Teleconnections (i.e., expected effects of El Niño) :

 

What are the scientific views about the existence and the strength of El Niño teleconnections to the Ethiopian region?

 

Interest in El Niño in Ethiopia began following the devastating 1983-84 drought.  In the 1980s, the research capacity at the NMSA on medium and long-range weather forecasting was improving (Kassahun 1999). Forecasters at NMSA were trying to come up with physical explanations for the recurrence of drought in Ethiopia (Kassahun 2000). ENSO was found to be one of the main factors that lead to drought in Ethiopia. Research at the NMSA began to focus on the links between the distribution of rainfall, Sea Surface Temperatures (SST) (including the Pacific, Indian and Atlantic Oceans), and Ethiopian weather (Kassahun 1999).

The NMSA has concluded that increases or decreases in the Sea Surface Temperature (SST) on the Pacific Ocean impact the amount and distribution of rainfall in Ethiopia (Kassahun 1999). Normal Kiremt rainfall is disrupted because El Niño weakens the atmospheric systems that strengthen rainfall and its distribution during the Kiremt season (Kassahun 1999).

 

The first application of El Niño to seasonal forecasting in Ethiopia was in 1987.  In its 1987 seasonal forecast, NMSA incorporated ENSO effects on Ethiopia for the belg and Kiremt seasons (Haile 1987a, 54). Haile stated that the  “occurrences of El Niño/Southern Oscillation (ENSO) phenomena and SST anomalies have a strong impact on the weather and climate variability” of Ethiopia. He said that the impacts of El Niño events on Ethiopian rainfall are real (Haile 1988). There is also an empirical pattern of association between El Niño and droughts in Ethiopia (Wolde-Georgis 1997, 5).

 

According to Ininda and el. al. (1987, 134), “… the years of strong El Niño, which are also years of negative SO anomaly, are generally characterized by below normal rainfall in Ethiopia.” Nicholls (1993, 1) believes that El Niño caused the great Ethiopian drought of 1888. About one third of the population and 90 percent of the cattle died because of the 1888 drought (Webb and Braun 1994, 20). bibliography.

 

There is a negative teleconnection between the floods of the Nile and Atbara rivers that originate in Ethiopia and ENSO Events (Attia and Abulhoda 1992).  Etahir (1996) also concluded that ENSO affects the flow of the Nile River that gets 85 percent of its water from highland Ethiopia; i.e. low flows of the Nile indicate drought in Ethiopia. Etahir recommended that knowledge of such information could help in the predictability of the annual flow of the Nile River. 

 

 The 1997 Kiremt droughts and the 1996 Kiremt floods also occurred during El Niño and La Niña years, respectively. According to Haile (1987a, 58), “the ENSO events have significant impacts on the rain-producing components over Ethiopia and its vicinity. The normalized rainfall analyses in northern Ethiopia clearly indicate that the drought years 1957-58, 1965, 1972-73, and 1984 strongly correspond to the recurrence of the ENSO events.”

 

According to Haile (1987b, 118), “The occurrence of the ENSO events as represented by the Southern Oscillation Index (SOI) … strongly corresponds to the drought incidence (in mid-1960s, early 1970s and early 1980s).” The teleconnection is a result of a complicated feedback mechanism with speculated sequences that cause drought incidences in Ethiopia. These sequences are:

 

Positive SST anomaliesÞ ENSO eventsÞ generation of high frequency of typhoonsÞ displacement and weakening of Tibetan highÞ the weakening of TEJ, vertical shear and easterly disturbanceÞ  drought” (Haile 1987b, 118).

 

Drought occurs in Ethiopia during ENSO years because the thermal gradient is weakened as a result of the SST anomalies over the Gulf of Guinea (Haile 1987b, 118). This weakening, in turn, reduces the warm-moist cross equatorial flow. Thus, the ENSO events (and the SST anomalies) … seem to strongly correspond to the reoccurrence of drought in Ethiopia” (Haile 1987b, 118). In addition to the influence of tropical weather systems, “the interaction between the above tropical systems and the extra tropical weather system makes the major active weather component of the country” (Kassahun 1987, 53). Rainfall patterns in Ethiopia are disturbed when El Niño weakens the high-pressure air mass on the South Indian and South Pacific Oceans that cause rainfall in Ethiopia (Kassahun 1999).

 

Impact of El Niño on the Kiremt Rainfall

 

El Niño affects the Kiremt seasonal weather in Ethiopia by weakening the rain-producing systems in the region and by distorting the amount and distribution of rainfall in Ethiopia (Kassahun 1999).  La Niña has the opposite effect on the Kiremt rainfall, as was seen in 1970, 1971, 1988 and 1996, when there was above-normal Kiremt rainfall in Ethiopia (Kassahun 1999). Regions that are usually affected by El Niño are northern, northeastern and eastern parts of Ethiopia. 

 

There is no single model of correlation between ENSO and the whole country; some areas are more correlated with ENSO than others. However, “All Ethiopian Kiremt rainfall is significantly correlated with May SOI” (Southern Oscillation Index) (Yebeshaw 1996, 5). bibliography. Comparisons of seasonal Kiremt rainfall totals with the corresponding seasonal mean of Southern Oscillation Index (SOI) appears to be deficient during ENSO events (Babu 1999b, 68). Droughts in Ethiopia outside El Niño years are the result of other physical factors that control rainfall.

 

Babu states the following on the relationship between ENSO and Kiremt rainfall:

Since SOI is closely related to ENSO or La Nina episodes, the Kiremt rainfall anomalies (1970-1988) were correlated with the June, July, August and September (JJAS) seasonal mean of SOI. It is found that the correlation is very high (0.8) and indicating a possible link between SOI and Kiremt rainfall. For the southern and northeastern regions the correlation was weak (0.3). It is worth to mention that southern regions do not experience rains during Kiremt seasons. Thus, low negative SOI links to below normal while high positive SOI corresponds to normal to above normal rainfall (Babu 1999b, 68).

 

Table 1. Correlation coefficient between rainfall and SOI for Ethiopia for different zones

MONTHS

RAINFALL REGIMES

 

Country

A

a1

B

C

 

 

 

 

b1

b2

 

JJAS

0.8

0.8

0.3

0.6

0.7

0.3

AM

0.7

0.6

-0.3

0.6

0.5

0.1

MA

0.6

0.4

0.1

0.5

0.3

0.0

MAM

0.5

0.2

-0.1

0.5

0.4

0.1

FMAM

0.4

0.3

-0.1

0.4

0.3

-0.1

FMA

0.2

0.2

-0.1

0.3

0.2

-0.2

Source: Amare Babu 1999b, 69.

 

As Table 1 shows, a fairly high value of 0.7, the SOI has a predictive value because of the correlation between Kiremt rains and the pre-seasonal months of SOI (Babu 1999b, 68). The correlation (-0.2) between Kiremt rainfall and SOI decreases as one moves to the southeastern and northeastern parts of Ethiopia. Thus, “pre-seasonal April and May (AM), March and April (MA) and March, April and May (MAM) months mean could be used to issue seasonal weather outlooks for the Kiremt seasons as one of the predictors” (ibid.). ENSO disturbs the Kiremt rain producing system in the tropics and sub-tropics, while “during La Niña episodes, the major features are well defined, strong and well organized, and a good amount of rainfall is experienced” (ibid.).

 

Impact of ENSO on the belg rains

 

The belg rainfall is generally considered variable in quantity and distribution. However, southwestern Ethiopia and some areas in the central highlands receive between 500-700 mm of rainfall during this season. The rest of the country gets a low amount of rainfall. The relation between Pacific SST and belg rainfall show a negative association with SOI (Babu 1990). The belg rainfall between 1970 and 1988 and its comparison with February to May SOI show that normal to above normal belg rainfall exists during negative SOI (ibid.). During ENSO years (e.g. 1982, 1987, 1992), the belg rains were above normal, while during La Niña years, the belg producing regions of Ethiopia received below normal rainfall, as shown in 1991, 1974, 1999 and 2000 (Kassahun 1999, 2000.). The relationship is created because “high negative or positive index” makes the weather system deviate from the norm by pushing or bringing that precipitation.  

 

The NMSA has classified ENSO types by their timing, and the timing of the significant increase in SST. Zang and Wang (1990) state that,  “In the first group, the anomaly increases considerably in the period from January to June, in the second during July to December.” The NMSA has used this classification to correlate the relationship between El Niño events and 223 rainfall stations for the 1969-1987 period. Based on this analysis, NMSA has reached the following conclusion on the impact of El Niño on Ethiopia (Bekele 1997, 106, Babu 1999b, 70):

 

§         ENSO events are associated with anomalous synoptic patterns during Kiremt and belg seasons.

§         La Niña  (negative SSTA) is strongly associated with deficient belg rainfall

§         El Niño (positive SSTA) is associated with normal or above normal belg rainfall

§         If the increase in the SST occurs between January-June (i.e. group A), it is always associated with severe Ethiopian drought in Kiremt: e.g. the 1972 and 1987 El Niños

§         If the increase in SST occurs between July-December (group B), it seems the El Niño has limited effect on the Kiremt rainfall: e.g. the 1982-83 El Niño.

§         El Niño is not the only cause of drought in Ethiopia.

 

The NMSA scientists state, “So far basic statistical research indicates that ENSO events influence the rainfall variability over Ethiopia. It seems that there is a possibility to predict seasonal rainfall in Ethiopia if the SST and ENSO events could be forecast in good time. In order to carry out these objectives, however, it is essential to acquire improved ENSO forecasts well in advance” (Babu 1999b, 70). 

 

In response to the FEWS report in early October 1997, the Early Warning department of the DPPC issued a special report on El Niño to show the relevance of El Niño events to Ethiopian rainfall. The report was extracted from the NOAA and other Internet sites. It affirmed that there is a historical association between El Niño and drought in Ethiopia.

 

El Niño episodes seem to be associated with drought occurrences in Ethiopia. The National Meteorological Services Agency (NMSA) has recognized the association for several years now and El Niño information is included in making seasonal forecasts. As a result, the quality of the forecasts has greatly improved. (DPPC, October 1997)

 

There is a teleconnection between El Niño events and climate-related hazards in Ethiopia as manifested by the droughts of 1888, 1957, 1965, 1973, 1983-84, 1987, 1993-94. There is a clear association when periods of El Niño from Quinn and Neal (1987) were compared with droughts in Ethiopia, as Table 2 illustrates (Wolde-Georgis 1997).

                                                                             

                  Table 2. Chronology of El Niño and Climate-Related Disasters in Ethiopia

El Niños

Disaster Years

Types

Affected

Killed

Regions

1953

1953

Drought

NA

NA

Tigray, Wollo

1957-58

1957-58

Drought

NA

NA

Tigray, Wollo

 

1965

Drought

  150,000

2000

 

 

1968

Flood

   10,000

    1

 

 

1969

Drought

  170,000

Unknown

 

1972-73

1973-74

Drought

  300,000

100,000

Tigray, Wollo

 

1976

Flood

   50,000

Unknown

 

 

1978-79

Drought

  250,000

  157

 

1982-83

1983-85

Drought

7,750,000

300,000

Ethiopia

1986-87

1987-88

Drought

  330,000

367

Ethiopia

 

1989-90

Drought

2,300,000

Unknown

N. Eth., Harer

 

1990

Flood

  350,000

 

N. Eth., Harer

1991-92

1991-92

Drought

6,160,000

 

 

1993

1993-94

Drought

6,700,000

 

Tigray, Wollo

1996

1996

Flood

 

 

East Ethiopia

1997

 

1997-98

Drought

Flood

 

 

N & E Ethiopia East Ethiopia

Complete models on the impact of El Niño on Ethiopian climate variability are not yet available. The existences of many microclimates due to the mountainous terrain, the lack of meteorological data and trained staff in Ethiopia are most likely some of the reasons for this lag in scientific studies. The NMSA has to develop its studies of the effect of the El Niño event on Ethiopia both by different regions and ecological zones as well as on different scenarios of the emergence of the El Niño event and its monthly effect on various parts of Ethiopia.

 

Climate Related Anomalies Associated with the 1982-83 El Niño.

 

There were many climate-related anomalies associated with the 1982-83 El Niño in Ethiopia. In addition to the misguided policy of the military government, it was the El Niño of 1982-83 that led to the devastating Ethiopian drought/famine of 1984-85. Some argue that the Ethiopian drought of 1983-84 was not as intense as the strength of the El Niño (the record SST at that time) (Ward and Yeshannew 1990). The 1984-85 Kiremt rains in Ethiopia failed because, “In 1984, a warm event of unusual amplitude affected the eastern tropical Atlantic during the relaxation phrase of the 1982-83 Pacific El Niño” (Bekele 1997, 109). Following the 1984 drought, thousands of people and animals died in Ethiopia due to hunger. Many families were displaced due to voluntary and forced migration, especially from Tigray, which was the center of military confrontation against the military government. One notable example was the mass migration of about 200,000 people from northern Ethiopia to the Sudan who were in search of food, at least until the weather improved (Hailu, et. al. 1994).

 

Climate-Related Social and Physical Impacts in 1997-98:

 

The development of El Niño in the Western Pacific and its impact on Ethiopia were clear in mid-1997. The NMSA gave its first forecast on May 29, 1997, and assessed the impact in August. In August 1997, the NMSA announced that El Niño had disrupted rainfall distribution in various parts of Ethiopia during the Kiremt. It stated that rainfall was below normal in many parts of the country, including southern Tigray, Northern Shewa and the SPNNS by 60, 30 and 40-50 percent respectively. The cause was El Niño (Ethiopian Herald 8/30/1997).

 

The El Niño led to climate anomalies on the Kiremt, Bega, and belg seasons of 1997-98 in Ethiopia. The 1997-98 Ethiopian rainfall was characterized as deficient and erratic (Babu 1999b, 69). This led to a decline in food production, the outbreak of disease, such as cholera and cerebral malaria, which killed hundreds of people, and displacement of people due to flooding (Babu 1999b, 70).

It was also reported that the temperature was abnormally high in July and August 1997. It was reported that 18 people were bitten by snakes--8 of the individuals fatally. This happened when the snakes were taking refuge in residential areas due to the scorching sun (Xinua 9/20/1997).

 

The 1997 Erratic Kiremt (June-September)

 

The Kiremt rains usually begin in the middle of June. In 1997, the rains arrived on time; but as the season progressed, precipitation was "generally low and distribution were uneven,” especially in the low altitude areas of the country (DPPC 1997b, 3).   The 1997 Kiremt was characterized by variability of rainfall in amount and distribution. This was particularly severe in the eastern part of the country (NMSA 1997b).  In the first ten days of July, there was a normal commencement of rainfall in the northern and northeastern parts of the country; however, rainfall was reduced in July and August in quantity and distribution (NMSA 1997b, 2). In many other areas, the rains were late by 10 to 15 days, which deviated from the normal and expected pattern. [1] In line with the forecasts of NMSA, the 1997 Kiremt rainfall in Ethiopia was “somehow erratic with some areas of the country receiving poor rains while others – mainly the western highlands – have received relatively good rainfall during the growing season” (UNDP – 9/1997). There were interruptions of rainfall in July and August, undermining the growth and development of crops during these crucial months. There was also “the early withdrawal of the Meher rains, when most crops were at critical stages in their development” (NMSA 1997c, DPPC 1997b, 3).

 

According to the NMSA almost all parts of Ethiopia had dry spells in the Kiremt months of July and August 1997 (NMSA November 1997). Of Ethiopia’s 33 zones, 18 reported that the Kiremt rains were unusually late, which caused further delay in planting and reduced the number of seed options. Land preparation was affected because “farmers sacrificed the quality of plowing for timeliness of sowing” (FAO/WFP 12/19/1997). In addition, the 15 zones that had a good start of the Kiremt rainfall in 1997 were not saved from the anomaly. As the season progressed, rainfall was becoming increasingly erratic “with variable dry spells recorded in August and/or September in most zones” (FAO/WFP 12/19/1997). Although the lowland parts of Ethiopia were more affected by the variability of rainfall than others, even the dega and weina dega agro-climatic zones were not spared from the impact of the low rainfall record. [2] This led to the premature desiccation of crops (FAO/WFP 12/19/1997). The FAO team that visited Ethiopia in 1997-98 reported that total precipitation from the first decade of January 1997 to the last decade day of September 1997 was 15-50 percent below 1996 rainfall [3] (FAO/WFP 12/19/1997).

 

The late arrival of the rains and their interruption during the 1997 Kiremt led to the late planting or replanting of crops, particularly affecting the planting of long maturing crops (UNDP 9/1997). At the end of September 1997, it was obvious that the normal development of crops would need an unseasonable extension of the rains by 2-4 weeks to save a fall in production (UNDP 9/1997).  The wish of many farmers for the unseasonable extension of the rains was realized when the rains were extended until late November, which became too much for the crops to bear normal production.

 

The heavy rainfall slowed the rate of desiccation of the crops. Furthermore, there was germination of wheat and sorghum and an increase in the rate of spoilage of the crops that were harvested. The possibility of fungal attack on standing and stored cereals also increased (FAO/WFP 12/19/1997). The area planted in the 1997 Kiremt season was estimated to have decreased by 9 percent from that of 1996.  Shortage of inputs, such as oxen power and seeds, reduced the size of cultivated land in 1997. Farmers without oxen normally rent from others who have finished planting their plots. Farmers without oxen usually miss ideal planting days. Replanting of plots several times following the arrival of rainfall also depleted the seed reserve of farmers. The 1997-98 FAO/WFP Mission noted that there was a noticeable seed shortage and seed price hikes in the northern zones in October and November 1997. Farmers were opportunistically attempting to cultivate fast maturing crops, such as barley, teff and chickpeas, in order to take advantage of the 1997 unseasonable rains (FAO/WFP 12/19/1997). Chickpeas are usually planted at the end of the rainy season and need less precipitation to mature. Chickpea prices were up by 25-300 percent because of the high demand for seeds (ibid.).

 

Yields and output in 1997 were disappointing because of drought during the Kemt and heavy rainfall during the harvest. Crop output was reduced because of  “delayed and reduced cultivation practices, poor rainfall at flowering, seed and grain fill, and early rain stop” that were reported in all regions of Ethiopia (FAO/WFP 12/19/1997). The 1997 heavy rainfall during the harvest season reduced crop yields. There was “physical damage, increased seed drop, vulnerability to fungal attacks, pre-harvest sprouting, delayed harvesting with associated increases in pest attacks, discoloration of grains, increased vulnerability to spoilage in on-field stacks and untimely threshing and storing of moist grains” (FAO/WFP 12/19/1997).  In 1997, it was reported that yield-reducing factors other than rainfall were very few.

 

One of the impacts of rainfall variability is pest infestation. The alternation between dry and wet conditions is conducive for the reproduction of pests and worms that attack crops. In 1997, there were reports of localized pest infestation, such as armyworms, stalk borer and grasshoppers, that attacked sorghum and maize in the lowland areas of Northern Gondar and Western Harerghe.

 

The 1997 drought had threatened livestock production in the pastoral areas. It is believed that about 10,000 domestic animals might have died in two districts of the Raya in Southern Tigray alone due to drought (UNDP 9/1997). Other localities affected included the lowland Weredas of southern Gondar and the vicinity of the Tekeze Valley. In Oromiya there was stress in the lowlands areas in July and August due to erratic rainfall (UNDP 9/1997). The price of cattle fell by 60-70 percent. There were premature migrations and an increase in morbidity and mortality (FAO/WFP 12/1997).  Fortunately, abnormal late rains reversed the situation for the animals.

 

The 1997 unseasonable bega floods:

 

Unseasonable heavy rainfall started in most parts of Ethiopia in October 1997 and continued until the end of November (FAO/WFP 12/19/1997). These months are normally dry, and Ethiopian farmers harvest their crops during this time of year. The areas most affected by flood were normally marginal in terms of rainfall.  The floods of 1997 were the worst in 40 years (Reuters, 12/5/1997).

 

The abnormally heavy rainfall of 1997 during the harvest season destroyed property and killed many people in Ethiopia. The heavy rains of September 1997 in Tigray caused floods and damaged 1,400 hectares of crops owned by 1,971 peasants (Addis Tribune, 9/5/1997). In Dollo Odo in eastern Ethiopia, 935 families were homeless and 5,700 cattle were lost to flood. In Melka Suftu, on the border of Kenya, floods displaced 2,500 people (Addis Tribune 11/7/1997). Heavy rainfall raised and flooded the banks of the Wabe Shebele and Genale Rivers in the Somali Regional State. The flood killed 38 people, over 9,000 head of livestock, destroyed. In addition, it destroyed 300 houses, displaced 1,300 people homeless and washed out roads  (IPS 11/3/1997, UN-DHA, 11/5/1997, ENA 11/13/1997). Somali refugees in eastern Ethiopia were the groups most severely affected by the flood.  The flood situation was bad in most countries of the Horn of Africa, including Kenya and Somalia. Table 3 is a conservative summary of the impact of abnormal weather in 1997.

 

 

 

Table 3. Summary of the Consequences of Abnormal Weather in 1997

 

Killed

Injured

Damaged Property

Damaged Crop

Flood

171 people

 5620 animals

16 persons

512 houses & residential units

805 hectares

Lightning

26 persons

9 oxen

6 persons

 

 

Torrential rain

18 persons

3258 animals

21 persons

151 houses & residential units

131,915 hectare

Land slide

22 persons

24 persons

 

 

Rain failure

800 animals

 

 

 

Total

237 persons

16,887 animals

67 persons

663 houses & residential units

132,732 hectare

Source: NMSA, Early Warning System Report, November 1997 (unpublished)

 

In October 1997, Ethiopia introduced nation-wide power rationing to offset water shortages in dams with hydroelectric stations (Reuters, 10/3/1997).  

 

The Ethiopian Electric and Power Corporation (ELPA) stated that the effect of El Niño on Ethiopian weather had reduced the volume of water in the reservoirs at the hydroelectric plants. It announced at the end of August 1997, which is the time the reservoirs usually reach their full supply volume, that water volume at Koka was reduced by 40 percent and at Melka-Wakena by 25 percent of their full capacity.  The ELPA announced, in a statement that it had divided the nine Federal States into 11 zones. Each would be without power between 7.00 a.m. and 8.00 p.m. on a rotating basis every 14 days (IPS 8/26/1997).

 

The 1997-98 El Niño affected the production of coffee, Ethiopia’s leading export. During normal weather conditions, Ethiopia’s annual coffee production ranges between 200,000 to 250,000 tons. In the 1997-98 season, however, Ethiopia produced only 128,000 tons (Reuters 10/1/ 1997). This made 1997 production 20 percent less than 1996. In September 1997, Mr. Tsegaye Berhanu, manager of Ethiopia’s Coffee and Tea Authority said that there would be some damage to Ethiopia’s coffee production, but it was too early to assess the damage (Reuters 9/1/1997).  He also added “the freak rains were somewhat light over the coffee growing region during October and September.”  His earlier optimism was dashed when, two months later, he lamented that  “Coffee berries ready to be picked from trees have been falling on the ground due to heavy rains accompanied by gale (force) winds prevailing in the area” (Reuters 12/9/1997).

 

At the beginning of 1998, there were hopes that the abnormally excessive precipitation in the bega season would continue to the belg season. However, the 1998 belg rains were very poor and below average in amount, throughout the southeastern, eastern, central and northern parts of Ethiopia in amount, distribution and duration (DPPC 1998d, 1).

 

 

 

Increase in the Price of Commodities

 

 At the beginning of September 1997, the price of all the major crops was substantially above that of 1996. The price hike ranged from 13 percent for teff to 53 percent for maize (FAO/WFP 12/1997). The anomalous weather in 1997-98 impacted the prices of commodities. Average prices of cereals between August-October 1997 were markedly higher than during the similar months in 1996. There were also regional differences in the prices of cereals. Teff prices ranged from 173 Birr per quintal in Hossana to 273 in Mekelle; the price of wheat ranged between 149 in Bale Robi to 263 in Mekelle; the price of maize ranged from 112 Birr in Chagni to 173 Birr in Alaba; and the price of sorghum ranged from 103 Birr in Chagni to 225 Birr in Mekelle (Grain Marketing Research Project [GMRP] November 1997).

 

Table 4. National Average Cereal Prices August-October 1996 and 1997

(In Birr/Quintal)

 

  1996

1997

% Change

Cereals

Aug

Sept.

Oct.

Aug.

Sept.

Oct.

Oct. 97 / Oct. 96

Mixed teff

177.90

174.79

187.81

186.58

187.84

189.18

+13

White barley

118.55

122.42

120.23

155.90

161.21

189.18

+57

White wheat

143.62

147.97

152.52

181.12

183.57

183.01

+20

White sorghum

129.54

133.99

137.64

159.15

161.74

166.91

+21

White maize

78.19

76.81

79.69

123.30

124.17

111.89

+40

Source: Grain Marketing Research Project (GMRP) New Market Information System (MIS) database, November 1997(Quoted by FAO/WFP 12/19/1997).

 

Impact on Food Production

 

The El Niño of 1997 damaged food production in Ethiopia by a combination of drought during the Kiremt and unseasonably excessive rainfall during the harvest season. As previously stated, total land cultivated, yield amount and total output in 1997 were lower than in 1996.

 

The total area of land cultivated in 1997-98 was estimated to be about 6.8 million hectares. For example, only 29 percent of farmland in the Oromiya zone of region 3 was planted due to drought (Addis Zemen Nehase 1, 1989 E.C [4] or August 10 1997).  An aggregated investigation of the damage indicates that teff was the most damaged crop in 1997-98 (CSA 1998, 22). The Central Statistical Authority (CSA) data indicates that 49 percent of the land for teff was damaged due to too much rain; while 12 percent was damaged due to too little rain. Only 21 percent was reported to be undamaged (ibid.). The yield for this important crop was also reduced by 46 percent due to a shortened period of rain and by 25 percent due to too much rainfall (ibid.). Wheat and barley, which are very important crops on the Ethiopian plateau, were also affected by the unusual weather. Too much rain affected 29 percent and 36 percent of wheat and barley, respectively. This lowered yields by 4 percent for wheat and 1 percent for barley. The undamaged areas for wheat and barley were 44 percent and 39 percent respectively (ibid. 23). The undamaged area had yields higher than 1996/97 by as much as 15 percent for wheat and 13 percent for barley (ibid. 23-24). Maize and sorghum are important crops in the lowland areas of Ethiopia. The CSA data indicates that 18 percent of maize areas and 21 percent of sorghum areas were damaged due to drought, while 22 percent of maize and 12 percent of sorghum were damaged due to too much rainfall (CSA 1998, 25). Too much rain also damaged 30 percent of other crops (ibid.).

 

Table 5.  Damage Assessment for the 1997-98 Season [5]

CAUSE OF DAMAGE

AREA

%

PRODUCTION ‘000 QT

%

YIELD

% CHANGE FROM NO DAMAGE

NO DAMAGE

2335.4

34.1

28954.12

39.33

12.40

-

INSECT AND PESTS

237.15

3.46

2179.21

2.19

9.19

-25.88

SHRTAGE OF RAIN

891.32

13.01

8087.49

10.98

9.07

-26.81

TOO MUCH RAIN

2256.56

32.94

22744.84

30.98

10.08

-18.70

OTHER

1129.07

16.48

1166.05

15.84

10.33

-16.70

TOTAL

6849.50

100.0

73626.71

100.0

10.75

-

SOURCE: Central Statistical Authority, 1998, page 22.

The FAO estimated that Ethiopia’s grain production in 1997-98 from the main season had declined by 26 percent from the previous year (FAO/GIEWS 12/1997) [6] . The FAO said the main reasons for the food production in 1997 were the erratic rainfall during the main season (i.e. June-September), torrential rains during the harvest season, and reduced fertilizer use by farmers (FAO/WFP 12/1997).

 

The 1997-98 climate anomaly had a significant impact on the Ethiopian economy, by reducing Ethiopia’s GDP growth from 5 percent in 1996/97 to 3 percent in 1997-98 (Thompson el. al., 98).  As Table 6 indicates, the output and yield of crops were reduced due to drought and excessive rainfall. Total area for cereals was reduced by 16.25 percent while output was down by more than 24 percent due to erratic weather. Yields were also down by about 10 percent, as Table 6 indicates. The yields of most pulses and other crops also declined due to the 1997-98 El Niño

events.

 

.Table 6. Estimates of 1996/97 and 1997-98 Area, Production and Yield of Major Crops for Private Peasants’ Holdings in Ethiopia (Meher Season)

CROP

Total area in (000 HA)

Total Output (000 QT.)

Yield (QT/HA)

 

96/97

97/98

%

Change

96/97

97/98

%

Change

96/97

97/98

%

Change

Cereal

Teff

Barley

Wheat

Maize

Sorghum

Millet

Oats

6,688.56

2,167.77

697.67

772.23

1,316.87

1,399.95

290.66

5,601.88

1,747.19

681.95

787.72

1,100.61

954.74

289.74

-16.25

-19.40

-2.25

2.01

-16.42

-31.80

-0.32

8,6293.32

20,019.93

7,423.85

10.015.90

25,320.03

20,073.46

2,961.65

6,4987.83

13,073.48

7,863.95

11,067.85

19,288.51

10,697.40

2,587.05

-24.69

-34.69

5.93

10.50

-23.82

-46.71

-12.65

12.90

9.23

10.64

12.97

19.23

14.34

10.19

11.60

7.48

11.53

14.05

17.53

11.20

8.93

-10.07

-18.93

8.38

8.33

-8.86

-21.87

-12.38

Pulses

Horse Beans

Field Peas

Haricotbeans

Chick Peas

Lentils

Vetch

905.35

329.31

158.11

112.81

147.90

52.81

104.41

837.61

266.30

119.88

92.19

169.97

47.09

142.17

-7.48

-19.13

-24.18

-18.28

14.92

-10.83

36.17

8026.28

3,206.76

1,063.03

947.64

1,264.61

344.87

1,199.36

6801.92

2,596.67

927.25

548.46

1,371.33

310.79

1,047.44

-15.25

-19.03

-12.77

-42.12

8.44

-9.88

-12.67

8.86

9.74

6.72

8.40

8.55

6.43

11.49

8.12

9.75

7.73

5.95

8.07

6.60

7.37

-8.35

0.11

15.10

-29.18

-5.64

2.64

-35.88

Others

Neug

Linseed

Rapeseed

Groundnuts

Sunflower

Sesame

Fenugreek

478.45

250.52

148.17

21.40

17.43

5.17

18.50

17.26

410.01

195.22

134.64

15.67

11.02

3.33

23.62

29.71

-14.30

-22.07

-9.13-26.78

-36.78

-35.59

27.68

72.13

2132.79

834.54

676.23

125.28

**

**

72.76

100.31

1836.96

735.79

633.46

86.94

77.76

12.57

98.25

162.19

-13.87

-11.83

-6.32

**

-37.93

**

35.02

61.69

4.46

3.33

4.56

**

7.19

**

3.93

5.80

4.48

3.77

4.79

5.55

7.06

3.77

4.16

5.46

0.45

13.18

3.18

**

-1.86

**

5.84

-5.88

ALL CROPS

8072.36

6849.50

-15.15

96452.39

73,626.71

-23.67

11.95

10.75

-10.05

Source: CSA, 1998

 

Impact of the 1998 belg season:

 

At the beginning of 1998, the prospects for the belg season were expected to be good. The good precipitation at the end of 1997 left some moisture in the soil that led to the early commencement of agricultural activities in some belg-producing areas (DPPC, May/June 1998). There was some rainfall at the beginning of the belg season. However, the rains abruptly stopped in March and the first half of April 1997 (ibid.) A 1998 report by the NMSA stated that the belg rains of 1998 were deficient in the eastern and southeastern parts of Ethiopia. This led to the shortage of fodder and water for the pastoralists who predominantly live in these areas. The rainfall resumed in April and May 1998, but good rainfall that came after April “did not lead to significant improvements in crop conditions in most areas” (DPPC May/June 1998). The dry spells in March and the first week of April 1998 had destroyed the belg crops. The good rains across all the belg growing areas from the second half of April until May were good for the maturation of those grains already “approaching flowering or grain filling stages” (ibid.). The irregularity of the belg rainfall in 1998 “substantially reduced the area planted to the relatively high-yielding but long-cycle crops,” such as maize and sorghum for the for the 1998 Kiremt season (DPPC 1998d, 1).

 

Table 7 below is a comparison between 1996-97 and 1997-98 grain production in various regions of Ethiopia. The FAO Mission put the total grain deficit for Ethiopia in 1998 at 530,000 tons with an affected population of over 5 million (Ibid.)

 

Table 7. Crop Production in Metric Tons by Region for 1996/97 and 1997-98

 

1997-98

1996/97

Tigray

522,881

751,036

Amhara

2,828,827

3,635,923

Oromiya

4,113,637

5,835,035

SNNPS

1,179,860

1,440,661

Benishangul

113,830

96,899

Gambella

9,008

11,130

Somali

17,822

27,300

Harar

12,573

16,627

Diredawa

800

6,153

Addis Ababa

6.055

10,245

Total

8,806,283

11,831,009

Source: Addis Tribune

 

At the end of August 1997, Ethiopia estimated that 154,107 Metric Tons (MT) of food was needed to feed an estimated 3.4-5 million people (Addis Tribune, 8/10/97 and  8/28/97 ).  The main cause of this food deficit was rainfall variability. The reason given for the intensity of the food shortage was a combined failure of the belg and the Kiremt rainfall. A joint USAID and EC report also estimated that in the marginal areas of Ethiopia, [7] which are usually drought prone, not more than 50 percent of the normal output would be produced in the 1997-98 harvest (Addis Tribune, 10/10/1997).

What is the reliability of those attributions?

 

Evaluations of the impact of the El Niño at the end of the 1997-98 season confirmed that the climate anomaly that caused the disasters was real. The NMSA compared the rainfall of the 1997 Kiremt with 1972, which was the analogue year selected during the forecast. It confirmed that the amount of rainfall was similar with some exceptions, as some areas received less rainfall than the 1972 Kiremt. The attributes that can be taken for reduced output are the supply of inputs. The zone governments were providing farmers with seed credits to take advantage of any precipitation. Inputs, such as oxen and seeds, were in short supply because of the abnormal weather that necessitated numerous replanting of the fields because of 1997’s erratic rainfall. There was an absence of subsidized fertilizers, but farmers stated that fertilizers are not useful with unpredictable rainfall.



[1] The rainy month of July that is very important for the sowing of short maturing crops and for the growth of long maturing crops. Based on 37 years of data, November is the driest month in the Ethiopian central plateau with an average of 1.9 mm. (Arrlee, 1965; Quoted by Abate 1984, 21).

[2] Those areas that received much below normal rainfall include “most of Tigray and Afar, Western, Northern and Southern and Northeastern Amhara, western and central parts of Benishangul Gumuz, Eastern, Southern and localized areas of Oromiya, the eastern lowlands of Somali and southwestern and central SNNPR and southern Gambela regional states” (NMSA  November 1997).

[3] In 1997, the FAO stated that there was 20-50 percent less rainfall than in 1996 at 28 weather stations, and a 15% decline compared to long-term average rainfall at 18 stations. The stations were selected throughout the country

[4] E.C., Ethiopian Calendar

[5] The twenty major crops included on the table do not include cash crops, such as coffee and chat fruits, vegetables, enset and other root crops.  (CSA 1998, 22).

[6] There was a reduction in fertilizer use due to the removal of subsidies by the government. However, there are those who argue that fertilizer use during erratic rainfall may in fact make the situation worst. Fertilizer is very responsive to adequate and controlled water supply. It can be argued that the non use of fertilizers by some Ethiopian farmers in 1997 may be a blessing in disguise.

[7] The areas that are usually considered marginal agricultural areas are East and West Harerghe, lowlands of Bale, North Wello, Wag Harma and South East Tigray Zones and the Northern highlands of the Somali Region (Addis Tribune 10/10/1997).