Rainforest GFM

The Rainforest Gap Filling Module calculates the deforestation risk associated with food products by tracing the amount of critical commodities (soy and palm oil) from problematic sources. This includes not only ingredients that go directly into the end product but also soy and palm oil used in animal feed for livestock products.

Quick Reference

Property	Description
Runs on	ModeledActivityNode with FoodProductFlowNode parent containing a product name
Dependencies	OriginGapFillingWorker
Key Input	Product name, origin country, certification status, product category
Output	rainforest_critical_products property with critical commodity quantities
Trigger	Product matched to critical commodity content data

When It Runs

The module triggers when:

1. A ModeledActivityNode has a parent FoodProductFlowNode with a valid product_name
2. The OriginGapFillingWorker has completed (to ensure origin information is available)
3. The product is linked to critical commodity content data in the glossary

Key Output

The module adds a rainforest_critical_products property to the flow node containing:

• Quantity of critical commodities (grams per 100g of product)
• Certification status: Rainforest conservation certified, certified but not specified for rainforest, or not certified

---

Scientific Methodology

The deforestation calculation traces the amount of soy and palm oil from problematic sources that was required in the production of the food product. The methodology accounts for:

1. Direct ingredients: Soy and palm oil used directly in the product
2. Indirect ingredients: Soy and palm oil used in animal feed for livestock products
3. Origin risk: Whether the product originates from high-deforestation-risk countries
4. Certification: Whether the product has certifications that protect against deforestation

Critical Amount Calculation

The critical amount is calculated based on:

Critical Amount = Soy Content + Palm Oil Content

Where the soy content contribution depends on:

• Product origin country
• Product category (fish, beef/pork/chicken/egg/milk, or other)
• Location data source (FAO statistics versus specified origin)

Origin-Based Risk Assessment

The module applies different rules based on product origin:

For products with unknown origin:

• Full soy content is included (assumes worst-case sourcing)
• Full palm oil content is included

For fish products:

• Full soy content is included (regardless of origin)
• Full palm oil content is included

For beef, pork, chicken, egg, and milk products:

• Soy content included unless origin is Switzerland (CH)
• Full palm oil content is included

For other products:

• Soy content included only if origin is Brazil (BR), Argentina (AR), or China (CN)
• Full palm oil content is included

---

Implementation Details

Covered Commodities

The module tracks two primary deforestation-linked commodities:

Commodity	Risk Factor	Primary Source Countries
Soy	Animal feed, direct ingredient	Brazil, Argentina, Paraguay, China
Palm Oil	Food processing, cosmetics	Indonesia, Malaysia

Critical Product Categories

Soy-Critical Products

All meat products typically fed with concentrated feed are assigned soy content based on:

• Inventory data from Life Cycle Assessment databases
• Allocation to different animal products (meat types, dairy products)
• Plant-based products containing soy directly

Palm Oil-Critical Products

Products likely to contain palm oil with assigned amounts:

Product Category	Palm Oil Content
Bread and pastries	10%
Chicken/meat broth	25%
Beef broth	15%
Margarine	26%
Plant-based fats/oils	33%
Chocolate sweets (pralines)	variable

Note: Soy or palm oil in amounts less than 1% by weight are ignored by the indicator.

Certification Categories

The module recognizes three certification states:

Term XID	Category	Description
EOS_rainforest_conservation_certified	Certified for rainforest	Product has certification specifically protecting rainforest
EOS_certified_rainforest_not_specified	Certified, not rainforest specific	Product is certified but not specifically for rainforest conservation
EOS_not_certified_for_rainforest	Not certified	No relevant certification

High-Risk Countries

Soy Production Risk Countries

Country Code	Country	Risk Level
BR	Brazil	High
AR	Argentina	High
CN	China	Moderate

Additional countries to consider (noted in source): Paraguay, Russia, Bolivia

Palm Oil Production Risk Countries

Primary risk from Indonesia and Malaysia (implicit in palm oil content assignment).

Product Category Detection

The module detects specific product categories using FoodEx2 terms:

Term XID	Category
EOS_Fish-and-seafood	Fish products
EOS_Beef-and-veal	Beef products
EOS_Poultry	Chicken products
EOS_Eggs	Egg products
EOS_Dairy	Dairy products

---

Recognized Certifications

Labels Protecting Against Deforestation

Based on labelinfo.ch (biodiversity indicator):

• Bio Knospe / Bio Suisse Knospe
• Coop Naturaplan / Coop Naturafarm
• Migros Bio / BIO natura / Bio Natur Plus
• Demeter
• KAGfreiland
• IP-Suisse
• Rainforest Alliance
• Fairtrade Max Havelaar
• EU-Bio / Bio-Siegel / Agriculture Biologique

Commodity-Specific Certifications

Soy: RTRS (Round Table on Responsible Soy) certified companies

Palm Oil: RSPO (Roundtable on Sustainable Palm Oil) member companies meeting WWF scorecard standards

Cocoa: Fairtrade, UTZ, Rainforest Alliance

Coffee: 4C Association, Fairtrade, Rainforest Alliance, UTZ, Organic

---

Full Code Reference

Critical Amount Calculation

The core logic for determining critical commodity amounts:

if (not flow_country_codes) or self.is_fish(amount_per_category_in_flow):
    if not flow_country_codes:
        warning_msg = (
            f"No location found for product {parent_flow_product_name_terms}."
            f"Assuming that the critical product comes from a problematic country."
        )
        logger.warn(warning_msg)
        calc_graph.append_data_errors_log_entry(warning_msg)
    critical_amount_per_100g = soy_content_per_100g + palm_oil_content_per_100g
else:
    critical_amount_per_100g = palm_oil_content_per_100g
    is_beef_pork_chicken_egg_milk = self.is_beef_pork_chicken_egg_milk(amount_per_category_in_flow)
    for flow_country_code, source in zip(flow_country_codes, sources):
        if flow_country_code and len(flow_country_code) == 3:
            flow_country_code = iso_3166_map_3_to_2_letter(flow_country_code)

        if not flow_country_code or not source:
            critical_amount_per_100g += soy_content_per_100g / len(flow_country_codes)
        elif source and source == LocationSourceEnum.fao_stat:
            critical_amount_per_100g += soy_content_per_100g / len(flow_country_codes)
        else:
            if is_beef_pork_chicken_egg_milk:
                if not flow_country_code == "CH":
                    critical_amount_per_100g += soy_content_per_100g / len(flow_country_codes)
            else:
                if flow_country_code in ("BR", "AR", "CN"):
                    critical_amount_per_100g += soy_content_per_100g / len(flow_country_codes)

Certification Detection

(
    matched_certified_term_xids,
    certified_term_containing_node,
) = parent_flow.get_prop_by_inheritance(
    "tag_term_xids",
    ignore_if=(
        lambda prop_name, tag_term_xids_to_search=self.gfm_factory.certification_term_xids:
            not has_one_of_predefined_tags(prop_name, tag_term_xids_to_search)
    ),
)

if (
    matched_certified_term_xids
    and certified_term_containing_node.product_name == parent_flow.product_name
):
    for label_xid in matched_certified_term_xids:
        if label_xid == "EOS_rainforest_conservation_certified":
            certified = True
            rainforest_specified = True
        if label_xid == "EOS_certified_rainforest_not_specified":
            certified = True

Output Property Creation

qty_prop = ReferencelessQuantityProp.unvalidated_construct(
    value=critical_amount_per_100g,
    unit_term_uid=self.gfm_factory.gram_term.uid,
)

if certified and rainforest_specified:
    quantities = {self.gfm_factory.rainforest_conservation_certified.uid: qty_prop}
elif certified:
    quantities = {self.gfm_factory.certified_rainforest_not_specified.uid: qty_prop}
else:
    quantities = {self.gfm_factory.not_certified_for_rainforest.uid: qty_prop}

rainforest_critical_products = RainforestCriticalProductsProp.unvalidated_construct(
    quantities=quantities,
    for_reference=ReferenceAmountEnum.amount_for_100g
)

---

Calculation Example

Scenario: 100g of chicken breast from Germany

Step 1: Identify Product Category

• Product matches EOS_Poultry category
• Classified as beef/pork/chicken/egg/milk product

Step 2: Look Up Critical Content

• Soy content from feed: 15g per 100g (from glossary link)
• Palm oil content: 0g (not typically in chicken)

Step 3: Apply Origin Rules

• Origin: Germany (DE)
• Not Switzerland, so soy content is included
• Palm oil always included

Step 4: Calculate Critical Amount

Critical amount = 15g (soy) + 0g (palm oil) = 15g per 100g

Step 5: Determine Certification Status

• No certification tags found
• Status: EOS_not_certified_for_rainforest

Final Output

The module adds to the flow node:

{
  "rainforest_critical_products": {
    "quantities": {
      "[not_certified_for_rainforest_uid]": {
        "value": 15.0,
        "unit_term_uid": "[gram_uid]"
      }
    },
    "for_reference": "amount_for_100g"
  }
}

---

Regulatory Context

European Union Deforestation Regulation

The EU Deforestation Regulation (EUDR) entered into force on June 29, 2023, with application starting December 30, 2024 (June 30, 2025 for small enterprises).

Covered Commodities:

• Palm oil
• Soy
• Coffee
• Cocoa
• Cattle (beef)
• Timber
• Rubber

The regulation requires that products placed on the EU market must be:

1. Deforestation-free (not produced on land deforested after December 31, 2020)
2. Produced in accordance with relevant legislation of the country of production
3. Covered by a due diligence statement

---

Known Limitations

Commodity Coverage

• Currently focuses primarily on soy and palm oil
• Cocoa and coffee not yet fully implemented despite regulatory relevance
• Timber and rubber not covered

Geographic Coverage

• Limited set of high-risk countries (Brazil, Argentina, China for soy)
• Paraguay, Russia, Bolivia noted as candidates for inclusion
• Sub-national risk differentiation not implemented

Certification Recognition

• System currently recognizes "certified" versus "not certified"
• Individual label recognition not fully implemented
• No differentiation between certification standards

Data Granularity

• Palm oil content assignments are category-based estimates
• Soy content in animal feed based on average inventory data
• No product-specific formulation data

Supply Chain Tracing

• Assumes worst-case sourcing when origin unknown
• FAO statistics-based origins treated as unknown
• No multi-tier supply chain visibility

---

References

1. World Resource Institute (2018). Global Forest Review - Deforestation linked to agriculture. https://research.wri.org/gfr/forest-extent-indicators/deforestation-agriculture

2. WWF (2020). Risky Business: The EU's dependence on imported biodiversity. https://forestsolutions.panda.org/uploads/default/report/WWF_Risky_business_eng.pdf

3. WWF (2021). Soy Scorecard - Assessing soy trader performance on deforestation-free production. https://wwfeu.awsassets.panda.org/downloads/wwf_soy_report_summary.pdf

4. WWF (2021). Palm Oil Buyers Scorecard. https://palmoilscorecard.panda.org/

5. WWF (2023). Chocolate Scorecard. https://www.wwf.de/fileadmin/fm-wwf/Publikationen-PDF/Landwirtschaft/chocolate-scorecard-2023.pdf

6. labelinfo.ch. Swiss label information platform - Biodiversity indicator. https://www.labelinfo.ch/

7. RTRS (2023). Chain of Custody certified companies. https://responsiblesoy.org/

8. European Commission (2023). Regulation (EU) 2023/1115 on deforestation-free products.