Minding the Gap: Echo Labs on Making Nature Count
November 18, 2026
Focused research organizations (FROs) are built to bridge gaps in the scientific landscape, solving bottlenecks and unblocking fields. Each Minding the Gap features one of the Convergent Research FROs. This time, we are featuring a short Q&A with Kaja Wasik, CEO of Echo Labs, a FRO funded by ARIA. Echo is building a set of tools and methods that make ecosystems measurable and usable across science, policy, and industry.
Q: What problem is Echo Labs working to solve?
A: Nature underpins our economies, yet it is largely invisible to the systems that allocate money and decide how land is used. At Echo we believe that this is largely a measurement problem: we cannot yet describe the state of an ecosystem in a form that is standardized and captures change over time.
We often understand complex systems by stacking layers of data: we did it for weather, combining layers such as pressure, wind and humidity to build forecasts. Ecology's best layers are borrowed: greenness indices such as NDVI come from satellites built for weather and were designed to track crops. Most of what we know about living things still comes from rich and very different data collected at single points on the ground that cannot be easily stacked or compared.
Echo is building the Ecosystem Fingerprint: a reproducible method for describing ecosystem state. We think of it as a stable set of core measures that can be collected anywhere and compared everywhere, plus context data for a particular place or question. This combination of data layers should be able to predict what matters about a place from the soil carbon to the species that live there.
There are a few problems that are central to this work. First, compression: an acoustic recorder, for example, produces more audio than anyone can analyze, so its output is usually reduced to a species list, discarding the rest. Instead, we want to create a compact signature that captures the site comprehensively, shifting only when the state of that site changes. Second, complementarity: when several sensors sit at the same site, we want to know what unique information each one adds and which are redundant. Third, going from sparse observations to dense data layers, where we use satellite data to connect measurements taken in different places and extend them into continuous layers.
Q: How did you get the idea for it? In other words, what were you doing when a gap became apparent to you?
A: The idea came while I was working between Kenya and Rwanda. In Kenya I lived on a conservancy, working with Molly (now my cofounder at Echo) and a team of ecologists who collected data about rangelands and how they responded to interventions. The approach was comprehensive and insightful but also labor-intensive and, being driven by specific research questions, hard to generalize and standardize.
In Rwanda I was working with Tosca (Echo’s second cofounder) on bringing nature closer to our economy. We gave mountain gorillas digital identities and wallets, so that other species, or more precisely their digital versions, could “participate” by spending real money based on their predicted interests. In an unusual reversal of the concept of who serves whom, it was the gorillas who paid for human services.
Together the two projects pointed to a gap. Ecology lacks system-level representation. What cannot be measured cannot be managed, and what is not represented is often overlooked. Nature is not integrated with our economy and therefore treated like a “free gift”. This is what we ultimately want to address.
Q: What has been a really interesting thing that has happened since Echo started?
A: Finding signals that we cannot yet name. We built a simple tool to explore our data after different models had compressed it: sound recordings, landscape photographs and satellite imagery of the same places. Each model turns the raw data into a list of numbers, and we wanted to see how those numbers of different sites compare.
For every model output there is clear structure. Places are not organized at random, instead they fall into patterns. But the models do not say what the patterns mean, and each one organizes the same places differently. We suspect the sound of a site carries information the satellite can’t pick up, and vice versa, but we cannot yet say what that is exactly.
Think about this as early maps of human genetic variation. They showed clear structure long before anyone could say which variants mattered and why. This is also an exciting validation that we are on the right track.
Stay tuned for more Minding the Gaps to learn about other Convergent Research FROs.

More from the blog
Sign up for our newsletter