Module 2 Exercise: Build a Multi-Step Calculation Pipeline

What you’ll build in this exercise:

Add 2 new event types (Feedstock Receipt, Lab Analysis) with appropriate datapoints
Expand your model to a 5-step calculation pipeline with intermediate outputs
Test the pipeline with varied data and verify intermediate and final results

This exercise expands the single-formula model you built in Module 1 into a multi-step calculation pipeline that processes data from three event types. By the end, your model will calculate net carbon sequestered through 5 distinct intermediate steps. Prerequisites: Complete Lesson 2.1: Calculation Architecture and Lesson 2.2: Expanding Your Data Model. Ensure your Mangrove Biochar training project has the Biochar Delivery event type and simple model from Module 1.

Part 1: Design (30 min)

Before building in Mangrove, design the expanded model on paper or in a spreadsheet.

1. Identify additional event types

Your project needs data from two new operational activities:

Feedstock Receipt

What goes into the pyrolysis process.

Feedstock type (text)
Wet mass in tonnes (number)
Moisture content as % (number)
Receipt date (date)

Lab Analysis

Quality testing of produced biochar.

Sample ID (text)
Carbon content as % (number)
H:C molar ratio (number)
Analysis date (date)

2. Map the 5-step calculation flow

For each step, identify the inputs and the formula:

Step	Calculation	Inputs	Formula
1	Dry feedstock mass	wet_mass, moisture_content	`wet_mass × (1 - moisture/100)`
2	Feedstock carbon content	dry_mass, carbon_pct	`dry_mass × (carbon_pct / 100)`
3	Biochar yield	feedstock_carbon, biochar_mass	`biochar_mass / feedstock_carbon` (efficiency ratio)
4	Permanence factor	hc_ratio	If H:C < 0.4 → 0.9; < 0.7 → 0.8; else → 0.7
5	Net carbon sequestered	feedstock_carbon, permanence, factor_3.67	`feedstock_carbon × permanence × 3.67`

3. Identify validation rules

Think about what values should be flagged or rejected:

Moisture content must be between 0% and 100%
Carbon content must be between 0% and 100%
H:C ratio for biochar is typically 0.1–1.0
Wet mass must be positive

Part 2: Build in Mangrove (1.5 hrs)

Step 1: Add 2 new event types

Open your Mangrove Biochar project and go to Data Inputs > Input Settings.
Add a Feedstock Receipt event type (slug: feedstock-receipt) with 4 datapoints:

Datapoint	Slug	Value Type
Feedstock type	`feedstock-type`	text
Wet mass (tonnes)	`wet-mass-tonnes`	number
Moisture content (%)	`moisture-content-pct`	number
Receipt date	`receipt-date`	date

Add a Lab Analysis event type (slug: lab-analysis) with 4 datapoints:

Datapoint	Slug	Value Type
Sample ID	`sample-id`	text
Carbon content (%)	`carbon-content-pct`	number
H:C molar ratio	`hc-ratio`	number
Analysis date	`analysis-date`	date

See configuring event types for details on setting up datapoints.

Step 2: Expand your model with a 5-step pipeline

Open the Models section and the Model Editor .
Build the multi-step calculation tree:

Step 1: Dry feedstock mass

Operator: product Children:

Input node: wet-mass-tonnes (from Feedstock Receipt)
Calculated node: 1 - moisture/100 (use a Keisan expression or intermediate nodes)

Expected result: If wet mass = 12 tonnes and moisture = 15%, dry mass = 12 × 0.85 = 10.2 tonnes

Step 2: Feedstock carbon content

Operator: product Children:

Intermediate: dry feedstock mass (from Step 1)
Input node: carbon-content-pct (from Lab Analysis, divided by 100)

Expected result: If dry mass = 10.2 tonnes and carbon = 80%, carbon mass = 10.2 × 0.80 = 8.16 tonnes C

Step 3: Biochar yield efficiency

Operator: quotient Children:

Input node: biochar-mass-tonnes (from Biochar Delivery)
Intermediate: dry feedstock mass (from Step 1)

Expected result: If biochar mass = 8 tonnes and dry feedstock = 10.2 tonnes, yield = 8 / 10.2 = 0.784 (78.4% yield)

Step 4: Permanence factor

Operator: keisan Expression: Conditional based on H:C ratio from Lab Analysis

if hc_ratio < 0.4
  0.9
elsif hc_ratio < 0.7
  0.8
else
  0.7
end

Expected result: If H:C = 0.35, permanence factor = 0.9

Step 5: Net carbon sequestered (tCO2e)

Operator: product Children:

Intermediate: feedstock carbon content (from Step 2)
Intermediate: permanence factor (from Step 4)
Constant: 3.67

Output datapoint type: net-carbon-tco2eExpected result: 8.16 × 0.9 × 3.67 = 26.95 tCO2e

The biochar yield (Step 3) is calculated for informational purposes — it doesn’t feed directly into the net carbon calculation in this simplified model. You could use it in more advanced models to adjust carbon content based on conversion efficiency.

Step 3: Test with varied data

Create test events — add events for each of the 3 event types with the following test data:

Event Type	Key Values
Feedstock Receipt	wet mass = 12 t, moisture = 15%
Lab Analysis	carbon content = 80%, H:C ratio = 0.35
Biochar Delivery	mass = 8 t, carbon content = 80%

Run the model in the Console and verify each intermediate step:
- Dry mass: 12 × 0.85 = 10.2 t
- Carbon mass: 10.2 × 0.80 = 8.16 t C
- Yield: 8 / 10.2 = 0.784
- Permanence: 0.9 (H:C < 0.4)
- Net carbon: 8.16 × 0.9 × 3.67 ≈ 26.95 tCO2e
Test with edge cases — create additional events with:
- High moisture (50%) — dry mass should drop significantly
- High H:C ratio (0.75) — permanence should be 0.7 (lowest tier)
- Different carbon content — verify proportional impact on output

If intermediate values don’t match expectations, check: (1) datapoint slugs match between event types and model nodes, (2) percentage values are divided by 100 where needed, (3) the Keisan expression handles all H:C ranges correctly.

Success criteria

You have completed the Module 2 exercise when:

Two new event types exist — Feedstock Receipt and Lab Analysis — each with appropriate datapoints in Data Inputs > Input Settings
Model expanded to a 5-step pipeline with visible intermediate calculations
Intermediate outputs are verifiable — dry mass, carbon content, yield, permanence factor all produce expected values
Final output (net carbon in tCO2e) matches manual calculation for your test data
Edge cases tested — varied moisture, H:C ratios, and carbon content produce correct adjusted results

What’s next

In Module 3 you’ll learn about batch partitioning — how to split your operational data into discrete, reportable batches, and add LCA emissions calculations to compute net carbon removal. Continue to Module 3: Batch Partitioning & LCA.

Overview

Production Accounting

Data Collection

Data Rules

Reporting

Inventory

Account Admin

Portfolio

Module 2 Exercise: Build a Multi-Step Calculation Pipeline

Accounting Academy

Part 1: Design (30 min)

1. Identify additional event types

Feedstock Receipt

Lab Analysis

2. Map the 5-step calculation flow

3. Identify validation rules

Part 2: Build in Mangrove (1.5 hrs)

Step 1: Add 2 new event types

Step 2: Expand your model with a 5-step pipeline

Step 3: Test with varied data

Success criteria

What’s next

Accounting Academy

Overview

Production Accounting

Data Collection

Data Rules

Reporting

Inventory

Account Admin

Portfolio

Accounting Academy

Documentation Index

​Part 1: Design (30 min)

​1. Identify additional event types

Feedstock Receipt

Lab Analysis

​2. Map the 5-step calculation flow

​3. Identify validation rules

​Part 2: Build in Mangrove (1.5 hrs)

​Step 1: Add 2 new event types

​Step 2: Expand your model with a 5-step pipeline

​Step 3: Test with varied data

​Success criteria

​What’s next

Accounting Academy

Part 1: Design (30 min)

1. Identify additional event types

2. Map the 5-step calculation flow

3. Identify validation rules

Part 2: Build in Mangrove (1.5 hrs)

Step 1: Add 2 new event types

Step 2: Expand your model with a 5-step pipeline

Step 3: Test with varied data

Success criteria

What’s next