Considering socioeconomic parameters in triggers for anticipatory action

Are you engaged in anticipatory action? What is your take on considering socioeconomic parameters when setting up thresholds and triggers for early actions ahead of extreme weather events? As a social scientist, I wonder how to understand socioeconomic parameters more fully, and how to better incorporate them into trigger mechanisms.

What is a trigger in anticipatory action?

In forecast-based financing, a trigger is used to activate early actions. For extreme weather events, triggers are set up based on weather forecasts. Once the forecast hits the specific threshold value, the early action protocol (EAP) for that hazard is triggered, and funding is automatically released to implement pre-defined early actions. The threshold varies depending on the type of hazard, as well as the coping capacity and other underlying socioeconomic factors of the exposed community.

What about socioeconomic parameters?

Although triggers for weather-related hazards are predominantly based on weather forecasts, certain socioeconomic parameters are important in analysing, setting up and activating triggers. For example, the Bangladesh Red Crescent Society’s EAP for Floods for the Jamuna River Basin is activated once the water level is forecast to be at 0.85m above the danger level at the Bahadurabad measuring station (the danger level set by the government). But there are also several socioeconomic aspects – such as the size of the dependent population, and the level of poverty they experience – that can influence the decision on whether and when to activate the EAP, and where to implement the early actions.

How does this work in practice?

The Bangladesh Red Crescent Society, the German Red Cross and the Red Cross Red Crescent Climate Centre have developed an ‘alert tool’ that helps to analyse weather forecasts and convert them into an impact-based forecast that defines the impact levels for a union (part of a sub-district), highlighting where early actions need to be implemented. This alert tool incorporates socioeconomic, exposure and vulnerability parameters to identify a list of unions for intervention.

When defining the trigger for floods in the Jamuna River Basin (or cyclones in the coastal zone), we considered:

• the percentage of the population expected to be affected
• the percentage of household assets expected to be damaged
• the level of exposure in different vulnerable ‘pockets’ (e.g., Char Island, which lies outside the protection of an embankment)
• the type of housing structure, such as brick-built (pucca), semi brick-built (semi-pucca) or temporary (kacha)
• people’s education and knowledge, which indicates their level of awareness of risk.

Through analysis of the secondary data available and a field investigation, we were able to assess the impact limits in terms of damage to household assets and the percentage of the population affected. Next, we synchronized the forecast threshold with the impact threshold. This led us to determine impact levels of 25 per cent or more for household asset damage, and 40 per cent or more for the population affected in the respective unions, while keeping the forecast threshold (river water level) at 0.85m.

These socioeconomic parametres still need to be further analysed and fully embedded within the EAP’s triggering process; at present, they are only considered as background parameters. For instance, Char Island and other low-lying area in the braided belt of the Jamuna, could be prioritized for early actions, given their high exposure.

How can socioeconomic parameters be mainstreamed into triggers?

There are several factors to consider here. First, socioeconomic parameters could be embedded with weather-related trigger values, or kept separate. If they are kept separate, triggers will have two components: weather forecasts and socioeconomic parameters. Each could be assigned a certain weighting when considering when to activate an EAP.

Second, socioeconomic parameters for triggers could be contextualized based on seasonality, with the possibility of different levels of damage to livelihoods and assets in different seasons. For example, a trigger threshold for flash floods in Haor, a wetland area in Bangladesh, could be tailored for the paddy-harvesting season, so that farmers can harvest early to avoid the loss of their crops.

Third, triggers do not yet consider the cumulative impacts of hazards. For example, some hazards have recurring (e.g., annual) impacts on a community; the floods in the Jamuna River Basin are an example. Cumulative impacts trap households and communities in a vicious cycle, reducing their coping mechanisms. How do we consider this in triggers? One way could be to analyse past disasters each year and consider them in the triggering process. The capacity to recover and/or the time needed for this differs for different households and communities; it would be useful to further explore how this could be incorporated into triggers.

Fourth, the return period – which describes how likely a hazard event is to occur at, or above, a specific intensity within a time frame – is a key component of anticipatory action mechanisms. However, a hazard that affects a wide area may cover regions with huge differences in terms of socioeconomic parameters. A region with a low performance in socioeconomic parameters might need a lower return period, or lower forecast trigger value, because even lower-intensity events might have major impacts for people there. For example, people living on Char Island are impacted by floods every year, or even several times in a season. But people who are better able to cope may not be so severely affected by floods of this intensity. Triggers for EAPs should therefore not only be determined by the scale of an extreme event, but also the underlying factors that determine its impact – which are mainly determined by socioeconomic conditions. The question is whether the weather-related trigger thresholds should be lower for these most vulnerable communities.

The way forward

There are existing sources of information that can support the process of considering socioeconomic parametres in triggers for anticipatory action. For example, the INFORM Index could be embedded within the triggering mechanism to identify early actions for the most vulnerable communities; the INFORM Severity Index and INFORM Warning (in development) are applicable to early warnings and early actions,  but need to be contextualized to local perceptions of risk.

Considering socioeconomic parametres in triggers is likely to be challenging, however, especially for early actions with short lead times. But anticipatory action is an evolving approach – and we must keep exploring how to make triggers more comprehensive and effective. The points listed here give us a place to start: some ‘low-hanging fruit’ to address before we move towards a more comprehensive mechanism for incorporating socioeconomic parameters in triggers.