Towards Forecast-based Preparedness Action (ForPAc): Probabilistic forecast information for defensible preparedness decision-making and action

Lead Research Organisation: University of Oxford
Department Name: Oxford Physics

Abstract

Drought and flood events remain a major threat to lives and livelihoods in sub-Saharan Africa, with significant impacts on long term development, due to the high exposure and vulnerability of populations. Existing early warning systems (EWS) whilst improving remain insufficient to protect vulnerable populations. Too often agencies and communities are only able to respond after a disaster has occurred rather than in advance, for a number of complex reasons. This project will address two of the primary limitations of EWS that hinder effective action in the face of hazard risks by;
1. Increasing the credibility and pertinence of hazard forecasts, by developing improved weather-climate forecasts and associated livelihood impacts over a range of 'seamless' lead times from days to seasons.
2. Overcoming barriers to preparedness action in advance of hazard events through the development and trialling of systematic and defensible approaches based on forecast information.
Our project consortium brings together world-renowned research institutes in the UK and East Africa with expertise in forecasting science, hazard impacts and vulnerability, with agencies responsible for EWS and humanitarian action. By building on strong and well-established existing partnerships we will ensure effective collaboration, co-production and integration of research directly into practical application. Our project focuses on a set of existing EWS for flood and drought in Kenya, providing a strong platform for operationalizing and rapid uptake of results, new approaches and tools. The EWS case studies include both urban and rural contexts and are characteristic of hazard and development situations across much of Africa. Through our extensive consortium networks and parallel projects we will share the benefits and lessons with similar systems across Africa and maximise added value and value for money.
Our scientific advances will include development and evaluation of state-of-the-art weather-climate risk forecasts expressed in a probabilistic form. The physical basis of forecast skill will be established. These products will be linked to decision-relevant impacts on agricultural and pastoralist livelihoods. Our hazard risk forecast products with associated skill assessment, will be co-developed together with those agencies who operate EWS and plan for preparedness actions. In this way we will ensure the pertinence and credibility of forecast products making them more suitable for uptake. We will incorporate these new forecast products into leading on-line hazard risk portals.
Successful uptake requires careful integration with other information sources and effective communication to the range of users. Critically, decision makers need to have confidence in the information but also effective ways of using the information for disaster risk reduction. To this end, we will develop and evaluate novel methodologies and practical tools of Forecast based Action that, by linking forecasts' attributes to risk reduction actions in well defined action plans, overcome institutional and technical barriers to preparedness action
The FbA approaches must take account of the reality of the decision-making context in which the various actors have complex agendas, priorities, and cultures related to risk. We will adopt a set of flexible strategies to apply these approaches within a range of EWS contexts. For this reason, our demonstration projects have been chosen to be characteristic of similar contexts elsewhere and lessons learned will be shared, and generalised where possible, to other regions.

Planned Impact

ForPAc is carefully designed to have direct and measurable impact during, and well beyond the timeframe and location of the project activities in Kenya. We will utilise a Participatory Impact Pathways Analysis (PIPA) framework and method to plan, monitor and evaluate the impact of our project. This will be rooted in, and flow from a coherent 'Theory of Change' established early in the project. Management structures include; 1). An impact sub-group led by co-I APHRC. 2) An advisory Panel (contingent on advice from SHEAR KE) providing strategic guidance, with representation from key regional organisations, civil society, donors and multi-laterals. ForPAc will generate impact because (i) we will improve existing EWS (ii) we include the key decision-making agencies and intermediaries to communities, building on partnerships already well established. The case study method of co-production of knowledge and stakeholder-driven deliberative processes will result in co-ownership of the process, direct applicability and development impact. We will strengthen the 'community of practice' in Kenya and beyond including 'champions' in relevant organisations, for continued legacy.

Benefits will accrue at 3 levels

(i) Local stakeholders. 1) 'delivery agencies' at Kenya county level e.g. government (agriculture, emergency services), and NGOs' operatives 2) communities themselves. Benefits: greater awareness of ForPAc innovations in early warning information and tools/approaches for preparedness planning. Method: enabling process of deep stakeholder engagement from the outset, starting with EWS mapping processes, and iterative process of EWS and portal development, within our continuous monitoring and evaluation.

(ii) National mandated agencies. The NDMA, KMD, Red Cross Kenya and NGOs within ALERT network. Operational practices will be directly influenced by the project's forecast-based action research e.g. improved methods of risk forecast production, early warning advice based on agreed decision "triggers" and improved standard operating procedures through new and improved tools. Through collaboration, iteration and co-production ForPAc will ensure a real sense of ownership of these innovations. Enhanced EWS capacity will strengthen national trust in the mandated agencies - encouraging uptake across government support in the future and will inform policy planning on DRR by government ministries. Outputs (products, portals training modules and decision tools) and lessons will be shared: Across IGAD nations through ICPAC the Greater Horn of Africa Climate Outlook Forum process; Across Africa through the wider networks of the team. In this way legacy impact is ensured.

(iii) 'Advocacy' level. Regional and international agencies that set agendas for DRR, such as ISDR, FAO, WMO, WFP, FEWSNet and national donors including DfID, utilising the extensive existing policy network of the project consortium. We will inform key policy makers and forums of best practice, cutting edge science, emerging challenges in the science and implementation of our innovations in EWS. Method: a programme of policy engagement including workshops and meetings with invited high-level agency representatives, interventions at forums, policy briefings.

Our UK/Kenya team has an excellent track record of delivering impact and policy engagement (see CVs and track records). Our Co-I's Pelling and Visman have experience in NERC Knowledge Exchange. ForPAc impact will draw on lessons from a wide array of related projects with Co-I involvement e.g. FCFA IMPALA and consortia; AdA consortium; ALERT; Urban-Ark; the Red Cross-Kenya Resilience Initiative; BRACED projects; WISER projects; International forecast initiatives of the WMO (SWFDP, S2S); EUROSIP and Copernicus seasonal predictions; Foresight 'Reducing Risks of Future Disasters'. These projects will also be cross-fertilised by ForPAc potentially multiplying our impact considerably.
 
Description We have assessed the forecast skill and reliability of rainfall forecasts for East Africa across a range of timescales. This includes:

Evaluation of the reliability of multi-day average for extreme rainfall for the Nzoia river basin for western Kenya, finding that the potential for early warning of flood-relevant rainfall events is high, supporting the extension of the operational early warning system beyond the current three days.

Assessment of the seasonal forecasts for the Greater Horn of Africa, showing that the short rains are predictable several months in advance of current early warnings. This information will be provided as a pilot in 2019 to support drought preparedness, alongside training in the use of probabilistic forecasts.

Set up of several climate model experiments, designed to test questions of predictability of East African climate. Once analyzed, these have the potential to provide the scientific basis for new sources of predictability and a more informed use of forecast information by project partners at regional climate prediction centers.
Exploitation Route As part of the ForPAc project the work on understanding predictability is directly informing 3 case studies in Kenya aiming to improve early warning systems for flooding in Nairobi, river flood in Nzoia and drought in Kitui.

These case studies are aiming at improved reliability of forecast information as well as extending lead time of current warnings. There are concrete plans in place to achieve this, such as the planned inclusion of reliability calibrated forecasts long lead seasonal forecasts and monthly forecasts in the National Drought Monitoring Authority monthly bulletins.

We have also been instrumental in facilitating climate information training workshops in Kenya. In these workshops (one in Kitui on drought and there other in Nairobi on flooding), relevant groups of cross-sector stakeholders have engaged for the first time with long lead probabilistic information. There are also plans to train key decision makers at NDMA in the use of the specific forecast products which will be provided as part of the ForPAc drought case study.
Sectors Agriculture, Food and Drink

URL https://www.sciencedirect.com/science/article/pii/S2212094717301986
 
Description Analysis of subseasonal forecast skill carried out at the University of Oxford is being used as part of the guidance for a pilot of a weekly real time subseasonal forecast coordinated by the UK Met Office and disseminated through the Kenya Meteorological Department.
First Year Of Impact 2018
Sector Agriculture, Food and Drink
Impact Types Societal,Economic,Policy & public services