Link: Skip banner
Commodity Intelligence Report
September 19, 2006

Heavy Rains during August Cause Massive Floods in Ethiopia

Ethiopia has appealed for millions of dollars in emergency relief for the country’s worst flooding in decades as hundreds of people have been killed and tens of thousands left homeless caused by flash floods along the Omo, Awash, and Blue Nile rivers.  Floods are relatively common in Ethiopia from June to September when rains peak during the main rainy season, but this year the country has experienced some of the heaviest and most intense rains on record during August.  Forecasters warn that the country faces further flood threats as dams are filled and rains are expected to continue until the end of the wet season in September.

Some of the worst affected regions include the southwestern Southern Omo river valley where 364 people have died since August 13th and the eastern town of Dire Dawa where 256 people were killed since August 6th. Figure 1 below shows two MODIS satellite images over southern Ethiopia and the general region where thousands of villagers are stranded in submerged villages or are living in temporary shelters due to ongoing heavy rains.

Heavy rains continue to hamper frantic relief efforts in southern Ethiopia where water levels in the upstream Gilgel Gibe I dam reached maximum levels, overflowed the spillway, and flooded numerous downstream lowland areas.  Serious flood damage to property and farmland is also reported in other parts of the country, (refer to affected regions map).


Flooded region in southern Ethiopia

Figure 1.  Two MODIS satellite images show numerous flooded regions
on August 21st near the Omo River and Lake Turkana.


Crop Assessment Tour in Ethiopia

Before the August floods started, USDA/FAS/PECAD-Washington performed a mid-season crop assessment and spatial model validation tour for Ethiopia from July 20-28, 2006, and in cooperation with USGS/FEWS-NET-Sioux Falls and FEWS-NET-Ethiopia personnel.  The purpose of the trip was to observe if various spatial models used by USDA and FEWS-NET closely resemble ground conditions and if any model improvements can be easily made to improve their accuracy. 

Over 1800 digital photos of crop conditions and landcover were taken.  These photos were then georeferenced with GPS track data as an easy method to display digital photos in ArcGIS software and a better method to archive year-to-year photos taken in the field (refer to Figues 2, and 4-6).  In addition, over 30 farmers were randomly interviewed during the crop assessment tour.  In general, farmers reported good belg harvests for regions with a belg season, and most belg crops were already harvested with only a few belg crops still drying in the fields in the very high altitude regions, as shown in Figure 2 below.

Photo showing fields with a mixture of different crop stages

Figure 2. Mixture of different crop stages in region
with both meher and belg growing seasons.


Northern and Southern Routes

The crop assessment route from July 20-28, 2006 covered northern and southern routes from Addis Ababa, as shown below in Figure 3.  End of July was the time period selected to conduct the crop tour in efforts to view the corn stage at mid-season.  Evaluating spatial model performance during mid-season is the preferred time period because mid-season crops are close to the critical flowering stage and crops are approaching maximum foliage.  Mid-season crop stage evaluations can also check if the Start of Season (SOS) and Length of Growing Period (LGP) models are performing correctly, by asking farmers and local agricultural experts questions about planting dates and expected harvest time. 

Crop Assessment Route

Figure 3.  Crop Assessment Route from July 20-28, 2006


Elevation is a critical component in monitoring Ethiopian agriculture and agro-climates, with sorghum and corn grown in the lower to mid-altitudes; teff and wheat grown in the higher altitude regions; and barley being grown at the highest altitudes (see Figure 4) of more than 3000 meters (9840 feet).  In general, corn and sorghum were observed or grown at elevations of 2000 meters and below.

Photo showing barley being harvested at elevation above two miles high.

Figure 4. Barley being harvested at elevations above two miles high.


Crops Benefit from Above-Average Rainfall In Ethiopia

For the current meher season, the keremt rains arrived timely and were well-distributed while crop condtions appeared favorable in most areas visited.  Long-cycle corn and sorghum crops were observed to range from vegetative to grain filling stages; with less mature crops (i.e., vegetative or pre-flowering stages) in the higher altitude regions and more mature crops in the lower elevations.  Current meher harvest prospects for corn and sorghum are above-average, assuming normal rainfall is received for the rest of the meher season. 

For most of Ethiopia, short-cycle crops (such as teff, wheat, and barley) are planted during the meher season in June/July and many fields were being plowed and planted for these crops at the end of July (see Figure 5 below).  It is, therefore, too early to make any predictions about the short cycle crops at this time.  However, high local cereal prices, improved seed availability, and favorable belg and meher rains have encouraged farmers to plant this year, which could increase crop area from last year.  Farmers also reported that corn seeds were readily available this year but fertilizers were expensive and limited in supply.

Photo showing a community planting teff.

Figure 5. Women stomp on freshly sown teff seed while men plow the soil with oxen at the end of July. Note that the corn in background (upper right) has tassels.


The most impressive corn belt region observed was located along the Alaba Kulito-Awasa road, where an above-average corn crop is expected, assuming normal rains for the rest of the season (see Figure 6 below).  Most of the corn in this region was planted by oxen, but the large spatial coverage and even plant/row spacings made the fields appear as though they were planted by tractors.  In fact, most of Ethiopia’s cropland is cultivated by oxen and only several small tractors were seen during the entire crop tour.  The total amount of area planted by draught power is impressive and an effective Ethiopian tradition practiced for hundreds of years. 

Photo showing large corn fields planted by oxen.

Figure 6. Typical corn field located along the Alaba Kulito-Awasa road where
the corn was planted by oxen but looks as though its was planted by tractors.


Large Government Discrepancies between Crop Area/Production Estimates

In a meeting with the USAID mission based in Addis Ababa, USAID personnel mentioned problems with Rainfall Estimates (RFE) for Ethiopia’s complex topography and discrepancies between crop estimate reports from two different Ethiopia government agencies.  The crop estimate discrepancies originate from the following two agencies that utilize different methods in compiling their crop area/production estimates:   

  1. Central Statistics Authority (CSA) provide comprehensive statistical data on agriculture through the organization and implementation of sample surveys
  2. Development agents of the Bureau of Agriculture and Rural Development (BoARD) of the Ministry of Agriculture and Rural Development (MoARD) collect data at grassroots level during plowing, sowing and harvesting time from the whole farming community. These BoARD data are processed through a hierarchical series of steps via Woreda, Zone and Region administration units to national level for use by agricultural specialists in their day-to-day activities.

The different methods used by these agencies are common methodologies used in both developed and developing countries, but the Ethiopian government does not have a formal protocol or committee mandated to reconcile different crop estimates at the end of each crop season.

The FAO/WFP Crop and Food Supply Assessment Mission (CFSAM) reports for Ethiopia have recorded discrepancies between CSA and BoARD since 1994, and each year FAO/WFP explains how they attempt to reconcile these difference crop estimates.  In general, CFSAM reports utilize the BoARD data set who is the established government partner in the CFSAM surveys.  In addition, CSA data traditionally are not available during mid-season or during harvest time when the CFSAM team conducts their crop production surveys and national food balance audits.  In contrast, crop production data from BoARD are presented to the CFSAM missions each year, and are discussed and audited during the CFSAM annual field visits during December.

Further analysis of the crop area discrepancies between CSA and BoARD during the 2005 crop season are shown below in Figure 7, where data from CSA and BoARD were taken from FAO/WFP CFSAM report of February 24, 2006.  Note that the largest crop area discrepancies lie in Zones located adjacent or fairly close to Addis Ababa.   Also, no discrepancies are reported in the Amhara Region, where nearly one-third of cereal production occurs, because the BoARD accepts all CSA crop estimates for this region.

Location of Area Estimate Discrepencies

 Figure 7.  Location of the Largest Crop Area Estimate Discrepancies between
CSA and BoARD data during the 2005 Crop Season

USDA’s official crop production estimates for Ethiopia (see PSD Online) have largely followed CFSAM reports for the past 10-years, as USDA also audits crop production estimates presented by national governments.  In addition, USDA’s Interagency Crop Estimates Committee (ICEC) has noted for the past several years that the greatest crop production discrepancies between CSA and BoARD tend to occur for wheat, corn, and sorghum commodities.  For example, the 2005 difference in corn, wheat and sorghum area estimates between CSA and BoARD are 600,000, 500,000, and 400,000 hectares, respectively, for a total difference in area of 1.5 million hectares.

The FEWS-NET office based in Ethioipia reported that FAO plans to implement a two-year project with CSA, BoARD, and Global Monitoring for Food Security (GMFS) in an attempt to reconcile government agency differences in crop area estimates.  GMFS will use radar satellite data to estimate crop area, but using satellite data to estimate crop area will be technically challenging in Ethiopia due to by small and fragmented fields, staggered planting dates, and similar leaf properties that cause crop classification confusion between sorghum and corn; wheat, barley, and teff; and most major pulses.

Part of the above project has motivated the Ethiopian government to formulate a crop estimates committee that might in the future be mandated to reconcile different crop production estimates from different government agencies after each crop season. Such a formal crop estimates committee should help to reconcile different crop area/production estimates after each season by consensus.  In addition, other related government agencies such as Ethiopian Grain Trade Enterprise and Disaster Prevention and Preparedness Agency (DPPA) should perhaps participate as members on this crop estimates committee to help reconcile conflicting crop estimates every season.

Current USDA area and production estimates for grains and other agricultural commodities are available at PSD Online.


Other Related Links and Images

Near-real Time MODIS Images from
PECAD's Global Agricultural Monitoring (GLAM) project

Recent (August 2006) Flood Damage in Ethiopia

Major Rivers which tend to flood in Ethiopia

Agroclimatic Map for Ethiopia

Crop Calendar for Ethiopia



For more information contact Curt Reynolds | | (202) 690-0134

Close Window