This piece is part of our report from UNCCD COP17 in Ulaanbaatar, where soil health got its own day for the first time. The two constraints named most often there were scale and standards.
What is actually blocking people: scale and standards
We asked what challenges people shared with her. The answer came fast, and it was not about science.
"Scalability is the main challenge across both public and private, because we do have a lot of scientific data. We have plenty of solutions that can be implemented. The problem is that we're not yet able to scale from the small pilot level to rolling out to the hundreds of thousands of hectares that we need to restore."
Knowledge is not the constraint. Restoration practice is reasonably well understood at plot scale. What is missing is a way to run it, and to verify it, across an area large enough to matter.
The second constraint follows directly from the first:
"Another challenge is standardization. When it comes to data collection and data interpretation, we need to interpret outcome data in a way that is comparable, but it's hard to do that if we don't have standardized protocols on how to do that."
Put the two together and the shape of the problem is clear:
- Scale
- What it looks like in practice
- Solutions proven on pilot plots, no route to hundreds of thousands of hectares
- What it blocks
- Restoration at the pace the targets assume
- Standards
- What it looks like in practice
- No agreed protocols for collecting and interpreting outcome data
- What it blocks
- Comparing results across sites, seasons and countries
Standards are the quieter of the two and, for anyone who works with data, the more consequential. Outcome data that cannot be compared is not evidence. It is an anecdote with a number attached. Two projects can both report improvement while measuring different things, at different depths, in different seasons, against different baselines.
"Really, the key word is scaling. We do have some solutions. We have to start somewhere. We don't know how to scale yet, to the level that we need to in order to restore enough land to meet the objectives that we set as a global community."
Where we think we can help, and where we are still working
We asked Berenice where Soilytix fits against those constraints. Her answer starts with the cost of measurement itself.
"Part of the scaling problem, at least of monitoring soil health, is the fact that it's very labor intensive and it can be expensive to do soil sampling and testing."
That is the honest bottleneck. If monitoring one hectare costs what it currently costs, monitoring hundreds of thousands of hectares is arithmetic that does not work. The directive in section 4 sharpens this, because a monitoring obligation is recurring by definition.
What we are doing about it, in her words:
"We're currently working on incorporating as much satellite and Earth observation layers as we can to optimize our sampling strategy. And we are, as Soilytix, doing this within agriculture, and with STRATERRA we are doing this within forestry."
"A key part of monitoring and scaling the monitoring of soil health is having an economically and logistically feasible sampling strategy."
STRATERRA is the forestry side of that work. It is a joint effort between Soilytix and TransparenC, developed under the European Space Agency's SOCBIO-MAP research programme, with Forest Stewardship Council Chile as lead pilot partner. It combines satellite layers (Sentinel-2 optical, Sentinel-1 radar, GEDI canopy height and a 30 metre elevation model) with field samples analysed for physicochemical properties and microbial eDNA, and produces wall-to-wall estimates of soil organic carbon stocks, below-ground biomass and biodiversity indicators for forest carbon projects. It replaces conventional grid sampling with a field campaign designed around what the satellite data already shows.
To be precise about status: that is a research programme with a lead pilot rather than a finished product. The idea is simple to state and hard to execute. Satellite and Earth observation layers do not replace soil samples, because the things that matter most in soil biology are not visible from orbit. What those layers can do is indicate where to sample. If environmental context is used to stratify a landscape sensibly, the same budget buys a more representative picture, or the same picture costs less.
We are not alone in that read, and it is worth saying that we are following a direction the field has already set rather than inventing one. The WEF paper devotes a box to exactly this shift. It describes the Land Degradation Surveillance Framework, now running with partners in more than 45 countries, which pairs statistically robust field sampling with standardised protocols to produce comparable georeferenced data. It describes soil spectroscopy, where one calibrated scan predicts several soil properties at a fraction of conventional laboratory cost. And it describes the combination of the two with satellite Earth observation, where field measurements supply the ground truth and satellite imagery supplies continuous spatial coverage, so that predictive maps can extend beyond the sampling points. [1]
The same box carries the caveat we would want attached to our own work. Different decisions need different precision. A 30 metre satellite pixel can be adequate for national monitoring and close to useless to a smallholder managing one or two acres. The challenge, in the paper's words, is not generating more soil data but building evidence systems that are trusted, affordable and appropriate to the decision they serve.
That is the lever we think is real: sampling design rather than sampling volume. It addresses the cost side of the scale problem, and it is worth something only if the protocol behind it stays consistent enough for results to be compared later. Which returns to the second constraint Berenice named, and that is the one we take most seriously.
"Earth observation helps us design smarter sampling strategies by showing where conditions vary across a landscape, so field samples can be placed where they provide the most representative evidence. It does not replace soil sampling or directly measure soil biology; it helps us use fieldwork more efficiently while maintaining a consistent, defensible protocol." Maurice, CTO Soilytix
FAQ
- What are the main barriers to land restoration at scale?
- Two came up repeatedly at COP17. Scalability, meaning the gap between solutions proven on pilot plots and the hundreds of thousands of hectares that need restoring. And standardisation, meaning the absence of agreed protocols for collecting and interpreting outcome data, without which results cannot be compared across sites.
- Why does standardised soil data matter?
- Because outcome data that cannot be compared cannot be used as evidence. Two projects can both report improvement while measuring different things, at different depths, in different seasons, against different baselines. Monitoring obligations assume comparability over time, which only holds if the protocol holds.
Sources
- UNCCD, Andrea Meza Murillo, Deputy Executive Secretary. unccd.int