Skills
skills/jpfielding/claude.pnge/epa-ghg

epa-ghg

SKILL.md

EPA Greenhouse Gas Reporting Program (GHGRP) Skill

Fetches and analyzes facility-level greenhouse gas emissions from the EPA GHGRP via the Envirofacts REST API (data.epa.gov/efservice/).

Credential

None required. The Envirofacts GHGRP views are fully public with no API key or authentication.

API Structure

Base URL: https://data.epa.gov/efservice/

Query pattern: Path-based filters chained in the URL:

https://data.epa.gov/efservice/{TABLE}/{COLUMN}/{VALUE}/.../{rows/M:N}/{FORMAT}

Primary table: V_GHG_EMITTER_SUBPART — facility emissions by subpart, gas, and year. This is the most detailed table available via the API.

Secondary table: V_GHG_EMITTER_GAS — facility emissions by gas only (no subpart breakdown). Useful for total facility emissions.

Key parameters:

Component Description Example
TABLE View name V_GHG_EMITTER_SUBPART
COLUMN/VALUE Exact match filter STATE/WV, YEAR/2022
COLUMN/CONTAINING/VALUE Substring match FACILITY_NAME/CONTAINING/EQT
rows/M:N Row range (0-indexed, inclusive) rows/0:99
count/JSON Record count only Appended instead of rows
Format suffix JSON, CSV, or XML Appended last

Multiple filters chain left to right:

/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/STATE/WV/YEAR/2022/rows/0:99/JSON

See references/api_reference.md for full column reference, all subpart codes, gas codes, and additional query patterns.

Key Subparts for PNGE

Subpart Category Use Case
W Petroleum and Natural Gas Systems Onshore/offshore production, processing, transmission, storage, LNG, distribution. Primary for O&G.
C Stationary Combustion Boilers, turbines, engines at any large facility.
Y Petroleum Refining Process emissions from catalytic cracking, coking, etc.
UU Injection of CO2 EOR and carbon storage injection quantities.
PP Suppliers of CO2 CO2 capture and supply facilities.
MM Suppliers of Petroleum Products Upstream supply-side accounting for refiners/importers.
RR Geologic Sequestration of CO2 Facilities with EPA-approved CO2 sequestration monitoring.

Workflow

Step 1 — Resolve Intent

Map the user's question to query parameters:

User asks about Subpart Gas filter Notes
O&G facility emissions W (all) Default for most PNGE queries
Methane leaks, venting W CH4 Methane is the key Subpart W concern
Refinery emissions Y (all) Texas, Louisiana dominate
CO2 injection / EOR UU (all) Supplier-type, reports injection volumes
Carbon storage RR (all) Sequestration monitoring
Total facility emissions (all or omit) (all) Use V_GHG_EMITTER_GAS for aggregate
Specific operator use CONTAINING filter (all) e.g., FACILITY_NAME/CONTAINING/EQT

State mapping: Use 2-letter postal codes (WV, TX, PA, OH, ND, NM, etc.)

Year range: Data available 2010-present. Most recent complete year is typically current year minus 2 (data for 2023 published ~October 2024).

Step 2 — Get Record Count (when total is uncertain)

Before fetching potentially large result sets, check the count:

curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/YEAR/2022/count/JSON"

Returns: [{"TOTALQUERYRESULTS": 5782}]

If count exceeds 1000, warn the user and paginate or narrow filters.

Step 3 — Fetch Data

Build the URL with appropriate filters. Default behavior:

  • Use V_GHG_EMITTER_SUBPART as the primary table
  • Filter by SUBPART_NAME/W for O&G queries unless user specifies otherwise
  • Include YEAR/{latest} — default to most recent available year
  • Add STATE/{code} when user specifies geography
  • Limit with rows/0:99 for initial fetch; expand if needed
  • Use JSON output format

Example — Subpart W facilities in West Virginia for 2022:

curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/STATE/WV/YEAR/2022/JSON" \
  | python3 -m json.tool

Example — Search by operator name:

curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/FACILITY_NAME/CONTAINING/EQT/SUBPART_NAME/W/YEAR/2022/JSON"

Example — Methane-only emissions in a state:

curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/STATE/PA/YEAR/2022/GAS_CODE/CH4/rows/0:99/JSON"

Example — Top emitters nationally (first page):

curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/YEAR/2022/rows/0:99/JSON"

Step 4 — Parse Response

Response is a flat JSON array. Each element is one row:

[
  {
    "facility_id": 1003461,
    "facility_name": "Dominion Energy Transmission - Cornwell Station",
    "state": "WV",
    "county": "KANAWHA COUNTY",
    "latitude": 38.48558,
    "longitude": -81.27167,
    "year": 2022,
    "subpart_name": "W",
    "subpart_category": "Petroleum and Natural Gas Systems",
    "co2e_emission": 4.6,
    "ghg_quantity": 4.6,
    "gas_code": "CO2",
    "gas_name": "Carbon Dioxide"
  }
]

Key fields:

  • co2e_emission — CO2-equivalent metric tons (GWP-weighted). Use this for comparisons and totals.
  • ghg_quantity — Raw mass in metric tons of the specific gas. Use this when analyzing gas composition (e.g., "how many tons of methane leaked").
  • Each facility appears as multiple rows (one per gas type per subpart). Aggregate by facility_id or facility_name for total facility emissions.

Aggregation logic (in bash with jq):

# Total CO2e by facility
curl -s "URL" | jq '[group_by(.facility_name)[] |
  {facility: .[0].facility_name, state: .[0].state,
   total_co2e: ([.[].co2e_emission] | add)}] |
  sort_by(-.total_co2e) | .[:10]'

Step 5 — Produce Output

Format: Ranked Facility Table + Gas Breakdown + Narrative

Present a markdown table of the top facilities ranked by CO2e, then a gas breakdown, then narrative analysis.

Example output structure:

## Subpart W Emissions — West Virginia (2022)

### Top Facilities by Total CO2e

| Rank | Facility | County | Total CO2e (MT) |
|------|----------|--------|-----------------|
| 1 | DTM Appalachia Gathering, LLC | Doddridge | 183,921.6 |
| 2 | HG Energy II Appalachia, LLC | Ritchie | 172,815.1 |
| 3 | Mountaineer Gas Company | Kanawha | 127,178.6 |
| ... | ... | ... | ... |

### Gas Breakdown (all WV Subpart W facilities combined)

| Gas | CO2e (MT) | % of Total |
|-----|-----------|------------|
| Carbon Dioxide | 601,604.8 | 53.7% |
| Methane | 518,635.5 | 46.3% |
| Nitrous Oxide | 593.0 | <0.1% |

**Summary:** West Virginia Subpart W (Petroleum and Natural Gas Systems)
facilities reported 1,120,833 MT CO2e in 2022 across 43 reporting facilities.
Methane accounts for 46% of CO2e despite being a small fraction by mass —
this reflects methane's 25x global warming potential. The top 3 facilities
(DTM Appalachia, HG Energy II, Mountaineer Gas) account for 43% of the
state's total Subpart W emissions. Gathering and processing operations
dominate, consistent with WV's role as a major Marcellus/Utica gas
production state.

**Units:** Metric tons CO2-equivalent (MT CO2e). Reporting threshold:
25,000 MT CO2e/year (facility-wide, not per-subpart).
**Data vintage:** 2022 final. Most recent available as of early 2025.

Pagination

Use rows/M:N for pagination (0-indexed, inclusive on both ends).

# Get count first
COUNT=$(curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/YEAR/2022/count/JSON" \
  | python3 -c "import json,sys; print(json.load(sys.stdin)[0]['TOTALQUERYRESULTS'])")

# Page through results
PAGE=0
PAGE_SIZE=500
while [ $PAGE -lt $COUNT ]; do
  END=$((PAGE + PAGE_SIZE - 1))
  curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/YEAR/2022/rows/${PAGE}:${END}/JSON"
  PAGE=$((PAGE + PAGE_SIZE))
done

Warn the user if dataset exceeds 5,000 rows and suggest narrowing by state, year, gas, or subpart.

Error Handling

Condition Response Pattern Action
Table not available {"error": "TABLE/...: The table is not available."} Use V_GHG_EMITTER_SUBPART instead. V_GHG_EMITTER_FACILITIES and V_GHG_EMITTER_SECTOR are not served by this endpoint.
Empty result [] Valid JSON, no records Verify filter values. Check that year has data (latest may not be published yet). Try removing one filter to broaden.
Timeout / slow response No response within 30s Narrow filters (add STATE, reduce row range). National full-year queries can be very large.
Invalid column name Server error or empty result Check column names in references/api_reference.md. Column names in URL are case-insensitive.
Row range too large Slow response or incomplete data Keep page size at 500-1000 rows. Use count first to plan pagination.

Multi-Year Trend Analysis

The API does not support year ranges in a single request. To build a time series, issue one request per year and merge results:

for YEAR in 2018 2019 2020 2021 2022; do
  curl -s "https://data.epa.gov/efservice/V_GHG_EMITTER_SUBPART/SUBPART_NAME/W/STATE/WV/YEAR/${YEAR}/JSON"
done | python3 -c "
import json, sys
# Merge JSON arrays from multiple requests
all_data = []
for line in sys.stdin:
    line = line.strip()
    if line:
        all_data.extend(json.loads(line))
# Aggregate by year
by_year = {}
for r in all_data:
    y = r['year']
    by_year[y] = by_year.get(y, 0) + r['co2e_emission']
for y in sorted(by_year):
    print(f'{y}: {by_year[y]:,.0f} MT CO2e')
"

PNGE Research Context

Carbon Intensity of Produced Water Operations

Subpart W covers emissions from produced water handling — separators, tanks, pneumatic devices, and flares at well sites. When analyzing carbon intensity of produced water treatment for lithium/magnesium recovery:

  1. Query Subpart W for the target basin/state to establish baseline O&G emissions per facility
  2. Compare against proposed DLE (Direct Lithium Extraction) energy inputs
  3. Consider that co-producing Li/Mg from existing brine handling may add minimal incremental emissions vs. standalone mining operations

Methane as the Key Concern

For petroleum and natural gas systems (Subpart W), methane typically accounts for 40-60% of CO2e despite being a small fraction by mass. This is because EPA uses a GWP of 25 for CH4 (AR4 basis, 100-year).

Key methane emission sources in Subpart W:

  • Pneumatic controllers and pumps
  • Equipment leaks (fugitive emissions)
  • Associated gas venting and flaring
  • Liquids unloading
  • Gathering and boosting compressor stations
  • Produced water tanks (flash emissions)

Cross-Referencing with Other Skills

  • usgs-produced-waters: Match GHGRP facility locations with produced water chemistry to correlate brine Li/Mg content with emissions profiles.
  • wvges-wells: Link GHGRP facilities to specific well permits via operator name and county.
  • fracfocus: Cross-reference GHGRP reporting facilities with hydraulic fracturing chemical disclosures.
  • eia-data: Compare GHGRP-reported emissions against EIA production volumes to calculate emissions intensity (MT CO2e per barrel or per Mcf).

Caveats and Data Limitations

  1. Reporting threshold: Only facilities emitting 25,000+ MT CO2e/year (facility-wide) are required to report. Thousands of smaller O&G facilities fall below this threshold and are not in the dataset.
  2. Self-reported: Data is submitted by facility operators. EPA conducts verification but errors and underreporting occur.
  3. Subpart W segment detail: The Envirofacts API returns subpart-level totals. Segment-level breakdowns (e.g., onshore production vs. transmission compression within Subpart W) are only available via EPA FLIGHT bulk downloads (https://ghgdata.epa.gov/ghgp/).
  4. GWP values: CO2e calculations use IPCC AR4 GWP values (CH4 = 25). Some current policy uses AR5 (CH4 = 28) or AR6 (CH4 = 29.8). Results are not directly comparable across GWP bases.
  5. Publication lag: Data for reporting year N is typically published October of year N+1. The most recent complete year is usually current year minus 2.
  6. Double counting risk: Facilities may report under multiple subparts (e.g., Subpart C combustion + Subpart W process). Summing across subparts can overcount if not careful about emission source boundaries.
  7. Supplier vs. emitter: Subparts UU, PP, MM are supplier-type (subpart_type = "S"). These report quantities entering commerce, not direct atmospheric emissions. Do not add supplier and emitter subpart totals together for facility emissions.

Implementation Notes

  • Use bash_tool with curl + jq for API queries and JSON parsing
  • Python client — see references/python_example.py (stdlib only, no dependencies). Includes GHGClient, GHGQuery, pagination, and aggregation helpers.
  • Default to most recent available year unless user specifies
  • Always aggregate by facility_id or facility_name before ranking — raw rows contain one record per gas per subpart per year
  • Cap output tables at 20 rows for readability; note total count
  • Always include units (MT CO2e) and data year in output
  • Note when a query returns facility-wide vs. subpart-specific results
Weekly Installs
1
Repository
jpfielding/claude.pnge
First Seen
4 days ago
Security Audits
Gen Agent Trust HubPass
SocketPass
SnykPass
Installed on
amp1
cline1
opencode1
cursor1
kimi-cli1
codex1