Water bucket protocol

Theory of the Water Bucket Protocol (WBP) for uplift biodiversity crediting

NOTE: This is a bricolage protocol; which means we are building it in public! Some pages are under construction — but check back soon its updating fast!

The Water Bucket Protocol is Savimbo's field method for collecting environmental DNA (eDNA) at agroforestry plots, used to quantify biodiversity uplift over time. It feeds the Interoperable Biodiversity Unit (IBU) — the only currently certified biodiversity credit instrument. This page describes the methodology and its limitations; for field implementation steps, see the Practice Guide.

We intentionally chose a simple name for a practical solution to an incredibly complex frontier-science problem. The multidisciplinary source of the protocol is described in Appendix B and Authors contributed significantly to its refinement.

For step-by-step instructions on practical field use, see Practice Guide Water Bucket Protocol

Purpose of the WBP

This protocol is meant to standardize frontier-science eDNA analysis in remote field sites operated by Indigenous and local youth.

The water bucket protocol standardizes eDNA data collection. By using filtered water, a standardized time period, and plot centroids, it reduces noise and enhances signal in eDNA data collection. To recap the Biodiversity calculation section, this offers a consistent approach for comparing sites over time to assess changes in ecosystem integrity.

The aim is to reduce costs and improve accuracy when using eDNA as a metric. Results will need to be statistically normalized to a per-project standardized curve for integrity uplift over time. The WBP simply makes that easier by reducing the available data to a more meaningful set, reducing he sample size necessary to meet the output requirements for Interoperable Biodiversity Units (IBU) and qualify for recurring revenue from uplift biodiversity credits (roughly $300/unit, see Biodiversity unit).

Limitations of the WBP

There are serious practical constraints to field testing eDNA. The absence of established eDNA reference levels across ecosystems remains the primary challenge for using eDNA in biodiversity uplift assessments via IBUs, compounded by the logistical difficulties of exporting samples and the costs of analysis to obtain eDNA metrics. One possible approach is to establish a baseline sampling of eDNA in untouched or old-growth forests, representing conservation in Fig. 4C. Simple traditional biodiversity metrics, such as species richness, are well-suited for community-based monitoring but have low statistical power and thus require substantial sampling effort to reliably detect ecological changes (Lamb et al. 2009).

Random error in eDNA samples may also be high and affect data integrity (Lahoz-Monfort et al. 2016). Cost-effective monitoring thus requires statistically robust sampling, ideally with regional normalized curves — though these may be more expensive to develop in more biodiverse zones.

That said, we believe in progressing limits where we find them, as practically as possible. The protocol below is intended to reduce sampling noise in a cost-effective way.

Sampling location

While our IBU calculations are designed for one-hectare areas, sampling is performed at the centroid of a plot (averaging five hectares) to optimize for the conditions of the agroforestry system itself rather than the biodiversity that tends to accumulate at the edges of farmed areas. Plot size should be used as a covariate in later statistical modeling (see Biodiversity calculations).