Frequently Asked Questions - Kumi Analytics
Learn more about our digital MRV, blue carbon monitoring, and satellite analytics solutions.
Digital MRV (dMRV) and Carbon Markets
What is the difference between dMRV and traditional MRV?
Traditional Monitoring, Reporting, and Verification (MRV) relies heavily on manual field surveys, which are time-consuming, expensive, and difficult to scale. Digital MRV (dMRV) automates this process using satellite imagery, remote sensing, and artificial intelligence. dMRV provides continuous, objective, and verifiable data on carbon stocks and ecosystem health, significantly reducing the time and cost required to issue carbon credits while improving transparency for buyers and registries.
How does satellite-based dMRV improve the accuracy of forest carbon credits?
Satellite-based dMRV improves accuracy by providing wall-to-wall coverage of a project area, eliminating the sampling errors inherent in manual field plots. By combining high-resolution optical imagery with Synthetic Aperture Radar (SAR) and LiDAR data, Kumi Analytics can measure forest structure, canopy height, and above-ground biomass with high precision. This continuous monitoring also detects disturbances like illegal logging or fires in near real-time, ensuring the permanence of the carbon credits.
Can Kumi Analytics' dMRV solutions be used for Verra and Gold Standard compliance?
Yes, Kumi Analytics designs its dMRV solutions to align with the rigorous requirements of major carbon registries, including Verra (VCS) and Gold Standard. Our platform provides the transparent, auditable, and scientifically validated data necessary to support project design documents (PDDs), establish dynamic baselines, and verify ongoing carbon sequestration for credit issuance.
Blue Carbon and Mangrove Intelligence
How do you measure mangrove above-ground biomass with satellites?
Measuring mangrove above-ground biomass from space requires a multi-sensor approach due to the complex structure of coastal wetlands. Kumi Analytics utilizes a combination of Synthetic Aperture Radar (SAR) backscatter data, which penetrates the forest canopy to measure structural volume, and high-resolution optical multispectral imagery. These data streams are processed through proprietary machine learning models trained on extensive field data to generate accurate, high-resolution biomass maps of mangrove ecosystems.
Why is remote sensing critical for blue carbon projects?
Blue carbon ecosystems, particularly mangroves, are often located in remote, difficult-to-access intertidal zones where traditional field surveys are dangerous and logistically challenging. Remote sensing provides a safe, cost-effective way to monitor these areas continuously. It allows project developers to accurately map historical mangrove extent, track restoration progress, and quantify carbon sequestration across vast coastal regions that would otherwise be impossible to measure manually.
Can satellite imagery detect changes in mangrove health before they die?
Yes, multispectral satellite imagery can detect subtle changes in mangrove health long before they are visible to the naked eye. By analyzing specific vegetation indices, such as the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI), Kumi Analytics can identify areas experiencing stress from changes in salinity, hydrology, or pollution, allowing project managers to intervene early and protect the ecosystem.
Forestry Carbon and Biomass Mapping
What satellite sensors do you use for forest biomass estimation?
Kumi Analytics employs a multi-sensor fusion approach for forest biomass estimation. We primarily use Synthetic Aperture Radar (SAR) data from Sentinel-1 and ALOS PALSAR, which can penetrate the forest canopy to measure structural volume. We combine this with high-resolution optical imagery from Worldview, Planet, Sentinel-2 and Landsat, and calibrate our models using LiDAR data or field measurements. This combination allows us to accurately estimate above-ground biomass across diverse forest types.
How do you establish historical baselines for REDD+ projects?
We establish historical baselines for REDD+ projects by analyzing decades of archived satellite imagery, primarily from the Landsat program. By applying machine learning algorithms to this historical data, we can accurately map past deforestation and forest degradation rates within the project area and a reference region. This data-driven approach provides a robust, verifiable baseline against which future emission reductions can be measured and credited.
Can your platform monitor selective logging and forest degradation?
Yes, monitoring forest degradation, such as selective logging, is a key capability of our platform. While clear-cut deforestation is easily visible in optical imagery, degradation requires more advanced techniques. We use high-resolution optical data combined with SAR time-series analysis to detect subtle changes in canopy structure and gaps, allowing us to quantify carbon losses from degradation that traditional methods often miss.
How are allometric models used in calculating biomass?
Allometric models are mathematical equations that relate easily measurable tree characteristics (like trunk diameter at breast height, or DBH, and tree height) to the total above-ground biomass (AGB) of the tree. In remote sensing, we use field-collected DBH and height data to calculate the actual biomass of sample plots using these allometric equations. We then train our machine learning algorithms to correlate the satellite data (like SAR backscatter and optical indices) with these field-calculated biomass values, allowing us to scale the biomass estimation across the entire project area.
How do different SAR wavelengths (L-band, C-band, X-band) help in understanding forest structure?
Different Synthetic Aperture Radar (SAR) wavelengths penetrate forest canopies to different depths, providing a 3D understanding of forest structure. X-band (shortest wavelength) reflects off the top of the canopy, providing information on leaves and small branches. C-band (medium wavelength, e.g., Sentinel-1) penetrates the upper canopy and reflects off secondary branches. L-band (longest wavelength, e.g., ALOS PALSAR) penetrates deep into the canopy, reflecting off the main trunks and large branches. By combining these, especially L-band, we can accurately estimate the structural volume and biomass of dense forests.
How is Soil Organic Carbon (SOC) calculated in nature-based carbon projects?
Soil Organic Carbon (SOC) represents the carbon stored in the soil, which is often a significant portion of the total carbon pool, especially in blue carbon ecosystems like mangroves and peatlands. Calculating SOC typically involves taking physical soil core samples at various depths across the project area and analyzing them in a laboratory to determine the carbon concentration and bulk density. While satellites cannot directly measure SOC deep underground, Kumi Analytics uses remote sensing to map above-ground vegetation types, soil moisture, and land cover changes, which serve as strong proxies to model and stratify SOC distribution across large landscapes.
Water Quality Monitoring from Space
How does hyperspectral satellite imagery improve coastal water quality analysis?
Hyperspectral satellite imagery captures data across hundreds of narrow, contiguous spectral bands, whereas traditional multispectral satellites only capture a few broad bands. This detailed spectral information allows Kumi Analytics to identify the specific optical signatures of different water constituents. We can accurately differentiate and quantify chlorophyll-a (indicating phytoplankton), suspended sediments, and colored dissolved organic matter (CDOM) in complex coastal and inland waters.
Can satellites detect harmful algal blooms (HABs)?
Yes, satellites are highly effective at detecting and monitoring harmful algal blooms (HABs). By analyzing the spectral reflectance of the water surface, Kumi Analytics can identify the specific pigments associated with different types of algae, such as cyanobacteria. Continuous satellite monitoring allows us to track the spatial extent, movement, and intensity of HABs, providing early warning systems for aquaculture operations, desalination plants, and coastal authorities.
What water quality parameters can be measured using remote sensing?
Remote sensing can accurately measure several optically active water quality parameters. These include Chlorophyll-a concentration (a proxy for algal biomass), Total Suspended Matter (TSM) or turbidity, Colored Dissolved Organic Matter (CDOM), and Sea Surface Temperature (SST). While satellites cannot directly measure non-optical parameters like dissolved oxygen or heavy metals, the optical parameters often serve as strong indicators of overall ecosystem health and pollution events.
KACSAT Platform and Methodology
What is the KACSAT platform?
KACSAT is Kumi Analytics' proprietary geospatial intelligence platform. It ingests massive volumes of multi-sensor satellite data (optical, SAR, LiDAR) and processes it using advanced machine learning algorithms to deliver actionable insights for carbon monitoring, environmental assessment, and climate risk finance. The platform is designed to provide transparent, verifiable, and scalable data to support Nature-Based Solutions (NbS).
How do you validate the accuracy of your satellite-derived models?
We validate our satellite-derived models through rigorous ground-truthing. We partner with local organizations, academic institutions, and project developers to collect high-quality field data, such as forest inventory plots, soil samples, and water quality measurements. This field data is used to train and calibrate our machine learning algorithms, ensuring that our remote sensing estimates accurately reflect on-the-ground realities. We transparently report the accuracy and uncertainty metrics of our models.
How frequently is the satellite data updated?
The update frequency depends on the specific satellite sensors used and the requirements of the project. For continuous monitoring of deforestation or algal blooms, we can provide near real-time updates using high-revisit satellites like Sentinel-1 and Sentinel-2, which capture data every few days. For more complex analyses, such as annual biomass accumulation, we typically provide quarterly or annual updates, synthesizing data over a longer period to account for seasonal variations and cloud cover.