561
Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986
Operational satellite data assessment for drought/disaster
early warning in Africa: Comments on GIS requirements
Hubertus L.Bloemer & Scott E.Needham
Ohio University, Athens, USA
Louis T.Steyaert
NOAA, NESDIS/AISC, Columbia, Mo., USA
ABSTRACT: The National Oceanic and Atmospheric Administration (NOAA/National Environmental Satellite Data and
Information Service (NESDIS) Assessment and Information Service Center (AISC) has developed operational climate
impact assessment for improved drought/disaster early warning in semi-arid regions of Africa. The system is
based on daily United States NOAA polar orbiting, Advanced Very High Resolution Radiometer (AVHRR) satellite
data used in combination with ten-day rainfall reports from ground stations throughout the region.
The current assessments are prepared using a "light table" G.I.S. approach for map/image overlay and
statistical time series analysis. As part of the United States Agency for International Development's (AID)
funded project, NOAA/AISC and Ohio University developed and defined requirements for a cost effective, reliable
computer based G.I.S. In addition to describing the assessment process, computer hardware and software
considerations to meet the needs of the spatial analysts are discussed. Remote sensing data processing and
G.I.S. capabilities are assessed according to various data handling proficiency and applicability of data.
This includes considerations of a variety of computer systems currently available, including "turn-key"
stations with G.I.S. packages as well as a comparison of the obtainable G.I.S. software packages for different
types of data sets.
1 INTRODUCTION
The National Oceanic Atmospheric Administration
(NOAA), National Environmental Satellite Data, and
Information Service (NESDIS) Assessment and
Information Service Center (AISC) has developed a
reliable and cost-effective program to support
disaster early warning and technical assistance
objectives of the Agency for International
Development (AID). The operational Early Warning
Program was developed at the request of the AID
Office of U.S. Foreign Disaster Assistance (OFDA)
and in cooperation with the University of Missouri-
Columbia, US/AID Missions, and host countries.
Qualitative climatic impact assessments routinely
provide early warnings of weather impacts on
subsistence agricultural crops and the potential for
drought caused food shortages with a lead-time of 3-
6 months before socio-economic impacts occur.
Recent advances in the operational use of daily NOAA
polar orbiting meteorological satellite data for
agricultural assessment have significantly upgraded
the existing system which is based primarily on
daily weather station reports. Global
meteorological satellite information increases the
spatial resolution for analysis and presents an
opportunity to incorporate a Geographic Information
Systems approach into the assessments. This paper
comments on some of the GIS requirements for the
upgraded assessment system.
2 BACKGROUND ON EARLY WARNING PROGRAM
Since July 1979, AISC has issued weekly and monthly
assessments of climatic impacts on food security for
developing countries in the Caribbean Basin, Africa,
South and Southeast Asia, and more recently, the
South Pacific Islands, Central America, and the
Andes countries of South America. Decision makers,
planners, and economists are provided with timely,
reliable information based on continuous monitoring
of environmental conditions. The AISC assessment
cables are sent through NOAA international
communication facilities to American Embassies, AIE
including overseas missions, the U.S. Department of
Agriculture, and various United Nations agencies
(e.g., Food and Agriculture Organization). Climatic
impact on potential food supplies, subsistence crop
conditions and field operations are assessed.
Information on crop calendars, estimated soil
moisture, and unusual or severe weather events such
as flooding, high winds and other meteorological
extremes are reported. If available, information
from ancillary sources is included to supplement
AISC analysis. The result is a cost-effective
increase in the lead-time for planning of food
assistance strategies and mitigation measures to
reduce the adverse impact of climate.
In addition, AISC prepares special assessment
reports for AID, e.g.: the 1981/82 drought impact
in Bangladesh, Sri Lanka, Malaysia, Tanzania,
Botswana; the 1981/83 drought problems in Nepal,
Philippines, Ethiopia, Somalia, Uganda, Sudan, Sahel
Region, southern Africa, Haiti, NE Brazil and the
South Pacific; and more recently Kenya, Tanzania and
Sahel countries. These were used by AID as one
input to estimate the magnitude of drought impact,
potential food shortage deficits and/or disaster
assistance needs. The Early Warning Program is
based on weekly rainfall/weather analysis and
climatic impact assessment models for more than 350
agroclimatic regions (i.e., regions which are
generally homogeneous with respect to agricultural
crops and climatic type). Regional rainfall
estimates are determined from ground station reports
received through an international communications
network. Satellite cloud data (e.g., METEOSAT) are
used to improve the accuracy of precipitation
estimates, particularly in those regions where
weather data are sparse and unreliable. Weather
data are then interpreted by regional agroclimatic
indices which indicate potential crop production in
relative terms. Finally, climatic impact and the
potential for abnormal food shortages are identified
from these indices.
Agroclimatic/crop condition indices are based on
derived climatic variables (e.g., soil moisture,
plant water deficit and moisture stress) which
directly determine the plant's response to
environmental conditions, hence productivity. The
selection of the regionally appropriate indices is
in part determined through the use of episodic event
data. For example, candidate indices are
