How Carbonleap Works

A Predictive Operating Model for High Performance Farming

Carbonleap is built on a structured, scalable data model that ensures accuracy, consistency, and isolation for every farm and field in the system. This model is the foundation behind satellite analytics, weather-driven insights, crop modeling, and alert generation.

---

The Core Hierarchy

Carbonleap organizes information using a simple but powerful structure:

Account
 └── Farm
      └── Field(s)

Each layer is independent but linked, allowing flexible management of multiple farms and properties.

---

Account Level

An account represents the user’s identity in the system. It includes:

• Basic profile information

• Password and authentication details

• Access to all farms the user owns or has been invited to

• Personal settings and notification preferences

Anyone can create an account at no cost, and accounts can host any number of farms.

---

Farm Level

A farm is the primary operational container in Carbonleap. Each farm is treated as its own isolated data pod, meaning:

• Satellite imagery is processed per farm

• Weather data is selected per farm

• Models and alerts run independently

• Subscription plans apply at the farm level

This guarantees clean separation between different properties or business units.

Subscription and Acreage

A farm’s subscription is determined by total acreage, calculated from all its fields. Acreage is rounded to the nearest whole acre to simplify billing.

Because subscriptions are farm-specific, users may operate:

• One farm under a basic plan

• Another farm under a premium plan

• Or separate parts of the same physical property under different farms/plans

This provides granular flexibility.

---

Field Level

A field is the smallest analytical unit in Carbonleap. Each field includes:

• A mapped boundary

• Automatically calculated acreage

• Assigned crop and variety

• Satellite-derived vegetation metrics

• Weather inputs

• Growth stage calculations

• Active alert pipelines

Accurate field boundaries are essential, as analytics and models depend heavily on geometry.

---

Data Sources and Processing

Once a farm is activated, Carbonleap continuously processes three major inputs:

Satellite Imagery

The system ingests the latest available imagery and extracts vegetation indices:

• NDVI

• NDMI

• NDRE

• MSAVI

Each index is offered in multiple variants (actual, change, percentiles) to reveal different crop conditions and trends.

Hyperlocal Weather

Carbonleap aggregates data from multiple providers and automatically selects the most accurate weather source for each farm.

Weather powers:

• GDD tracking

• Phenological modeling

• Seven-day growth forecasting

• Condition-based alerts

Modeling Engine

Every farm runs its own instance of:

• Growth stage models

• Forecasted GDD models

• Water deficit analysis

• Operational alerts (e.g., spray windows)

All models update whenever new satellite or weather inputs arrive.

---

Change Orders

When users edit farm or field information, Carbonleap processes the update through a change order, ensuring data integrity.

During this process:

• Edited farms or fields become temporarily read-only

• The system recalculates models and acreage

• Subscription changes (if any) are previewed

• The farm returns to active status once updates finalize

This workflow ensures accurate and consistent data across the platform.

---

Why This Architecture Matters

Carbonleap’s hierarchy and processing model enable:

• Independent analytics per farm

• Scalable multi-farm operations

• Reliable, consistent satellite metrics

• Accurate weather-driven predictions

• Granular subscription flexibility

• Clean team and permission management

This structure ensures that every farm—regardless of size or location—receives precise, actionable intelligence throughout the growing season.

---