What is the right spatial and temporal scale needed for a meaningful CBWT?

Alexis Morgan — WWF Global Water Stewardship Lead and Rylan Dobson — Water Stewardship Consultant

Views expressed in this article are those of WWF and don’t represent those of other organisations working in the CBWT initiative

© Day’s Edge Productions

As a geographer and hydrologist, the concept of spatial scale is something that is often at the forefront of my mind. Catchments (aka watersheds or river basins) can range from the size of a continent down to nearly your backyard. As companies increasingly embrace the concept of water stewardship, and shift their lens to that of the catchment, the question naturally arises: what is the appropriate scale of action for a company to work at?

This issue arose during the development the Alliance for Water Stewardship (AWS) and as we’ve begun working on Context-Based Water Targets (CBWTs), a similar question has arisen: what is the “right” size of catchment to use to make a CBWT meaningful? Simplistically, and setting aside aquifers, smaller catchments are more manageable and allow for a bigger influence. However, smaller catchments may not be sufficiently large to impact on the status of the catchment and therefore not address the shared water challenge. Larger catchments may be more meaningful, and allow for collective actors to collaborate in a manner that will affect the status of the catchment, but can be more difficult (costly in terms of time and resources to engage), especially as the catchment gets very large.

Further complicating the matter is the fact that when it comes to CBWT setting, spatial scale is not the only “scale” variable to consider; we must also consider temporal scale. Not only does water vary throughout the year, but it varies year to year as well. Accordingly, asking: what the appropriate temporal scale for a CBWT is — both in terms of setting the target and then delivering the target — is an equally relevant question.

Working at multiple scales

To begin to answer these questions, it’s important to put yourself in the shoes of the practitioner on the ground, and to reflect upon what is already employed. Water users, most often businesses, are accustomed to operating at various spatial scales already. A production site has its own factory walls, its property, its community, city, state/province and country. It operates within these spatial scales already with different requirements at each level to a certain extent, though it should be noted that jurisdictional requirements are typically broken up (e.g., water pricing is set at the municipal level, while granting of water use licenses at sub-national or national levels). Similarly, businesses are also used to working at multiple time scales — be it daily production figures, monthly status updates, quarterly earning announcements, annual budgets and reports. In short, business has a long history of operating at multiple spatial and temporal scales, so they should be comfortable with the broad concept.

© Xiaodong Sun / WWF-UK

Water authorities too, typically work with various scales. Regulatory allocation of water resources is typically cascaded down from the largest catchment scales, while other approaches to allocation aggregate up from smaller catchments to larger ones (e.g., the US Environmental Protection Agency’s Total Maximum Daily Load — or TMDL — process, can employ cumulative loading thresholds for nutrients).

Returning to CBWTs, we at WWF believe that there is a case to be made for employing multi-scale CBWTs. In other words, to set targets at more than one spatial scale and more than one temporal scale. Allow me to explain to start with a single catchment.

A simplified solution to spatial scales

In our experience, when you talk to a site about their catchment, a site typically immediately considers its most local “named” catchment (think of this as the area that feeds a stream within a neighbourhood). If they’ve considered water issues more thoroughly, they will also know the next order or two up as well (say “city” and “sub-region/province/state”) but few sites consider how they impact, or depend upon, catchments at the continental-scale.

So, the first level of catchment scale to consider is one that is, as a rough estimate, often similar to the community (or part of a larger city) from which the site draws its staff. Using the global catchment dataset “HydroSHEDS”, we would equate this to roughly a HydroBASIN Level 9 basin. The second level of catchment scale to consider needs to be large enough to account for upstream dependencies and downstream impacts. This should be guided by considering one’s water source and discharge locations, but also needs to account for the size of the operation (how much withdrawal and effluent is under consideration). Once withdrawal and discharge locations are identified, all areas downstream should be selected as well as those areas upstream that are relevant and/or contribute a disproportionate amount of flow. This will likely select one or more HydroBASIN Level 7 or 6 basins.

Beyond Level 6, the catchments become too large to affect. Similarly, beyond Level 9, the catchments generally begin to become too small to be meaningful, and ultimately, a site would end up back with just its own boundaries as the catchment. In between (between level 7 and level 9) lies a bit of a “goldilocks zone” — large enough to be material for impacts and dependencies, but small enough to be feasible for sites to engage local stakeholders and realistically engage on collective action. This range is illustrated here for Southern California (around Los Angeles) in Figure 1. For smaller catchments, the target should be a greater proportion, while for larger catchments, the target should be a lesser proportion. In theory, the target should be transferrable between different scales.

©Thor Morales_WWF Mexico

Taking this logic a step further: in theory, for more advanced sites, we could also envision multiple targets being set for different scales — with each reflecting the proportionate impact/contribution at the relevant scale. For example, pollution effluent may be at one spatial scale, while water use may be at another scale.

Questions that a site can contemplate when choosing the most appropriate scale include:

  • What catchments are my site embedded within? See HydroSHEDS to review this question (especially levels 7–9).
  • Where is my site positioned in relation to the source of water that I rely on and the mouth of the river?
  • Where is the source of the water that my site withdraws located?
  • Where do I discharge my effluent and if into a river what water users are downstream?
  • If sourcing from or discharging to a third party, where do they source/discharge their water?
  • How far downstream does it take until that effluent is fully diluted?
  • Proportionally, am I a big user of water or emitter of pollution in the context of the catchment I’ve selected?
Figure 1: HydroBASIN Level 7 in Dark blue, Level 9 in light blue

Testing thinking on temporal scales

With some initial guidance in place for how to address spatial scale, the next question is that of temporal scales. Like spatial considerations, there is an optimal and variable range. Hourly-to-daily time steps are generally, from a management perspective, overly prescriptive, though still potentially relevant to setting meaningful water targets (water quality sometimes varies significantly over the course of hours). Weekly-to-monthly time steps are likely to be most useful from a management and budgeting perspective (typically tied to volume), while annual time steps are also useful for planning/budgeting as well (e.g., many jurisdictions do annual water allocations). Not unlike cash flows, water must be planned for and managed during times of abundance or scarcity — both at a monthly basis, but also an inter-annual basis. If it becomes overdrawn in one year, then it needs to be adjusted the following year to enable replenishment in the system.

Stitching it all together

With priority locations selected and shared water challenges identified, it would give a sense of what issues to cover and the spatial and temporal nature to consider. For example, perhaps you had a production site in a catchment (e.g., San Gabriel) near Los Angeles (Figure 1) that was concerned about scarcity issues. Since scarcity is a function of an annual cycle (rain/snow & drought cycle), month-by-month management makes sense. Spatial scales will run up to the source waters, which in the case of LA is challenging since it covers not only the local mountains (likely captured by the Level 7 basin in the figure), but also transfers from the Central Valley/Sierra Nevada Mountains as well as the Colorado River (which are in fact at a scale closer to HydroBASIN Level 4). In such a case, maintaining base flows

(especially during the dry summer months) for the downstream component of the San Gabriel river could form part of the target for the Level 7 catchment. In addition, there may be some aim to minimize use that draws from the Colorado & Central Valley during their low-flow periods.

Scales — both spatial and temporal — are a challenging aspect to context-based water targets. Both spatial and temporal scale will need to be addressed during in the development of a CBWT in a way that results in a meaningful target. This will mean respecting dynamic spatio-temporal contextual thresholds while at the same time, being simple and consistent enough to offer viable management plans/actions.

It is not a simple challenge, but WWF is actively working with its partners to pilot some initial thinking with respect to guidance for spatial and temporal scale selection and we will seek to publish this when it has been developed further. If one thing is for certain, CBWTs are a learning journey for us all!