VALORIZATION OF DATA
Key Prerequisite for Continued Data Collection
Above, we have shown that any monitoring campaign carries a price tag. The question in this context then becomes how the data that are acquired can be properly valorized. While acknowledging that answers to this question are highly context specific, we continue our argument for the case of irrigation agriculture and the application arenas there since this conforms to the original iMoMo Project focus.
THE PERSPECTIVE OF INDIVIDUAL WATER USERS
The valorization of data is concerned with turning data into direct and discernible benefit for stakeholders. In agriculture, several possible pathways exist for doing so. For example, data on nutrient levels in soils can help to guide fertilization practices and thus improve crop out-comes. Knowledge about the type of pest that has affected a particular crop on the field allows for targeted spraying of pesticides. And a good understanding of soil moisture levels and crop water requirements can minimize crop water stress during the critical crop growth stages, thus avoiding negative impacts on yields. While the availability of these types of data all allow theoretically improvements in income levels, proper valorization depends not just on their timely availability and quality but also on the proper knowledge on how to turn data into actionable information. For example, the diagnosis of a soil moisture deficit on a field may ultimately cause overwatering and water logging because of overwatering by the farmer whose field is affected by the water deficit. Data hence carries a risk in itself if wrong conclusions are drawn from their analyses.
Another example is the reduction of water fees when, for example, a water fee charged by a flat rate tariff can be reduced by proper monitoring of abstracted water for volumetric pricing. Similarly, in the case where there is volumetric pricing, the monitoring of soil moisture can lead to a reduction in wasteful water use and thus lower the total price for the production factor water.
In all these examples, the valorization of data works in the sense that data can help to optimize the application of production factors for best-possible crop outcomes. This however depends on several factors, one of which was already mentioned above, i.e. the ability for stakeholders to turn data into actionable knowledge. Another precondition that is common to the examples described above is the presence of production factor pricing. In situations where water has no price, there is no incentive for conservation and also no eagerness to better understand the use of the resource in relation to actual field-level needs and consideration of economic effectiveness.
It appears thus that for a proper valorization of data in irrigation agriculture, input needs to be priced, ideally reflecting also the scarcity values of the resources under consideration. The latter point is relevant especially in the case of water that is normally, for various reasons, heavily underpriced. Resource pricing carries, however, a risk for the data-keen water user since his/her costs for resources use could possibly rise under variable and flexible volumetric monitoring as compared to the fixed permit price if the latter was not very low or the user simply abstracted much more as compared to what his/her permit allowed.
Agriculture, however, is very peculiar in another regard. Namely, farmers are confronted with delayed rewards. That is to say that while the agricultural season demands the attention of the farmer mainly during the initial and middle crop growth stages, the ultimate yield outcome can only be harvested with considerable delay, i.e. at the end of the growing season, usually weeks or months later. So many things can still go wrong between the last watering and application of fertilizer and herbicides/pesticides and the point at which the harvested goods are sold either on the market, to a contractor or other middle men. For example, locusts can wipe out late season fields, an extreme hailstorm destroy the tedious work of the previous months, a herd of elephants trample through the vegetable field and so on. The link between input and output, in the context of delayed reward that is an innate feature in agriculture, is far from obvious, even under the best possible conditions. Clearly, in many developing nations, crop insurance protects farmers against unexpected losses. Often however, these risk minimization vehicles are not available for farmers in developing regions.
In short, unlike in the case of the owner of a groundwater well who can expend energy to directly get the benefit of having more water available for irrigation and drinking, the simple availability of data on water flows today does not guarantee equal amounts or even more water tomorrow and the day after and so on.
In circumstances as described above, it thus becomes imperative to properly understand the local stakeholders needs for data and also see to what extent he or she can turn this data into value, either through a reduction in expenses or an increase in income or both. If this cannot be convincingly elicited at the beginning of a monitoring campaign through local involvement, there is very high likelihood that the campaign will turn out to be a damp squib and collapse after its initial setup soon thereafter, but certainly after the termination of a donor-supported intervention project.
Despite the complexities surrounding data valorization, one has to emphasize that projects, including for research which need quick and robust results in terms of making large amounts of data available over a predefined and limited time horizon can greatly benefit from the flexibility and ease of deployment of crowd-sensing technologies, also and especially due to their low investment costs.