Data Sources and Methodology for MVP

[tylerhuntington]

1. Data Sources

1.1 LandIQ Crop Mapping Data

• Description: High-resolution, field-level crop mapping dataset for California, produced by LandIQ for specific years.
• Content: Polygon geometries representing agricultural fields, with associated crop type classification.
• Use in BioSiting: Serves as the spatial framework for estimating agricultural residue availability. Each polygon represents a discrete agricultural field with a known crop type for the year the dataset was produced.

1.2 USDA National Agricultural Statistics Service (NASS) – Census of Agriculture

• Description: Comprehensive, nationwide agricultural dataset collected every five years by USDA NASS, reporting acreage, yields, and production statistics for major crops at county and state levels.
• Content: Crop-specific yield data (e.g., tons per acre) aggregated at the county level.
• Use in BioSiting: Provides baseline yield values for primary crops, which are then converted into residue estimates using feedstock-specific residue factors.

1.3 BioCirv Aim 1/2 Feedstock Composition Data

• Description: Dataset containing compositional attributes for various biomass feedstocks, compiled as part of the BioCirv project under Aims 1 and 2. Includes lignin content, cellulose, hemicellulose, ash content, moisture content, and other relevant biochemical properties.
• Content: Feedstock-specific average composition values derived from experimental measurements and literature sources.
• Use in BioSiting: Enables estimation of not only the quantity of available feedstock but also its quality and suitability for various bioenergy or bioproduct pathways.

---

2. Draft Methodology for Data Integration

1. Acquire LandIQ Data

   • Obtain the most recent available LandIQ crop mapping dataset for California.
   • Ensure each polygon has an associated crop type code.

2. Acquire USDA Census of Agriculture Data

   • Download the most recent Census of Agriculture data for California counties.
   • Extract county-level yield data for crops of interest.

3. Apply Residue Factors

   • For each crop type, assign a residue factor (ratio of residue mass to primary crop yield) based on literature or industry references.
     • Example: Wheat – 1.3 tons residue per ton grain yield; Corn – 1.0 ton residue per ton grain yield.
   • Multiply county-level yield values by the residue factor to estimate residue yield per acre for each crop.

4. Merge Yield and Residue Data with LandIQ Polygons

   • Join county-level residue yield values to LandIQ polygons based on county location and crop type.
   • For each polygon:
     • Determine polygon area (acres).
     • Multiply area by residue yield per acre to compute total residue mass for that field.

5. Merge Feedstock Composition Attributes (BioCirv Aim 1/2 Data)

   • Match each crop type in the LandIQ–USDA merged dataset to its corresponding entry in the BioCirv Aim 1/2 composition dataset.
   • Append compositional fields such as:
     • Lignin content (% dry weight)
     • Cellulose content (% dry weight)
     • Hemicellulose content (% dry weight)
     • Ash content
     • Moisture content
     • Other relevant biochemical or physical properties
   • This step enables downstream analyses that consider both quantity and quality of biomass for specific conversion technologies.

6. Generate Final Feedstock Dataset

   • Output dataset includes:
     • Field polygon geometry
     • Crop type
     • Estimated residue yield per acre
     • Total estimated residue mass
     • Compositional attributes from BioCirv Aim 1/2
   • Dataset can be aggregated spatially (e.g., by county, watershed, or custom region) for BioSiting analyses.

[atambay37]

Thank you! @tylerhuntington ! This looks great. Looking forward to discussing in today's meeting. My first comment is that we can constrain data 1.1 and 1.2 to the North San Joaquin River Valley.

[atambay37]

@petercarbsmith and co discussion: let's aim to support temporal data eventually: harvest dates, etc.

[lsetiawan]

Req based on meeting conversation:

• Need to be able to get all set of data based on the geospatial data from LandIQ
• Temporal data needed for various time ... e.g harvest date, processing data date (Future, out of scope)
  • Current national tool doesn't support temporal dataset

[atambay37]

@tylerhuntington if you could describe the existing infrastructure data from the national tool, that would be great.

[atambay37]

@tylerhuntington @lsetiawan we will bring with us to this new database and resulting API all applicable CA bounded data from the national tool.

[lsetiawan]

Found this interesting open data for pupulation density stuff: https://hub.worldpop.org/

[tylerhuntington]

I've pulled together an initial set of the infrastructure data that we want to include in the tool and stored them in this Google Drive folder. They can also be directly downloaded from the Data Downloads page of the National Biositing Tool, which provides references for each one. We may want to add more infrastructure data down the line. Having good pipelines in place for handling these types of datasets would be valuable. What I've put together so far is a starting point, but I imagine the need for adding additional datasets as they become available in the future.