We are devoted to making any product developed by Quick Carbon transparent, open-source, and accessible to researchers and practitioners alike. We see the opportunity to fill the void of rapid carbon measurement as something much greater than our own initiative — hence, we're sharing what we're doing and welcoming your feedback. If you are equally passionate about the need for this technology, we’d love to work with you to accomplish our objectives faster and more effectively.
To build a tool that allows users to make spatially explicit estimates of soil C concentrations and stocks fully in-field with a degree of accuracy that enables quantification of management interventions or landscape variation.
This tool will remain transparent, open, affordable, user-friendly, and can be applied by land managers and scientists.
We intend to meet our objectives through four work-streams. While each work-stream is dependent on the others, this process allows us to take a modular approach to collaboratively build an open project.
Spectrometer Optimization. The reduced-size VNIR sensors used by Quick Carbon employ only a handful of the wavelengths that bench-top VNIR spectrometers do, but our data reveal their capacity to estimate SOC with reliable accuracy. In this first work-stream we will work with (a) publicly-available datasets of VNIR spectra to select wavelength combinations that have the most power to estimate SOC concentrations and potential predictive variables (e.g. clay content); we will also work with (b) soils sampled at much higher spatial densities, within-landscapes, to test how spatial variation (or scale) affects the choice of wavelengths, given the potential that among-site analyses can fail to capture predictors of soil carbon dynamics that dominate within-sites.
Sample Design. We will continue to build and enhance our ability to intelligently sample large (and small) landscapes for soil carbon. At the core of this work-stream is building web-based tools that are: a) accessible to both land managers and researchers; b) dynamic and flexible for multiple landscapes and contexts; and c) robust and statistically sound. We will accomplish this through the design and testing of web-based mapping tools like Stratifi.
Dynamic Analysis and Visualization. Under this work-stream we will develop analytical and visualization applications that return readily-interpretable information about SOC – in real time – to the scientist and land manager in the field. These applications will apply machine-learning approaches to integrate data from field application of the spectrometers dynamically with GIS data sources and soil spectral libraries to produce spatially-explicit estimates of SOC concentrations with measurement uncertainty. The collective aim for work-streams 1-3 is to develop a sensor system that enables production of landscape maps of SOC in real-time without the laborious process of returning soils to the lab for elemental analysis.
Implementation. This involves many different aspects to scale Quick Carbon and make our approach broadly accessible: field testing across different regions and agricultural practices, designing and testing new sampling tools, building spectral libraries, and implementing and highlighting valuable use cases. This fourth work-stream is focused on the application of technology and methods from work-streams 1-3. However, we believe that structures and tools for application must proceed concurrently with the fundamental research necessary to optimize sensors, visualize data, and design sampling schemes. New data collected in the field through pilot projects, or through collaborations with organizations that hold existing soil archives, will be critical in providing training information for each of the first three work-streams. Workstream 4 may include designing ‘kits’ for dispersed research projects, overseeing data collection and testing of field equipment, and working with partners to seek out soil archives across the country.
current vs future work flow
Application to other ecosystems (e.g. forests and other high SOC systems)
Increased capability at depth (e.g. 1 meter)
Also measure in-field other soil variables important to understanding SOC (e.g. pH, texture, SIC)
Provide additional info related to SOC (e.g. WHC, aggregation, etc)
Integration w/ process-based models
Broad adoption beyond just us