Services

Krotovina Data Insights provides quantitative soil, carbon, and environmental analysis for organizations working in carbon markets, climate risk, and agricultural sustainability. Every project is built on reproducible, defensible methods and transparent treatment of uncertainty.

Predictive Soil Mapping

Spatial prediction of soil properties: organic carbon, texture, pH and salinity across fields, regions, or entire landscapes. By combining field observations with terrain, climate, and remote sensing covariates in machine-learning frameworks, I produce continuous maps with explicit uncertainty at the resolution your project needs. Well suited to establishing soil carbon baselines, delineating management zones, and extending sparse sampling to full spatial coverage.

Remote Sensing Data Analysis

Analysis of satellite and airborne imagery at scale using Google Earth Engine and open geospatial tooling. Applications include vegetation productivity and biomass estimation, land-cover and land-use change detection, and construction of long time series from the MODIS, Landsat, and Sentinel archives. This turns vast imagery collections into the specific indicators a project needs such as productivity trends, conversion events, and calibrated model inputs.

Biogeochemical Process Modelling

Simulation of soil organic carbon dynamics and greenhouse gas fluxes using process-based models (DayCent/CENTURY) calibrated to local conditions. Bayesian calibration quantifies parameter uncertainty and propagates it through to predictions, so scenario projections come with credible bounds rather than single-point estimates. Supports additionality assessment, management-scenario comparison, and forward-looking projections of carbon outcomes.

Carbon Intensity and MMRV

Measurement, monitoring, reporting, and verification support for carbon and biofuel projects. I build quantification workflows that align with recognized protocols, combine measured and modeled data defensibly, and carry uncertainty through to reported values producing credit-grade, audit-ready estimates. Applicable to soil carbon crediting, avoided-conversion accounting, and carbon-intensity scoring for agricultural systems.

Agricultural and Environmental Scientific Data Analysis

End-to-end analysis of complex agricultural and environmental datasets, from experimental design through to publication or regulator-ready results. Services span statistical modeling (mixed models, Bayesian inference, spatial statistics), reproducible R and Python pipelines, and processing of large datasets on high-performance computing infrastructure. For teams that have the data but need rigorous, transparent analysis they can stand behind.