Forecasting by Analogy :
If a perfect forecast
had been available as early as October 1996 (knowing what is now known about
the actual impact), what could have been done differently? (Do not take
into consideration at this time any restrictions on possible actions.)
a.
About information flow?
Societies have
“the propensity to prepare for the last climate anomaly by which they are
affected” (Glantz 1988, 4). Current Ethiopian policy makers have constantly
engaged in forecasting by analogy in the area of climate-related impacts to
avoid repeating the 1973-74, 1980, 1984-85, 1987 and 1993-94 experiences that
reduce food in varying degrees. However, they do it with imperfect information.
One of the
problems with weather forecasting in Ethiopia is the lack of timely and
reliable information. As mentioned earlier, traditional meteorological
information on the various Ethiopian climatic zones is not available. If there
were perfect information available about the arrival of El Niño as early
October 1996, there would have been more time between the dispatch of
information to the users and the actual impact. In the case of Ethiopia, the
season that was impacted was the Meher season when plantation of long maturing
crops begins as early as April and May. With El Niño information that had about
nine months of lead time, the Ethiopian Early Warning System could have had the
time to spread the word to the regions and the villages.
The lack of
perfect information forces the NMSA to delay its forecast. In 1997, for
example, it heard about the El Niño in March, but issued the weather forecast
at the end of May 1997. We can even argue that the May forecast does not have
much utility for those farmers who plan to plant long maturing crops. Mr.
Ibrahim Mohammed, Head of the Extension Department of the Ministry of
Agriculture, stated that “the forecast usually comes at the beginning of the
sowing season with no time gap to prepare any counter measures to counter the
drought.” The NMSA sends its forecast at the end of May after the drought has
already affected the first or second stage of the planting process.
[1] The late arrival of early warning
information was also one of the main complaints of the Early Warning Officers
in Tigray Region. Thus, there is very little time left for the National Early
Warning System to meet, study how to respond, and reach the farmers and
herders, who are the main users in Ethiopia.
If we examine
the time frame between the 1997 El Niño-based forecast in Ethiopia and the flow
of information based on our information from the Tigray Regional State, we find
that more than a month had elapsed before the Wereda level functionaries
held their first meeting. The NMSA forecast was released on May 29, 1997 and,
they received the NMSA forecast at the end of June. On July 8 1997 they
gathered the entire zone Disaster Prevention and Preparedness Department heads
of the region for a briefing. The latter returned to their zones and briefed
the Wereda Early Warning Committee members that have direct contact with
farmers and herders. Thus, it took a minimum of approximately five weeks for
the information to flow between the NMSA forecast and the briefing of the Zone
Early Warning Committee by the Regions. In Tigray, the Regional Early Warning
Committees and the Zone EWC met every two weeks to report about events and to
evaluate the response until the impact of the drought was over. These
activities were emergency in nature due to the late arrival of the forecast
from the NMSA. Perfect information in October 1996 could have given much needed
time to study the problem and choose the best alternative responses. The
information could have reached the farmers long before the beginning of the
planting season so that they could decide what alternate actions to take.
Perfect information could have also have also created credibility between the
users of information, such as the farmers and the distributors of information, such
as the local development agent. The development agent meets with the farmers
even when the information he delivers is wrong.
b.
About preparing for the forecast impacts?
Lack of access
by Ethiopians to water, infrastructure, markets, information and inputs are the
most important constraints that are exacerbated during drought. With perfect
forecast information many months ahead of time, the society could focus all of
its energy and deal with the above problems. It could dig or construct dams in
all parts of the country to prepare for the impact. Drought resistant seeds
could be identified and collected for later distribution for farmers. There
would also be time to pass the information on how to cope with the impact.
Farmers would also get advice on where and when to plant their crops. This
would reduce their lost energy and seeds that are used for planting and
replanting. One of the pieces of advice from the Ministry of Agriculture to
farmers in 1997 was to replant crops. This is hardly any support as farmers
have been doing this all the time.
As stated
earlier, the Ethiopian government advised farmers to adapt to the nature of the
weather and reduce losses by planting drought-adaptive crops. It also mobilized
and distributed resources. However, time is a very important factor for the
successful implementation of both. A forecast with a lead time of more than six
months would have given the farmer and the government ample time to communicate
and implement their responses. The responses would not have the tone of an
emergency. As stated by Mr. Ibrahim of the Ministry of Agriculture, Ethiopian
forecasts usually come after the beginning of the planting season even for the
fast maturing crops. Table 9 shows the planting and harvesting time of some of
the crops in the Tigray region of Ethiopia at the beginning of the Kiremt
season. Table 9 does not include the long maturing crops, such as maize,
sorghum, sigem hagai, taff Ginbot (May teff) etc.
Table 9. Planting and Harvesting Dates
|
Crop variety |
Plant |
Harvest |
|
Barley Tsa'idaa
sigem Sa-isa'a |
June 25-July
2 July 8-July 15 |
Sept 11-Oct
10 Sept 11-Oct 10 |
|
Wheat Ashihaan
Desaalegn
Tselim
sindaay Ayiqurtem
|
June 8-June
15 July 8-July
22 July
22-Aug 6 Aug 7-Sept 6 |
Sept 11-Oct
10 Oct 11-Nov
9 Nov 10-Dec
9 Dec 2-Dec 9 |
|
Oats 'Ares
|
Aug 7-Aug 15
|
Nov 10-Dec 9 |
|
Flax Intati
|
June 8-Aug 6
|
90 days later |
|
Legumes Inguayaa |
Sept 1-Sept 7
|
Nov 1-Nov
10 |
|
Chick pea Shimbraa
|
Sept 6-Sept 10 |
Nov 10-Dec
9 |
Source: Bauer (1976, 71).
Mr.
Ibrahim stated that if “we were to get the El Niño or La Niña information
a few months ahead of the beginning of the farming process for that season,
we could cope and reduce the drought impact.” He also added “anything
you do to prepare for drought does not hurt.” Some of the actions that can
be taken to reduce the impact of drought are creating dams, saving seeds,
and buying food. These are actions that are helpful to the society and the
country even under normal weather conditions. The worst thing is “to sleep
idle saying that there is no drought.” Mr. Ibrahim stressed that Ethiopia
needs to stop “working in the form of emergency or relief and deal with the
drought problem in a planned manner.”
Therefore, a
longer time frame between forecast and action introduces the idea of planning.
With time, people could be encouraged to introduce the diversification of
crops. He wants to “shift from annual cereals to other crops, such as
vegetables,” which are dismally absent in Ethiopia. The opportunities created
by the drought should be used to change the culture of the people in order
introduce the gardening culture into their activities. Mr. Ibrahim cited the
case of the bumper potato harvest in Debre-Berhan, which was not subsequently
followed by the farmers because it is not a culture that is interested in
eating potatoes.
The Ministry of
Agriculture official stated that “access to perfect information with many
months of lead time would help us to advise our nomads to convert their cattle
into other forms of property, instead of waiting until the drought comes and
forces them to sell their livestock cheaply.” The extra lead-time could have
given the herders time to sell their animals.
One of the problems of imperfect information is the tendency for trial and
error. If there were perfect information, the Head of the Tigray Regional State
Early Warning Department, Mr. Hadera said, “The region could have done more
than what it did in 1997.” The canals and water harvesting holes could have
been built extensively, and drought resistant seeds could have been selected in
advance. He said that it would have also been possible to sell animals before
they starved to death.
A perfect
forecast for the 1997-98 El Niño, as early as October 1996, could have been an
incentive for the government and farmers to spend more resources on the
post-harvest lose [losses] (?) of Ethiopian agriculture in 1996, which was a
record production-year. In addition, Ethiopia would not have exported some of
its 1996-97 surplus food production to the neighboring countries. It would have
stockpiled it for the period of food deficit in 1997 and 1998. The 170,000 MT
food exports were going on in March 1997, two months earlier than the NMSA
forecast of May 29, 1997 (Addis Zemen July 4, 1997 or Sene 24,
1991 EC). In March 1997, there were two contradictory news items in Addis
Zemen. One referred to food exports and the other to the possibility of
food insecurity in the country. Additional El Nino information about the
possibility of the failure of the Kiremt rainfall would have lent
additional information to policy-makers that the food insecurity that might be
witnessed would reach disaster proportions. Donors were also depleting the
successful Ethiopian Emergency Food Security Reserve in Ethiopia because of
loans, for their programs in the neighboring countries as well as Ethiopia, since
the beginning of 1997. The Food Security Reserves should not have been opened
until the impact of the 1997-98 El Niño was clear. When it was a necessity to
get food later in 1997-98, it was depleted and there was very little reserve
food for emergency. Donors could have been very careful before they borrowed
grains from the Food Reserves. It could have also been possible to buy more
food from the surplus producing areas and increase the capacity of the
reserves. At this time, food was very cheap because of the record harvest in
Ethiopia. Thus, if a perfect weather forecast was available ahead of time, the
government could use the information to manage the import and export of food
crops, as well as manage land use policies and advise farmers on what, when and
where to farm. If there had been perfect information about El Niño and
its impact on Ethiopia, the government would not have been in emergency mode to
tackle the problem.
Credible and
perfect forecasting could also lead farmers to re-organize their annual
agricultural activities and household budgets. They could manage their
resources by reducing non-mandatory expenditures, such as weddings and feasts
in order to save for the expected reduction in food. With this information,
farmers could have saved their drought resistant crops, such as chickpeas. The
price of such crops rocketed by 300 percent in some areas in 1997. Others could
have sold their animals before the prices were depressed and bought food when
it was cheap. With accurate information on drought, many farmers would not
waste their energy and seeds. Others reported that they could have looked for
non-farm activities, such as petty trade or construction, to supplement the
expected reduction in output.
Effective
response to drought is related to development. Ethiopia is frequently impacted
by climate hazards because its crop and animal production are dependent on
rainfall and natural resources. The contribution of irrigated agriculture in
Ethiopia to the food system is insignificant. The Ethiopian market is not
integrated, and cattle herders in Ethiopia have a mobile wealth that is
destroyed by a single drought. Thus, even under normal weather conditions, the
people are confronted with widespread poverty. What is absent under normal
conditions is the mass famine and death as a result of starvation.
Therefore, the
various actions that ought to have been taken to reduce the impact should be
seen within the contours of the government’s capacity to finance and develop
the various possible actions to counter the impact. Even under normal
conditions, the Ethiopian government has been encouraging farmers to get
involved in development activities relating to soil and water conservation.
This is done through terracing and the construction of micro-dams. Even if
financing were available from external sources, there might not be the capacity
to implement it in between six and nine months.
An important
constraint regarding the implementation of the ideal El Niño forecast is the
fact that it will still be difficult to provide accurate information at the
local level. Many farmers in Tigray were angry at the forecast that the rains
would start late because, in some areas, the rains began as normal. However,
some farmers did not use the opportunity because of the advice they received
from the government. Thus, unless the relationship of El Niño with the local
climate is definitely stated, it would be problematic to give specific warning
to farmers about the timing of rains.
Ethiopian
governments in the past did not have the institutional capacity or the desire
to forecast droughts. In such cases, it is not useful to have perfect
information about the future if you are neither interested nor willing to
prevent it. In 1973-74, the Monarchy had the resources to respond to the hungry
people in northern Ethiopia by providing relief. At this time, there were
stocks of grain in the country. But the government did not consider that as its
job. In fact, it considered drought and famine as an act of God that was going
to happen once in a while. It is alleged that the government knew about the
disaster but hid it in order not to tarnish the “image” of the Emperor of Ethiopia.
Government
commitment, access to information, and the viability of its institutions,
infrastructure and resources are important elements in the success of
forecasting by analogy. The information required includes knowledge about
previous disasters in the country and the lessons learnt in dealing with them.
Availability of effective societal institutions, such as an effective and
stable government, manpower, and infrastructure and material resources are some
of the important components of effective response. It should be noted that
planning for the future based on the unknown is a very difficult task for any
government, especially when it requires spending scarce resources.
Current
Ethiopian leaders have learned from past experience to cope with future
droughts. In 1997, for example, the government at the highest level gave
instructions and follow-up meetings to all concerned institutions in order to
respond to the imminent disaster. This would minimize or prevent its impact.
They learned from the past and avoided those strategies that were not
effective. The response to the 1997 drought was effective because the
government had been working to build institutions with the objective of
preventing disasters. The sole objective of the Disaster Prevention and
Preparedness Commission (DPPC) is to provide early warning and organize
response to disasters. It has learnt from the disasters of 1972 and 1984-85. It
also learned from the floods of 1996 and put in infrastructure to prevent the
1997 anomalous floods. It also learned to apply the various traditional and
modern agronomic techniques of farming to reduce the impact of below-normal
rainfall.
One of the
beliefs of the current Ethiopian policy makers regarding disaster response is
the use of local manpower and resources. The preventative responses should be
done at the local level so that people can learn for the future. Food should be
distributed at the village level to avoid distress at migration relief camps.
The expansion of food security reserves is something that evolved from past
experiences. The DPPC makes its appeals to the local and international
community by telling them that unless they donate in time, the conditions of
the past (i.e. 1973-74 and 1984-85) could return again.
The amount of information discovered on the
relationship between El Niño and climate variability in Ethiopia needs to be explicitly
included into the national disaster plan. As we saw previously, the National
Meteorological Services Agency has been issuing weather forecasts based on El
Niño information.
However, the extent of awareness on the part of
the population about El Niño is very small. As we know, the various disaster
management documents, such as the five-year Plan of the DPPC, do not even
mention El Niño as a hazard. Such explicit inclusion would have made people
aware of the dangers, and they could respond to future disasters accordingly.
An explicit consideration can be useful because, when El Niño is on the scene,
different organizations in the private and public sectors could plan their
responses without waiting for the government. Importers and exporters, hydroelectric
distributors and livestock traders could take mitigating actions on their own
before the onset of El Niño.
Ethiopia has been a food deficit country even at
the end of 1996-97, when there was a record harvest. The causes for this kind
of food insecurity include over-population and lack of productivity of the
agricultural sector. A timely and credible El Niño-based forecast is important
for a government with perennial food insecurity because of the severity of
climate-related disaster. It is believed that there was no indication that
donors and NGOs were using El Niño information for program planning for the
year 1998 (Thompson, et. al. 1998, 99). This probably was due to their
continuous involvement in food aid, even outside El Niño years.
Identify the strengths and
weaknesses in the way Ethiopia's government system responds to El Niño-related
climate anomalies.
Ethiopia has a wide experience in disaster management. It has a
network of institutions, such as the NMSA, the DPPC and various disaster
prevention and preparedness committees at all levels of government, from the Wereda
to the Office of the Prime Minister. The DPPC is an established agency that
assesses the needs of people impacted by a disaster and mobilizes and
distributes relief. The relationship between the government and
non-governmental organizations involved in relief and development has been
improving from time to time. There is a political will on the part of the
government to mitigate drought and floods in the country. Ethiopia has also
been able to establish good relations with donors after the replacement of the
former Military Government in 1991.
The issuance of
forecasts close to the plantation period is a visible weakness of the use of El
Niño in Ethiopia. The Ministry of Agriculture, for example, complains that El
Niño-based forecasts from the NMSA usually coincide with the period of the
cultivation of crops, which makes early response to the impact of the climate
anomaly on the crops quite difficult. For example, to those farmers who planned
to plant long maturing crops, the early warning on May 29, 1997 was late by one
month. As the example of the Tigray National State shows, although the NMSA
forecast was released on May 29, the first meeting of the regional Disaster
Prevention and Preparedness Committee was on July 8, 1997. We were informed
that the Tigray Regional Early Warning Committee met two weeks after the
information arrived in the region from the head office in Addis Ababa. In July,
the rainy season had already started and the information was hardly a drought
early warning (Hadera 2000). The Tigray region tried to do everything possible
in spite of the contradictory information they were getting about the weather
outlook.
Ethiopia also
needs scientific models regarding the relationship between El Niño and
Ethiopian weather for various local areas. Ethiopia has a diverse local
climate, but it does not have accumulated meteorological information. There is
also a lack of trained manpower, especially at the regional level.
The existence
of a policy framework for inter-departmental collaboration of the committees is
limited at the local level due to a lack of common understanding of the issues
(DPPC 1998a, 6). In its periodic evaluations, the NDPPC has reached the
conclusion that some members of the early warning committee consider this task
as secondary to their departmental work. This was despite the provision of a
clear mandate by the government on the importance of inter-departmental
cooperation on early warning and preparedness. The DPPC states that a lack of a
skilled and experienced work force at the local level hinders the
implementation of policies related to climate-related impacts. There is also
lack of research interest outside the DPPC and NMSA. There is no ongoing
research at the universities about El Niño. There is also little awareness of
El Niño by the media or the public.
Ethiopia is
vulnerable to the legacy of decades of civil conflict, land mines, over-population,
and lack of agricultural productivity, which can change one season’s abnormal
rainfall into a disaster. The number of people affected by drought disaster in
Ethiopia is usually in the millions, which makes it difficult for the Ethiopian
people and government to respond from its internal resources only.