Documentation Index
Fetch the complete documentation index at: https://docs.mangrovesystems.com/llms.txt
Use this file to discover all available pages before exploring further.
What you’ll learn in this lesson:
- Understand uncertainty propagation and how it affects reported values
- Work with system ledgers for embodied emissions and renewable instruments
- Handle methodology versioning and updates in existing projects
- Design multi-project configurations for complex operations
Uncertainty propagation
Carbon accounting methodologies often require reporting uncertainty ranges alongside point estimates. Uncertainty propagates through calculations — if your input measurement has ±5% uncertainty, the output will also have uncertainty.How uncertainty compounds
When you multiply values with uncertainty, the relative uncertainties add (approximately):- Mass: ±5% uncertainty
- Carbon content: ±5% uncertainty
- Result: approximately ±7.1% uncertainty (√(5² + 5²))
Different methodologies have different rules for uncertainty propagation. Some use simple addition of relative uncertainties, others use root-sum-of-squares (RSS), and some specify fixed deductions. Always follow the methodology’s approach.
Implementing uncertainty in Mangrove
You can model uncertainty as:- Separate calculation branches — one for the point estimate, one for the upper bound, one for the lower bound
- Deduction factors — a methodology-specified discount (e.g., report the 95% lower confidence bound instead of the mean)
- Keisan expressions — for complex uncertainty formulas
Example: Uncertainty deduction as a model node
Example: Uncertainty deduction as a model node
If the methodology requires a conservative approach (report at 95% confidence):Where
uncertainty_deduction is a static input defined by the methodology (e.g., 0.05 for 5% deduction).System ledgers
Beyond production ledgers that track physical material, some projects need system ledgers for tracking non-physical quantities:- Embodied emissions ledger
- Renewable instruments ledger
Tracks the amortized carbon cost of facilities and equipment.
- Credits: Annual amortization amount (e.g., 50 tCO2e/year from facility construction)
- Debits: Monthly allocation to production batches
- Purpose: Ensures embodied emissions are spread evenly and don’t distort individual batch calculations
Methodology versioning
Methodologies evolve — emission factors get updated, calculation requirements change, new sources must be included. Your project needs to handle these transitions.Key challenges
| Challenge | Impact |
|---|---|
| Updated emission factors | Historic batches used old factors; new batches should use new ones |
| New calculation requirements | New model nodes needed without breaking existing batches |
| Changed reporting boundaries | More or fewer emission sources included |
Handling version transitions
Create a new model version
In the Model Editor, create a new version of the model with the updated calculations. Previous versions remain linked to historic batches — they aren’t affected by the change.
Update static inputs
Change emission factors and methodology parameters in the resource library. Set effective dates if the platform supports it, so historic and future calculations use the correct values.
Document the change
Record what changed, when, and why. This is critical for verification — auditors will want to see that the transition was handled correctly.
Multi-project configurations
Some operations span multiple projects — for example, a company with three biochar facilities, each configured as a separate Mangrove project.When to use multiple projects
- Different facilities with different operational parameters
- Different methodologies or registries
- Separate reporting requirements
- Independent mass balance tracking
Shared components
Even with separate projects, some components can be shared:| Component | Can Be Shared? | How |
|---|---|---|
| Emission factors | Yes | Same static inputs across projects |
| Event type definitions | No | Each project has its own event types |
| Model logic | Partially | Similar structure but separate instances |
| Feedstock types | Yes | Common feedstock definitions |
| Reporting templates | Yes | Standardized report formats |
Multi-project configurations are an advanced topic. For most implementations, start with a single project and only split into multiple projects when the operational differences genuinely require it. See the Guide to Production Accountingfor project configuration guidance.
Check your understanding
Why should you never modify a model version that's linked to generated batches?
Why should you never modify a model version that's linked to generated batches?
Modifying an existing model version would retroactively change historic batch outputs, breaking the audit trail. Verifiers expect that generated batches reflect the calculations as they were at the time of generation. Always create a new model version for methodology updates — this preserves historic batches while applying new logic to future ones.
What is a system ledger and when would you use one?
What is a system ledger and when would you use one?
A system ledger tracks non-physical quantities like embodied emissions or renewable energy certificates. You’d use one when: (1) the emission source isn’t tied to a specific production event (like annual facility construction costs amortized monthly), or (2) you need to track instruments like RECs to avoid double-counting when offsetting grid emissions across batches.
Next, prepare your project for production deployment in Lesson 6.3: Production Readiness.