Drought is a global problem that affects an estimated 1.5 billion people, particularly those in the Southern Hemisphere. Between the 1970s and the early 2000s the percentage of the earth’s landmass affected by severe drought has more than doubled.

Drought has many unique identifiers that distinguish it from other natural hazards. First, drought is a creeping phenomenon whose onset and end are difficult to determine with precision. The residual effects of drought linger for a considerable period after the event’s end. Secondly, the absence of a standard definition of drought adds to the confusion regarding its recognition, classification, and the measure of its severity. Lastly, drought impacts are usually non-structural and spread over more expansive geographical space, mostly crossing frontiers uninvited.

Experience across Africa and elsewhere has shown that drought results in significant impacts regardless of the development level of a state, resulting in water and food insecurity, land degradation, agricultural losses, human migration, and conflicts.

However, the character of the impact differs profoundly. Drought affects the vulnerable regions of the arid and semi-arid lands of the world most because the social and economic support systems cannot withstand its effects, and the people have limited options and inadequate coping strategies during times of water and forage shortage.

This non-structural nature of drought has hindered the development of an accurate, timely, and reliable assessment of its severity, geographical magnitude, and ultimately the formulation of sound contingency plans by the government and its development partners. On the other hand, there is confusion in the scientific and policy community about the character and manifestation of drought. Drought management is primarily a policy question. Until we fully crystalize the status of drought policy and preparedness, proactive and effective drought management will remain a hit-and-miss endeavour.

The cost of drought

In Kenya, fifteen droughts were recorded between 1960 and 2016, with the 1970s and 80s being described as decades of lack. Six drought episodes (40 per cent) occurred after 2000, an indication that droughts are increasing in both frequency and severity.

The economic costs of drought in the 1998-2000 period were estimated at US$2.8 billion, while the post-disaster needs assessment for the extended 2008-2011 drought estimated the total damage and losses to the economy to be a staggering US$12.1 billion.

Part of the reason why previous droughts have caused considerable losses is because of how planning and interventions are conceptualized. Traditionally, drought is viewed as a one-off disaster that requires an emergency response focused on a few short-term activities like delivery of food aid and other life-saving humanitarian support, rehabilitation of boreholes, emergency vaccination campaigns, etc.

The understanding of drought and the logic of drought management have gradually shifted, however. Today, drought is viewed primarily as a cyclical process, and drought interventions are now structured along a cycle of four warning phases: normal, alert, emergency, and recovery.

This shifting view comes with some advantages. First, matching activities to a specific stage of the drought cycle improves the timeliness, appropriateness, and ultimately, the effectiveness of the drought response. Secondly, the common framework offers space where humanitarian, development, and advocacy work can be aligned to complement each other.

Although this fits in well with the programmers’ and the pastoralists’ own understanding of the drought cycle, and by extension with the improvement of the overall coordination of drought management in Kenya, we still have a huge gap in tackling drought impacts in the short and long term. In addition to high-cost short-term emergency interventions to save lives and livelihoods, the Kenyan government launched the Medium-Term Plan for Drought Risk Management and Ending Drought Emergencies (EDE) for 2013-2017. The EDE strategy committed the government to end drought emergency by 2022. However, a huge gap remains due to a number of factors. Despite improvements to early warning and contingency planning systems, drought management remains largely reactive, rather than being an anticipatory and preventive risk management undertaking. There is a gap between the information provided by the early warning systems about impending threats and the ability of the government to act to reduce those threats.

Until we fully crystalize the status of drought policy and preparedness, proactive and effective drought management will remain a hit-and-miss endeavour.

Moreover, although county drought management plans are well developed in a participatory manner, they largely contain pre-prepared “off-the-shelf” projects whose implementation is triggered by the early warning systems, but without sustainable funding for their execution. The availability of financial resources should be guaranteed, and the money held in a contingency fund mechanism that is operational at the community and county levels.

However, the National Drought Emergency Fund is yet to be operationalized, thereby delaying timely and appropriate emergency responses. The current drought management system remains centralized and lacks ownership, resources, and technology at the county level to pre-empt or minimize the magnitude of the losses that often follow the onset of droughts. Moreover, the drought management system is not strongly connected to county-level resource allocations.

Linking drought indicators and actions

Past drought mitigation efforts have been largely reactionary and have done little to influence preparedness for future droughts. This is largely attributed to a number of concerns: the unfavourable status of drought preparedness and, in particular, weak institutional capacity; the bottleneck in plans and policy development, especially the level of understanding of the policy and planning needs; and a weak drought warning system, among others. Assessments across African countries have identified problems common to some drought-affected countries including Kenya. These include low levels of interest in planning beyond the relatively short window of opportunity that follows successive drought cycles. Mostly, interest in drought contingency planning wanes in the post-drought phase as soon as precipitation moves to normal or slightly to above normal, whereas drought preparedness planning should be an integral part of development planning and the institutional structures necessary should be in place.

Responsive drought warning systems 

Currently, we have wide gaps between meteorology and hydrology and between hydrology and humanitarian need in most drought-affected counties in Kenya. We need to galvanize real-time planning that could potentially prevent drought from turning into a crisis. To this end, our weather forecasting should provide informative and operational data that characterizes the development of a drought. A robust system of indicators that can identify and diagnose anomalies and provide the basis for early detection of drought events is the basis of any drought management plan, with operational data used as indicators to trigger specific actions. Operational variables can be linked with activities in the drought plan. Indicator systems will allow for early warning of the drought events and time to activate the programme of measures established for the emergency. Local or indigenous weather condition indicators can also be integrated to complement other scientific indicators.

The current drought management system remains centralized and lacks ownership, resources, and technology at the county level.

Early warnings without proper data and mechanisms for utilizing the data could prove futile. For example, the poor rains in late 2010 in East Africa were successfully predicted, but the failure of the subsequent anticipated longer rainy season that led to severe drought in 2011 was not predicted. Although early warning systems gradually increased, we also need to develop a system that is multidimensional. For example, one could focus on climate and water conditions and the other on food security. We could borrow useful lessons from Canada, which has officially developed a monthly drought-monitoring portal, with an interactive map and narratives visualizing and describing drought conditions by region. If we emulate such a system, we could present the severity, spatial extent, and impacts of drought by county on a monthly basis and generate judicious choices of appropriate interventions that can be applied in a timely manner. This can done if more funding goes to research and strategic investments are made in building a superior drought monitoring system.

Reducing drought risk through land restoration

Nature-based solutions have significant potential to reduce drought risks in arid and semi-arid areas. Such an approach to drought management should be centred on at least two broad programme areas: reviving hydrological functions through rehabilitation and sustainable land management to reduce the severity of drought; restoring biodiversity to foster adaptation and diversification. Land restoration and conservation of water catchments will restore hydrological processes and reduce the overall frequency and severity of drought. In some places, lack of rain is compensated by access to underground water, artificial reservoirs or moisture stored in soils across watersheds. As such, programmes oriented towards land restoration and sustainable water management are important for the long-term recovery of the ecosystem.

The development of strategic water sources for sustainable rangeland utilization is also key to drought mitigation. Small-scale strategic water resources like sub-surface dams, sand-dams, water pans and limited duration boreholes that can be closed during certain periods and adjusted to the seasonal availability of adjacent pastures, are critical in boosting herd survival during drought and should be a key priority in drought mitigation.