and Sarah Dapul-Weberman

Due to the variable nature of the monsoon season, it is important to understand the climate of the Sahel region of Africa. Much effort has

been put into creating accurate climate simulations of the region in both the short and long term. The climate model used in this project, the RM3

Regional Model at the Center for Climate Systems Research of Columbia University and the NASA Goddard Institute for Space Studies (GISS), has been configured to simulate weather and climate forecasts over West Africa by taking initial and lateral boundary conditions from

either a global model or reanalysis data. The model simulations are used to investigate the impact of sea surface temperatures (SSTs) in the Atlantic Ocean on the movement of storms and intensity of the Inter-

tropical Convergence Zone (ITCZ). Certain patterns of SSTs favor Sahel drought. In this project, the forcing data and the SSTs used in the RM3 come from Reanalysis 2, which are provided by the National Centers for Environmental Prediction’s (NCEP), National Atmospheric and Oceanic Administration (NOAA). One simulation uses 2006 SSTs, another uses

climatological SSTs, and another tests the impact of cold SST anomalies.

Hovmoller

Model Forcings

Differences

•“Regional Model Nesting within FFS Daily Forecasts Over West Africa.” Druyan, Leonard M., Matthew Fulakeza, Patrick Lonergan and Ruben Worrell, The Open Atmospheric Science Journal, January 2010.•National Atmospheric and Oceanographic Administration (NOAA) - NCEP Reanalysis II •“SST Forcings and Sahel Rainfall Variability in Simulations of Twentieth and Twenty-First Centuries.” M. Biasutti, IM Held, A.H. Sobel, and A. Giannini. 15 July 2008.•“Coupled Model Simulations of the West African Monsoon System: 20th Century Simulations and 21st Century Predictions.” Cook, Kerry H, and Edward K. Vizy. 1 August 2006.•http://www.sciencedaily.com/releases/2004/06/040623082622.htm•http://www.iac.ethz.ch/groups/knutti/research

Correlation Coefficient: 5°N - 15°N

TRMM RM3 Climatol

ogical

RM3 ATL3K

RM3 Control .77 .91 .89RM3 ATL3K .71 .92

 

RM3 Climatologica

l

.72 

The RM3 (Regional Model 3) developed at GISS is a climate

simulation model that uses boundary conditions in order to

predict rainfall, temperatures and circulation over a specified region.

The area is specified by 20°S -35°N and 50°W - 20°E. The model

uses .5 º (~ 50km) spacing between grid points. In order for the model to

run, it needs to be provided with boundary data such as wind speed,

wind direction, humidity, sea surface temperatures, etc. The lateral boundary conditions are

provided by the National Center for Environmental Prediction (NCEP).

The Tropical Rainfall Measuring Mission (TRMM) is a polar

orbiting satellite launched in June of 1997 through a joint collaboration between NASA

and JAXA (the Japanese equivalent of NASA) with the

objective of accurately measuring rainfall over a domain from -50ºS to 50ºN in the tropics and subtropics. TRMM data are recorded in near-real time, but

are updated by calibrating them against gauges.

Sea Surface Temperatures (SSTs) in the Atlantic are highly correlated to the

precipitation that occurs over the Sahel region. Certain patterns of SSTs have

been correlated to many droughts in West Africa. The image on the left shows the skin temperatures (temperatures at sea

level and ground level) from June to September of 2006, as taken from the

National Center for Environmental Prediction (NCEP). Sea Surface

Temperature is one of the many variables that is inputted into the RM3 Model.

Precipitation over the West African climate is highly

dependent on wind patterns. The African

Easterly Jet steers low pressure systems and their

associated precipitation across the continent and

the Atlantic Ocean. Occasionally these systems develop into tropical storms or hurricanes that arrive in

the Americas.

The images above and left show the differences of the RM3 precipitation as

compared to the TRMM satellite data. The TRMM data is subtracted from the RM3 data, which means that where the plotted values

are negative, the RM3 predicts less rain than the TRMM, and where the values are positive, the RM3 predicts more rain than the TRMM. However, some of the areas in which there is

zero difference in rainfall are misleading because no rainfall occurs there. To

accurately test for correlation, the Hovmoller time-longitude distributions are used.

In order to calculate a meaningful correlation coefficient between the RM3 models simulations and the TRMM, the area with the most rainfall is considered. In this case, the data from 10°N -

15 °N are averaged. The data is taken over the 2006 monsoon season (June to September) and the time series for each longitude 30°W to 10°E is graphed. The TRMM shows more extreme

maxima and minima, while the RM3 favors middle values more. The bands of precipitation show the storms moving across the Atlantic over time with significant space-time correlations between

simulated data by the RM3 and observed data by TRMM (see table).

By altering only one of the input variables in the RM3 model forcings, the effect on a dependent variable can be observed. In this experiment the effects of sea surface temperatures (SSTs) on precipitation in the West African region are tested. The RM3 Model is forced using NCEP

Reanalysis 2 data that differs only in SST values. The RM3 Control (left) uses NCEP SST values of what actually occurred in June to September of 2006. The RM3 Climatological uses a climatological average of SSTs from 1998 – 2009, in the months June to September (below, left). The RM3 ATL3K uses SSTs that are 3° lower than the RM3 Control model

(below, right) from 0° to 15°N. As expected, when the graphs are compared, the lower sea surface temperatures in the ATL3k correspond

to lower amounts of precipitation in the model. These model forcings are also compared to TRMM satellite data (far left.)

•Comparisons between different forcings: ATL3K to Control and CLM• Plot winds over contour precipitation to see effect of wind forcings•Studying the relationship between sea surface temperatures and the continuing drought within the Sahel region of Africa

• The effect of GHGs (Green House Gases) emissions + anthropogenic sources on SSTs

• Aerosols + changing SST’s force a differential in the hemispheres above and below equator

The monsoon season in West Africa occurs annually from June to September and is

characterized by winds blowing from the ocean onto the Sahelian belt. The area where North

and Southeast winds merge is called the Intertropical Convergence Zone (ITCZ). It shifts

Northward during the monsoon season, causing precipitation in the Sahelian belt. The

local economy, which includes a large agricultural sector, depends greatly on these

seasonal rains.

The simulation of the U-Winds NCEP data for the 2006 monsoon season (above, left) shows much similarity to the

graph for the RM3 U-Wind simulations for 2006 (above, right).

New York City Research Initiative