| name | bsee-data-extractor |
| description | Extract and process BSEE (Bureau of Safety and Environmental Enforcement) data including production, WAR (Well Activity Reports), and APD (Application for Permit to Drill) data. Use for querying production data, well activities, drilling permits, completions, and workovers by API number, block, lease, or field with automatic data normalization and caching. |
BSEE Data Extractor
Extract and process production data from BSEE (Bureau of Safety and Environmental Enforcement) databases for oil & gas analysis in the Gulf of Mexico.
When to Use
- Querying BSEE production data by API number, block, or lease
- Downloading and parsing BSEE ZIP file archives
- Normalizing production data across different time periods
- Building production timelines for specific wells or fields
- Tracking well status changes over time
- Preparing data for economic analysis (NPV, decline curves)
- Analyzing Well Activity Reports (WAR) for drilling and completion history
- Tracking drilling operations, workovers, and sidetracks
- Reviewing APD (Application for Permit to Drill) records
- Calculating drilling and completion durations
- Building drilling timelines for rig scheduling analysis
Core Pattern
Query Parameters → BSEE API/Files → Parse → Normalize → Cache → Output
Data Types Supported
| Data Type | Source | Size | Update Frequency |
|---|
| Production | ProductionRawData.zip | ~15-50 MB | Monthly |
| WAR | eWellWARRawData.zip | ~120+ MB | Weekly |
| APD | APDRawData.zip | ~5-10 MB | Weekly |
Implementation
Data Models
from dataclasses import dataclass, field
from datetime import date
from typing import Optional, List, Dict, Any
from enum import Enum
import pandas as pd
class ProductType(Enum):
"""Production fluid types."""
OIL = "oil"
GAS = "gas"
WATER = "water"
CONDENSATE = "condensate"
class WellStatus(Enum):
"""Well operational status."""
PRODUCING = "producing"
SHUT_IN = "shut_in"
TEMPORARILY_ABANDONED = "ta"
PERMANENTLY_ABANDONED = "pa"
DRILLING = "drilling"
COMPLETING = "completing"
@dataclass
class WellIdentifier:
"""Unique well identification."""
api_number: str
well_name: Optional[str] = None
lease_number: Optional[str] = None
area_code: Optional[str] = None
block_number: Optional[str] = None
@property
def api_10(self) -> str:
"""Return 10-digit API number."""
return self.api_number[:10] if len(self.api_number) >= 10 else self.api_number
@property
def api_14(self) -> str:
"""Return full 14-digit API number."""
return self.api_number.ljust(14, '0')
@dataclass
class ProductionRecord:
"""Single month production record."""
well_id: WellIdentifier
production_date: date
oil_bbls: float = 0.0
gas_mcf: float = 0.0
water_bbls: float = 0.0
condensate_bbls: float = 0.0
days_on_production: int = 0
status: WellStatus = WellStatus.PRODUCING
@property
def oil_bopd(self) -> float:
"""Oil production in barrels per day."""
if self.days_on_production > 0:
return self.oil_bbls / self.days_on_production
return 0.0
@property
def gas_mcfd(self) -> float:
"""Gas production in MCF per day."""
if self.days_on_production > 0:
return self.gas_mcf / self.days_on_production
return 0.0
@property
def boe(self) -> float:
"""Barrels of oil equivalent (6:1 gas conversion)."""
return self.oil_bbls + self.condensate_bbls + (self.gas_mcf / 6.0)
@dataclass
class WellProduction:
"""Complete production history for a well."""
well_id: WellIdentifier
records: List[ProductionRecord] = field(default_factory=list)
first_production: Optional[date] = None
last_production: Optional[date] = None
def to_dataframe(self) -> pd.DataFrame:
class ActivityType(Enum):
"""Well activity types from WAR data."""
DRILLING = "drilling"
COMPLETION = "completion"
WORKOVER = "workover"
SIDETRACK = "sidetrack"
PLUG_ABANDON = "plug_abandon"
TEMPORARY_ABANDON = "temporary_abandon"
RECOMPLETION = "recompletion"
STIMULATION = "stimulation"
LOGGING = "logging"
TESTING = "testing"
class APDStatus(Enum):
"""APD application status."""
PENDING = "pending"
APPROVED = "approved"
DENIED = "denied"
WITHDRAWN = "withdrawn"
EXPIRED = "expired"
@dataclass
class WARRecord:
"""Well Activity Report record."""
well_id: WellIdentifier
activity_type: ActivityType
start_date: date
end_date: Optional[date] = None
spud_date: Optional[date] = None
rig_name: Optional[str] = None
water_depth_ft: Optional[float] = None
total_depth_md: Optional[float] = None
total_depth_tvd: Optional[float] = None
target_formation: Optional[str] = None
operator_name: Optional[str] = None
status: Optional[str] = None
remarks: Optional[str] = None
@property
def duration_days(self) -> Optional[int]:
"""Calculate activity duration in days."""
if self.end_date and self.start_date:
return (self.end_date - self.start_date).days
return None
@dataclass
class APDRecord:
"""Application for Permit to Drill record."""
well_id: WellIdentifier
application_date: date
approval_date: Optional[date] = None
status: APDStatus = APDStatus.PENDING
permit_number: Optional[str] = None
proposed_spud_date: Optional[date] = None
proposed_total_depth: Optional[float] = None
well_type: Optional[str] = None # exploration, development, injection
operator_name: Optional[str] = None
surface_location: Optional[str] = None
bottom_hole_location: Optional[str] = None
@dataclass
class WellActivity:
"""Complete activity history for a well."""
well_id: WellIdentifier
war_records: List[WARRecord] = field(default_factory=list)
apd_records: List[APDRecord] = field(default_factory=list)
first_activity: Optional[date] = None
last_activity: Optional[date] = None
def to_war_dataframe(self) -> pd.DataFrame:
"""Convert WAR records to DataFrame."""
data = []
for rec in self.war_records:
data.append({
'api_number': rec.well_id.api_number,
'activity_type': rec.activity_type.value,
'start_date': rec.start_date,
'end_date': rec.end_date,
'duration_days': rec.duration_days,
'spud_date': rec.spud_date,
'rig_name': rec.rig_name,
'water_depth_ft': rec.water_depth_ft,
'total_depth_md': rec.total_depth_md,
'total_depth_tvd': rec.total_depth_tvd,
'target_formation': rec.target_formation,
'operator_name': rec.operator_name,
'status': rec.status
})
return pd.DataFrame(data).sort_values('start_date') if data else pd.DataFrame()
def to_apd_dataframe(self) -> pd.DataFrame:
"""Convert APD records to DataFrame."""
data = []
for rec in self.apd_records:
data.append({
'api_number': rec.well_id.api_number,
'application_date': rec.application_date,
'approval_date': rec.approval_date,
'status': rec.status.value,
'permit_number': rec.permit_number,
'proposed_spud_date': rec.proposed_spud_date,
'proposed_total_depth': rec.proposed_total_depth,
'well_type': rec.well_type,
'operator_name': rec.operator_name
})
return pd.DataFrame(data).sort_values('application_date') if data else pd.DataFrame()
@property
def total_drilling_days(self) -> int:
"""Total days spent drilling."""
return sum(r.duration_days or 0 for r in self.war_records
if r.activity_type == ActivityType.DRILLING)
@property
def total_completion_days(self) -> int:
"""Total days spent on completions."""
return sum(r.duration_days or 0 for r in self.war_records
if r.activity_type == ActivityType.COMPLETION)
WAR/APD Data Models (continued)
def to_dataframe(self) -> pd.DataFrame:
"""Convert production records to DataFrame."""
data = []
for rec in self.records:
data.append({
'date': rec.production_date,
'oil_bbls': rec.oil_bbls,
'gas_mcf': rec.gas_mcf,
'water_bbls': rec.water_bbls,
'condensate_bbls': rec.condensate_bbls,
'days': rec.days_on_production,
'oil_bopd': rec.oil_bopd,
'gas_mcfd': rec.gas_mcfd,
'boe': rec.boe,
'status': rec.status.value
})
return pd.DataFrame(data).sort_values('date')
@property
def cumulative_oil(self) -> float:
"""Total cumulative oil production."""
return sum(r.oil_bbls for r in self.records)
@property
def cumulative_gas(self) -> float:
"""Total cumulative gas production."""
return sum(r.gas_mcf for r in self.records)
@property
def cumulative_boe(self) -> float:
"""Total cumulative BOE."""
return sum(r.boe for r in self.records)
BSEE Data Client
import requests
import zipfile
import io
from pathlib import Path
from typing import Optional, List, Dict, Generator
import pandas as pd
import logging
logger = logging.getLogger(__name__)
class BSEEDataClient:
"""
Client for accessing BSEE production data.
Supports both API access and file-based data extraction.
"""
# BSEE data URLs
BASE_URL = "https://www.data.bsee.gov"
PRODUCTION_URL = f"{BASE_URL}/Production/Files"
# Column mappings for BSEE data files
PRODUCTION_COLUMNS = {
'API_WELL_NUMBER': 'api_number',
'PRODUCTION_DATE': 'production_date',
'OIL': 'oil_bbls',
'GAS': 'gas_mcf',
'WATER': 'water_bbls',
'CONDENSATE': 'condensate_bbls',
'DAYS_ON_PROD': 'days_on_production',
'WELL_STAT_CD': 'status_code'
}
# WAR (Well Activity Report) column mappings
WAR_COLUMNS = {
'API_WELL_NUMBER': 'api_number',
'ACTIVITY_TYPE': 'activity_type',
'START_DATE': 'start_date',
'END_DATE': 'end_date',
'SPUD_DATE': 'spud_date',
'RIG_NAME': 'rig_name',
'WATER_DEPTH': 'water_depth_ft',
'TOTAL_DEPTH_MD': 'total_depth_md',
'TOTAL_DEPTH_TVD': 'total_depth_tvd',
'TARGET_FORMATION': 'target_formation',
'OPERATOR_NAME': 'operator_name',
'STATUS': 'status'
}
# APD (Application for Permit to Drill) column mappings
APD_COLUMNS = {
'API_WELL_NUMBER': 'api_number',
'APPLICATION_DATE': 'application_date',
'APPROVAL_DATE': 'approval_date',
'STATUS': 'status',
'PERMIT_NUMBER': 'permit_number',
'PROPOSED_SPUD_DATE': 'proposed_spud_date',
'PROPOSED_TD': 'proposed_total_depth',
'WELL_TYPE': 'well_type',
'OPERATOR_NAME': 'operator_name',
'SURFACE_LOCATION': 'surface_location',
'BHL_LOCATION': 'bottom_hole_location'
}
# Data URLs
WAR_URL = f"{BASE_URL}/Well/Files/eWellWARRawData.zip"
APD_URL = f"{BASE_URL}/Well/Files/APDRawData.zip"
def __init__(self, cache_dir: Path = None):
"""
Initialize BSEE data client.
Args:
cache_dir: Directory for caching downloaded data
"""
self.cache_dir = cache_dir or Path("data/bsee_cache")
self.cache_dir.mkdir(parents=True, exist_ok=True)
self.session = requests.Session()
def download_production_file(self, year: int, month: int = None) -> Path:
"""
Download BSEE production data file.
Args:
year: Production year
month: Optional month (downloads full year if not specified)
Returns:
Path to downloaded file
"""
if month:
filename = f"ogoramon{year}{month:02d}.zip"
else:
filename = f"ogoramon{year}.zip"
cache_path = self.cache_dir / filename
if cache_path.exists():
logger.info(f"Using cached file: {cache_path}")
return cache_path
url = f"{self.PRODUCTION_URL}/{filename}"
logger.info(f"Downloading: {url}")
response = self.session.get(url)
response.raise_for_status()
cache_path.write_bytes(response.content)
logger.info(f"Saved to: {cache_path}")
return cache_path
def extract_production_data(self, zip_path: Path) -> pd.DataFrame:
"""
Extract production data from BSEE ZIP file.
Args:
zip_path: Path to ZIP file
Returns:
DataFrame with production records
"""
with zipfile.ZipFile(zip_path, 'r') as zf:
# Find the data file (usually .txt or .csv)
data_files = [f for f in zf.namelist()
if f.endswith(('.txt', '.csv', '.dat'))]
if not data_files:
raise ValueError(f"No data file found in {zip_path}")
with zf.open(data_files[0]) as f:
# Try different delimiters
try:
df = pd.read_csv(f, delimiter='\t', low_memory=False)
except:
f.seek(0)
df = pd.read_csv(f, delimiter=',', low_memory=False)
# Rename columns to standard names
df = df.rename(columns={
k: v for k, v in self.PRODUCTION_COLUMNS.items()
if k in df.columns
})
return df
def query_by_api(self, api_number: str,
start_year: int = None,
end_year: int = None) -> WellProduction:
"""
Query production data for specific API number.
Args:
api_number: Well API number (10 or 14 digit)
start_year: Start year for query
end_year: End year for query
Returns:
WellProduction object with complete history
"""
from datetime import datetime
end_year = end_year or datetime.now().year
start_year = start_year or (end_year - 20) # Default 20 years
well_id = WellIdentifier(api_number=api_number)
records = []
for year in range(start_year, end_year + 1):
try:
zip_path = self.download_production_file(year)
df = self.extract_production_data(zip_path)
# Filter for API number (match on first 10 digits)
api_10 = well_id.api_10
df_well = df[df['api_number'].astype(str).str[:10] == api_10]
for _, row in df_well.iterrows():
records.append(self._row_to_record(row, well_id))
except Exception as e:
logger.warning(f"Error processing year {year}: {e}")
continue
production = WellProduction(well_id=well_id, records=records)
if records:
dates = [r.production_date for r in records]
production.first_production = min(dates)
production.last_production = max(dates)
return production
def query_by_block(self, area_code: str, block_number: str,
year: int) -> List[WellProduction]:
"""
Query all wells in a specific OCS block.
Args:
area_code: OCS area code (e.g., 'GC', 'WR', 'MC')
block_number: Block number
year: Year to query
Returns:
List of WellProduction objects
"""
zip_path = self.download_production_file(year)
df = self.extract_production_data(zip_path)
# Filter by area and block (column names may vary)
area_col = next((c for c in df.columns if 'AREA' in c.upper()), None)
block_col = next((c for c in df.columns if 'BLOCK' in c.upper()), None)
if area_col and block_col:
df_block = df[
(df[area_col].astype(str).str.upper() == area_code.upper()) &
(df[block_col].astype(str) == str(block_number))
]
else:
logger.warning("Could not find area/block columns")
return []
# Group by API number
wells = {}
for _, row in df_block.iterrows():
api = str(row.get('api_number', ''))
if api not in wells:
well_id = WellIdentifier(
api_number=api,
area_code=area_code,
block_number=block_number
)
wells[api] = WellProduction(well_id=well_id, records=[])
wells[api].records.append(self._row_to_record(row, wells[api].well_id))
return list(wells.values())
def _row_to_record(self, row: pd.Series, well_id: WellIdentifier) -> ProductionRecord:
"""Convert DataFrame row to ProductionRecord."""
prod_date = row.get('production_date')
if isinstance(prod_date, str):
# Parse various date formats
for fmt in ['%Y%m', '%Y-%m', '%Y/%m', '%Y%m%d', '%Y-%m-%d']:
try:
prod_date = pd.to_datetime(prod_date, format=fmt).date()
break
except:
continue
elif hasattr(prod_date, 'date'):
prod_date = prod_date.date()
else:
prod_date = date.today()
return ProductionRecord(
well_id=well_id,
production_date=prod_date,
oil_bbls=float(row.get('oil_bbls', 0) or 0),
gas_mcf=float(row.get('gas_mcf', 0) or 0),
water_bbls=float(row.get('water_bbls', 0) or 0),
condensate_bbls=float(row.get('condensate_bbls', 0) or 0),
days_on_production=int(row.get('days_on_production', 0) or 0),
status=self._parse_status(row.get('status_code', ''))
)
def _parse_status(self, code: str) -> WellStatus:
"""Parse BSEE status code to WellStatus enum."""
code = str(code).upper().strip()
status_map = {
'P': WellStatus.PRODUCING,
'SI': WellStatus.SHUT_IN,
'TA': WellStatus.TEMPORARILY_ABANDONED,
'PA': WellStatus.PERMANENTLY_ABANDONED,
'DR': WellStatus.DRILLING,
'CO': WellStatus.COMPLETING
}
return status_map.get(code, WellStatus.PRODUCING)
# ==================== WAR Data Methods ====================
def download_war_file(self, use_cache: bool = True) -> Path:
"""
Download BSEE Well Activity Report (WAR) data file.
Args:
use_cache: Use cached file if available
Returns:
Path to downloaded ZIP file
"""
cache_path = self.cache_dir / "eWellWARRawData.zip"
if use_cache and cache_path.exists():
logger.info(f"Using cached WAR file: {cache_path}")
return cache_path
logger.info(f"Downloading WAR data from: {self.WAR_URL}")
response = self.session.get(self.WAR_URL, timeout=2400) # 40 min timeout
response.raise_for_status()
cache_path.write_bytes(response.content)
logger.info(f"Saved WAR data: {len(response.content) / (1024*1024):.1f} MB")
return cache_path
def extract_war_data(self, zip_path: Path) -> pd.DataFrame:
"""
Extract WAR data from BSEE ZIP file.
Args:
zip_path: Path to WAR ZIP file
Returns:
DataFrame with WAR records
"""
with zipfile.ZipFile(zip_path, 'r') as zf:
data_files = [f for f in zf.namelist()
if f.endswith(('.txt', '.csv', '.dat'))]
if not data_files:
raise ValueError(f"No data file found in {zip_path}")
with zf.open(data_files[0]) as f:
try:
df = pd.read_csv(f, delimiter='\t', low_memory=False)
except:
f.seek(0)
df = pd.read_csv(f, delimiter=',', low_memory=False)
# Rename columns
df = df.rename(columns={
k: v for k, v in self.WAR_COLUMNS.items()
if k in df.columns
})
return df
def query_war_by_api(self, api_number: str) -> WellActivity:
"""
Query WAR data for specific API number.
Args:
api_number: Well API number (10 or 14 digit)
Returns:
WellActivity object with WAR records
"""
zip_path = self.download_war_file()
df = self.extract_war_data(zip_path)
well_id = WellIdentifier(api_number=api_number)
api_10 = well_id.api_10
df_well = df[df['api_number'].astype(str).str[:10] == api_10]
war_records = []
for _, row in df_well.iterrows():
war_records.append(self._row_to_war_record(row, well_id))
activity = WellActivity(well_id=well_id, war_records=war_records)
if war_records:
dates = [r.start_date for r in war_records if r.start_date]
if dates:
activity.first_activity = min(dates)
activity.last_activity = max(dates)
return activity
def query_war_by_block(self, area_code: str, block_number: str) -> List[WellActivity]:
"""
Query WAR data for all wells in a block.
Args:
area_code: OCS area code (e.g., 'GC', 'WR', 'MC')
block_number: Block number
Returns:
List of WellActivity objects
"""
zip_path = self.download_war_file()
df = self.extract_war_data(zip_path)
area_col = next((c for c in df.columns if 'AREA' in c.upper()), None)
block_col = next((c for c in df.columns if 'BLOCK' in c.upper()), None)
if area_col and block_col:
df_block = df[
(df[area_col].astype(str).str.upper() == area_code.upper()) &
(df[block_col].astype(str) == str(block_number))
]
else:
logger.warning("Could not find area/block columns in WAR data")
return []
wells = {}
for _, row in df_block.iterrows():
api = str(row.get('api_number', ''))
if api not in wells:
well_id = WellIdentifier(
api_number=api,
area_code=area_code,
block_number=block_number
)
wells[api] = WellActivity(well_id=well_id)
wells[api].war_records.append(self._row_to_war_record(row, wells[api].well_id))
return list(wells.values())
def _row_to_war_record(self, row: pd.Series, well_id: WellIdentifier) -> WARRecord:
"""Convert DataFrame row to WARRecord."""
def parse_date(val):
if pd.isna(val):
return None
if isinstance(val, str):
for fmt in ['%Y-%m-%d', '%Y%m%d', '%m/%d/%Y']:
try:
return pd.to_datetime(val, format=fmt).date()
except:
continue
elif hasattr(val, 'date'):
return val.date()
return None
activity_map = {
'DRILL': ActivityType.DRILLING,
'COMP': ActivityType.COMPLETION,
'WORK': ActivityType.WORKOVER,
'SIDE': ActivityType.SIDETRACK,
'PA': ActivityType.PLUG_ABANDON,
'TA': ActivityType.TEMPORARY_ABANDON,
'RECOMP': ActivityType.RECOMPLETION,
'STIM': ActivityType.STIMULATION,
'LOG': ActivityType.LOGGING,
'TEST': ActivityType.TESTING
}
activity_str = str(row.get('activity_type', '')).upper()
activity_type = ActivityType.DRILLING
for key, val in activity_map.items():
if key in activity_str:
activity_type = val
break
return WARRecord(
well_id=well_id,
activity_type=activity_type,
start_date=parse_date(row.get('start_date')) or date.today(),
end_date=parse_date(row.get('end_date')),
spud_date=parse_date(row.get('spud_date')),
rig_name=row.get('rig_name'),
water_depth_ft=float(row.get('water_depth_ft', 0) or 0) if pd.notna(row.get('water_depth_ft')) else None,
total_depth_md=float(row.get('total_depth_md', 0) or 0) if pd.notna(row.get('total_depth_md')) else None,
total_depth_tvd=float(row.get('total_depth_tvd', 0) or 0) if pd.notna(row.get('total_depth_tvd')) else None,
target_formation=row.get('target_formation'),
operator_name=row.get('operator_name'),
status=row.get('status')
)
# ==================== APD Data Methods ====================
def download_apd_file(self, use_cache: bool = True) -> Path:
"""
Download BSEE APD (Application for Permit to Drill) data file.
Args:
use_cache: Use cached file if available
Returns:
Path to downloaded ZIP file
"""
cache_path = self.cache_dir / "APDRawData.zip"
if use_cache and cache_path.exists():
logger.info(f"Using cached APD file: {cache_path}")
return cache_path
logger.info(f"Downloading APD data from: {self.APD_URL}")
response = self.session.get(self.APD_URL, timeout=600) # 10 min timeout
response.raise_for_status()
cache_path.write_bytes(response.content)
logger.info(f"Saved APD data: {len(response.content) / (1024*1024):.1f} MB")
return cache_path
def extract_apd_data(self, zip_path: Path) -> pd.DataFrame:
"""
Extract APD data from BSEE ZIP file.
Args:
zip_path: Path to APD ZIP file
Returns:
DataFrame with APD records
"""
with zipfile.ZipFile(zip_path, 'r') as zf:
data_files = [f for f in zf.namelist()
if f.endswith(('.txt', '.csv', '.dat'))]
if not data_files:
raise ValueError(f"No data file found in {zip_path}")
with zf.open(data_files[0]) as f:
try:
df = pd.read_csv(f, delimiter='\t', low_memory=False)
except:
f.seek(0)
df = pd.read_csv(f, delimiter=',', low_memory=False)
df = df.rename(columns={
k: v for k, v in self.APD_COLUMNS.items()
if k in df.columns
})
return df
def query_apd_by_api(self, api_number: str) -> List[APDRecord]:
"""
Query APD records for specific API number.
Args:
api_number: Well API number
Returns:
List of APDRecord objects
"""
zip_path = self.download_apd_file()
df = self.extract_apd_data(zip_path)
well_id = WellIdentifier(api_number=api_number)
api_10 = well_id.api_10
df_well = df[df['api_number'].astype(str).str[:10] == api_10]
records = []
for _, row in df_well.iterrows():
records.append(self._row_to_apd_record(row, well_id))
return records
def _row_to_apd_record(self, row: pd.Series, well_id: WellIdentifier) -> APDRecord:
"""Convert DataFrame row to APDRecord."""
def parse_date(val):
if pd.isna(val):
return None
if isinstance(val, str):
for fmt in ['%Y-%m-%d', '%Y%m%d', '%m/%d/%Y']:
try:
return pd.to_datetime(val, format=fmt).date()
except:
continue
elif hasattr(val, 'date'):
return val.date()
return None
status_map = {
'PEND': APDStatus.PENDING,
'APPR': APDStatus.APPROVED,
'DENY': APDStatus.DENIED,
'DENY': APDStatus.DENIED,
'WITH': APDStatus.WITHDRAWN,
'EXP': APDStatus.EXPIRED
}
status_str = str(row.get('status', '')).upper()
status = APDStatus.PENDING
for key, val in status_map.items():
if key in status_str:
status = val
break
return APDRecord(
well_id=well_id,
application_date=parse_date(row.get('application_date')) or date.today(),
approval_date=parse_date(row.get('approval_date')),
status=status,
permit_number=row.get('permit_number'),
proposed_spud_date=parse_date(row.get('proposed_spud_date')),
proposed_total_depth=float(row.get('proposed_total_depth', 0) or 0) if pd.notna(row.get('proposed_total_depth')) else None,
well_type=row.get('well_type'),
operator_name=row.get('operator_name'),
surface_location=row.get('surface_location'),
bottom_hole_location=row.get('bottom_hole_location')
)
Production Aggregator
from typing import Dict, List, Tuple
import pandas as pd
import numpy as np
from datetime import date
class ProductionAggregator:
"""
Aggregate production data across wells, fields, or time periods.
"""
def __init__(self, wells: List[WellProduction]):
"""
Initialize aggregator with well production data.
Args:
wells: List of WellProduction objects
"""
self.wells = wells
self._combined_df = None
@property
def combined_dataframe(self) -> pd.DataFrame:
"""Get combined production data as DataFrame."""
if self._combined_df is None:
dfs = []
for well in self.wells:
df = well.to_dataframe()
df['api_number'] = well.well_id.api_number
df['well_name'] = well.well_id.well_name
dfs.append(df)
self._combined_df = pd.concat(dfs, ignore_index=True)
return self._combined_df
def monthly_totals(self) -> pd.DataFrame:
"""Aggregate production by month."""
df = self.combined_dataframe.copy()
df['year_month'] = pd.to_datetime(df['date']).dt.to_period('M')
return df.groupby('year_month').agg({
'oil_bbls': 'sum',
'gas_mcf': 'sum',
'water_bbls': 'sum',
'boe': 'sum',
'days': 'sum',
'api_number': 'nunique'
}).rename(columns={'api_number': 'well_count'}).reset_index()
def annual_totals(self) -> pd.DataFrame:
"""Aggregate production by year."""
df = self.combined_dataframe.copy()
df['year'] = pd.to_datetime(df['date']).dt.year
return df.groupby('year').agg({
'oil_bbls': 'sum',
'gas_mcf': 'sum',
'water_bbls': 'sum',
'boe': 'sum',
'days': 'sum',
'api_number': 'nunique'
}).rename(columns={'api_number': 'well_count'}).reset_index()
def well_summary(self) -> pd.DataFrame:
"""Get summary statistics per well."""
summaries = []
for well in self.wells:
df = well.to_dataframe()
if len(df) == 0:
continue
summary = {
'api_number': well.well_id.api_number,
'well_name': well.well_id.well_name,
'first_production': df['date'].min(),
'last_production': df['date'].max(),
'months_producing': len(df),
'cumulative_oil': df['oil_bbls'].sum(),
'cumulative_gas': df['gas_mcf'].sum(),
'cumulative_boe': df['boe'].sum(),
'peak_oil_bopd': df['oil_bopd'].max(),
'peak_gas_mcfd': df['gas_mcfd'].max(),
'avg_oil_bopd': df['oil_bopd'].mean(),
'avg_gas_mcfd': df['gas_mcfd'].mean()
}
summaries.append(summary)
return pd.DataFrame(summaries)
def decline_curve_data(self, well: WellProduction) -> Dict:
"""
Prepare data for decline curve analysis.
Returns time on production and rate data.
"""
df = well.to_dataframe()
df = df[df['oil_bopd'] > 0].sort_values('date')
if len(df) == 0:
return {'time': [], 'rate': []}
# Calculate months from first production
first_date = df['date'].iloc[0]
df['months'] = ((pd.to_datetime(df['date']) - pd.to_datetime(first_date))
.dt.days / 30.44).astype(int)
return {
'time': df['months'].tolist(),
'rate': df['oil_bopd'].tolist(),
'dates': df['date'].tolist()
}
Activity Aggregator
from typing import Dict, List
import pandas as pd
class ActivityAggregator:
"""
Aggregate WAR and APD data across wells, fields, or time periods.
"""
def __init__(self, activities: List[WellActivity]):
"""
Initialize aggregator with well activity data.
Args:
activities: List of WellActivity objects
"""
self.activities = activities
self._combined_war_df = None
self._combined_apd_df = None
@property
def combined_war_dataframe(self) -> pd.DataFrame:
"""Get combined WAR data as DataFrame."""
if self._combined_war_df is None:
dfs = []
for activity in self.activities:
df = activity.to_war_dataframe()
if not df.empty:
df['area_code'] = activity.well_id.area_code
df['block_number'] = activity.well_id.block_number
dfs.append(df)
self._combined_war_df = pd.concat(dfs, ignore_index=True) if dfs else pd.DataFrame()
return self._combined_war_df
def activity_summary(self) -> pd.DataFrame:
"""Summarize activities by type."""
df = self.combined_war_dataframe
if df.empty:
return pd.DataFrame()
return df.groupby('activity_type').agg({
'api_number': 'count',
'duration_days': ['sum', 'mean', 'min', 'max']
}).round(1)
def drilling_timeline(self) -> pd.DataFrame:
"""Get drilling activity timeline."""
df = self.combined_war_dataframe
if df.empty:
return pd.DataFrame()
drilling = df[df['activity_type'] == 'drilling'].copy()
drilling = drilling.sort_values('start_date')
return drilling[['api_number', 'start_date', 'end_date', 'duration_days',
'rig_name', 'water_depth_ft', 'total_depth_md', 'operator_name']]
def rig_utilization(self) -> pd.DataFrame:
"""Analyze rig utilization from WAR data."""
df = self.combined_war_dataframe
if df.empty:
return pd.DataFrame()
rig_summary = df.groupby('rig_name').agg({
'api_number': 'nunique',
'duration_days': 'sum',
'start_date': 'min',
'end_date': 'max'
}).rename(columns={
'api_number': 'wells_drilled',
'duration_days': 'total_days'
})
return rig_summary.sort_values('total_days', ascending=False)
def operator_activity(self) -> pd.DataFrame:
"""Analyze activity by operator."""
df = self.combined_war_dataframe
if df.empty:
return pd.DataFrame()
return df.groupby('operator_name').agg({
'api_number': 'nunique',
'activity_type': 'count',
'duration_days': 'sum',
'water_depth_ft': 'mean',
'total_depth_md': 'mean'
}).rename(columns={
'api_number': 'unique_wells',
'activity_type': 'total_activities',
'duration_days': 'total_days',
'water_depth_ft': 'avg_water_depth',
'total_depth_md': 'avg_total_depth'
}).round(0)
def depth_statistics(self) -> Dict:
"""Calculate depth statistics from drilling data."""
df = self.combined_war_dataframe
drilling = df[df['activity_type'] == 'drilling']
if drilling.empty:
return {}
return {
'water_depth': {
'min': drilling['water_depth_ft'].min(),
'max': drilling['water_depth_ft'].max(),
'mean': drilling['water_depth_ft'].mean(),
'median': drilling['water_depth_ft'].median()
},
'total_depth_md': {
'min': drilling['total_depth_md'].min(),
'max': drilling['total_depth_md'].max(),
'mean': drilling['total_depth_md'].mean(),
'median': drilling['total_depth_md'].median()
}
}
Report Generator
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from pathlib import Path
class BSEEReportGenerator:
"""Generate interactive HTML reports for BSEE data."""
def __init__(self, aggregator: ProductionAggregator):
"""
Initialize report generator.
Args:
aggregator: ProductionAggregator with well data
"""
self.aggregator = aggregator
def generate_field_report(self, output_path: Path, field_name: str = "Field"):
"""
Generate comprehensive field production report.
Args:
output_path: Path for HTML output
field_name: Name of the field for report title
"""
fig = make_subplots(
rows=3, cols=2,
subplot_titles=(
'Monthly Oil Production',
'Monthly Gas Production',
'Cumulative Production',
'Well Count Over Time',
'Water Cut Trend',
'GOR Trend'
),
vertical_spacing=0.08
)
monthly = self.aggregator.monthly_totals()
monthly['date'] = monthly['year_month'].dt.to_timestamp()
# Monthly oil production
fig.add_trace(
go.Bar(x=monthly['date'], y=monthly['oil_bbls']/1000,
name='Oil (Mbbls)', marker_color='green'),
row=1, col=1
)
# Monthly gas production
fig.add_trace(
go.Bar(x=monthly['date'], y=monthly['gas_mcf']/1000,
name='Gas (MMCF)', marker_color='red'),
row=1, col=2
)
# Cumulative production
monthly['cum_oil'] = monthly['oil_bbls'].cumsum() / 1e6
monthly['cum_gas'] = monthly['gas_mcf'].cumsum() / 1e6
fig.add_trace(
go.Scatter(x=monthly['date'], y=monthly['cum_oil'],
name='Cum Oil (MMbbls)', line=dict(color='green')),
row=2, col=1
)
fig.add_trace(
go.Scatter(x=monthly['date'], y=monthly['cum_gas'],
name='Cum Gas (BCF)', line=dict(color='red')),
row=2, col=1
)
# Well count
fig.add_trace(
go.Scatter(x=monthly['date'], y=monthly['well_count'],
name='Active Wells', fill='tozeroy'),
row=2, col=2
)
# Water cut
df = self.aggregator.combined_dataframe.copy()
df['date'] = pd.to_datetime(df['date'])
monthly_avg = df.groupby(df['date'].dt.to_period('M')).agg({
'oil_bbls': 'sum',
'water_bbls': 'sum',
'gas_mcf': 'sum'
}).reset_index()
monthly_avg['date'] = monthly_avg['date'].dt.to_timestamp()
monthly_avg['water_cut'] = (monthly_avg['water_bbls'] /
(monthly_avg['oil_bbls'] + monthly_avg['water_bbls'])) * 100
monthly_avg['gor'] = monthly_avg['gas_mcf'] / monthly_avg['oil_bbls'].replace(0, np.nan)
fig.add_trace(
go.Scatter(x=monthly_avg['date'], y=monthly_avg['water_cut'],
name='Water Cut (%)', line=dict(color='blue')),
row=3, col=1
)
# GOR
fig.add_trace(
go.Scatter(x=monthly_avg['date'], y=monthly_avg['gor'],
name='GOR (MCF/bbl)', line=dict(color='orange')),
row=3, col=2
)
fig.update_layout(
height=1000,
title_text=f"{field_name} Production Analysis",
showlegend=True
)
# Write report
output_path.parent.mkdir(parents=True, exist_ok=True)
fig.write_html(str(output_path))
return output_path
def generate_well_report(self, well: WellProduction, output_path: Path):
"""Generate individual well production report."""
df = well.to_dataframe()
fig = make_subplots(
rows=2, cols=2,
subplot_titles=(
'Daily Oil Rate',
'Daily Gas Rate',
'Cumulative Production',
'Production Decline'
)
)
# Oil rate
fig.add_trace(
go.Scatter(x=df['date'], y=df['oil_bopd'],
name='Oil (BOPD)', line=dict(color='green')),
row=1, col=1
)
# Gas rate
fig.add_trace(
go.Scatter(x=df['date'], y=df['gas_mcfd'],
name='Gas (MCFD)', line=dict(color='red')),
row=1, col=2
)
# Cumulative
df['cum_oil'] = df['oil_bbls'].cumsum() / 1000
df['cum_gas'] = df['gas_mcf'].cumsum() / 1000
fig.add_trace(
go.Scatter(x=df['date'], y=df['cum_oil'],
name='Cum Oil (Mbbls)'),
row=2, col=1
)
# Semi-log decline plot
df_producing = df[df['oil_bopd'] > 0]
fig.add_trace(
go.Scatter(x=df_producing['date'], y=df_producing['oil_bopd'],
name='Rate (BOPD)', mode='markers'),
row=2, col=2
)
fig.update_yaxes(type='log', row=2, col=2)
fig.update_layout(
height=800,
title_text=f"Well {well.well_id.api_number} Production History",
showlegend=True
)
fig.write_html(str(output_path))
return output_path
YAML Configuration
Query Configuration
# config/bsee_query.yaml
metadata:
task: bsee_data_extraction
created: "2024-01-15"
query:
type: by_block # by_api, by_block, by_lease, by_field
# API query parameters
api_number: "1771049130"
# Block query parameters
area_code: "GC" # Green Canyon
block_number: "640"
# Time range
start_year: 2010
end_year: 2024
data_options:
include_gas: true
include_water: true
include_condensate: true
normalize_rates: true # Convert to daily rates
cache:
enabled: true
directory: "data/bsee_cache"
expiry_days: 30
output:
format: "csv" # csv, parquet, json
path: "data/results/bsee_extract.csv"
report:
enabled: true
path: "reports/bsee_production.html"
title: "BSEE Production Analysis"
WAR Query Configuration
# config/war_query.yaml
metadata:
task: war_data_extraction
created: "2024-01-15"
query:
type: by_block # by_api, by_block, by_area
# API query parameters
api_number: "1771049130"
# Block query parameters
area_code: "GC" # Green Canyon
block_number: "640"
activity_filters:
types:
- drilling
- completion
- workover
- sidetrack
start_date: "2020-01-01"
end_date: "2024-12-31"
analysis:
drilling_timeline: true
rig_utilization: true
operator_activity: true
depth_statistics: true
cache:
enabled: true
directory: "data/bsee_cache"
expiry_days: 7 # WAR data updates more frequently
output:
format: "csv"
war_data: "data/results/war_extract.csv"
drilling_timeline: "data/results/drilling_timeline.csv"
report:
enabled: true
path: "reports/war_activity.html"
title: "Well Activity Report Analysis"
APD Query Configuration
# config/apd_query.yaml
metadata:
task: apd_data_extraction
query:
type: by_area
area_code: "WR" # Walker Ridge
status_filters:
- approved
- pending
date_range:
start: "2023-01-01"
end: "2024-12-31"
analysis:
approval_timeline: true
operator_summary: true
well_type_breakdown: true
output:
format: "csv"
path: "data/results/apd_extract.csv"
Multi-Well Configuration
# config/field_analysis.yaml
metadata:
task: field_analysis
field_name: "Lower Tertiary Development"
wells:
- api_number: "1771049130"
name: "Well A-1"
type: "producer"
- api_number: "1771049131"
name: "Well A-2"
type: "producer"
- api_number: "1771049132"
name: "Well B-1"
type: "injector"
analysis:
production_summary: true
decline_analysis: true
water_cut_trend: true
gor_trend: true
time_range:
start: "2015-01-01"
end: "2024-12-31"
output:
summary_csv: "data/results/field_summary.csv"
well_reports: "reports/wells/"
field_report: "reports/field_production.html"
CLI Usage
Basic Queries
# Query by API number
python -m bsee_extractor query --api 1771049130 --output data/well_production.csv
# Query by block
python -m bsee_extractor query --area GC --block 640 --year 2023
# Query multiple years
python -m bsee_extractor query --api 1771049130 --start-year 2010 --end-year 2024
Report Generation
# Generate well report
python -m bsee_extractor report --api 1771049130 --output reports/well_report.html
# Generate field report from config
python -m bsee_extractor report --config config/field_analysis.yaml
WAR Queries
# Query WAR data by API number
python -m bsee_extractor war --api 1771049130 --output data/war_well.csv
# Query WAR data by block
python -m bsee_extractor war --area GC --block 640 --output data/war_block.csv
# Filter by activity type
python -m bsee_extractor war --area WR --block 758 --activity drilling,completion
# Generate drilling timeline report
python -m bsee_extractor war-report --area GC --block 640 --output reports/drilling_timeline.html
APD Queries
# Query APD data by API number
python -m bsee_extractor apd --api 1771049130 --output data/apd_well.csv
# Query APD data by area
python -m bsee_extractor apd --area WR --status approved,pending --output data/apd_area.csv
# Recent APD applications
python -m bsee_extractor apd --area GC --since 2023-01-01 --output data/recent_apd.csv
Data Export
# Export production to CSV
python -m bsee_extractor export --api 1771049130 --format csv --output data/export.csv
# Export to Parquet (for large datasets)
python -m bsee_extractor export --area GC --block 640 --format parquet --output data/block_production.parquet
# Export combined production + WAR data
python -m bsee_extractor export --api 1771049130 --include-war --format csv --output data/combined.csv
Usage Examples
Example 1: Single Well Analysis
from bsee_extractor import BSEEDataClient, BSEEReportGenerator
# Initialize client
client = BSEEDataClient(cache_dir=Path("data/bsee_cache"))
# Query well production
well = client.query_by_api("1771049130", start_year=2010, end_year=2024)
# Get production DataFrame
df = well.to_dataframe()
print(f"Records: {len(df)}")
print(f"Cumulative Oil: {well.cumulative_oil:,.0f} bbls")
print(f"Cumulative Gas: {well.cumulative_gas:,.0f} MCF")
# Generate report
from production_aggregator import ProductionAggregator
aggregator = ProductionAggregator([well])
reporter = BSEEReportGenerator(aggregator)
reporter.generate_well_report(well, Path("reports/well_analysis.html"))
Example 2: Block-Level Analysis
# Query all wells in Green Canyon Block 640
wells = client.query_by_block("GC", "640", year=2023)
print(f"Found {len(wells)} wells in GC 640")
# Aggregate production
aggregator = ProductionAggregator(wells)
# Monthly totals
monthly = aggregator.monthly_totals()
print(f"\n2023 Production:")
print(f" Total Oil: {monthly['oil_bbls'].sum():,.0f} bbls")
print(f" Total Gas: {monthly['gas_mcf'].sum():,.0f} MCF")
# Well summary
summary = aggregator.well_summary()
print(f"\nWell Rankings by Cumulative Oil:")
print(summary.sort_values('cumulative_oil', ascending=False)[
['api_number', 'cumulative_oil', 'peak_oil_bopd']
].head(10))
# Generate field report
reporter = BSEEReportGenerator(aggregator)
reporter.generate_field_report(
Path("reports/gc640_production.html"),
field_name="Green Canyon 640"
)
Example 3: Decline Curve Preparation
# Get decline curve data for type curve analysis
well = client.query_by_api("1771049130")
decline_data = aggregator.decline_curve_data(well)
# Prepare for Arps decline fitting
import numpy as np
time = np.array(decline_data['time'])
rate = np.array(decline_data['rate'])
# Filter to producing periods only
mask = rate > 0
time_producing = time[mask]
rate_producing = rate[mask]
# Export for external decline curve software
decline_df = pd.DataFrame({
'months': time_producing,
'oil_bopd': rate_producing
})
decline_df.to_csv("data/decline_input.csv", index=False)
Example 4: WAR Activity Analysis
from bsee_extractor import BSEEDataClient, ActivityAggregator
# Initialize client
client = BSEEDataClient(cache_dir=Path("data/bsee_cache"))
# Query WAR data for a specific well
well_activity = client.query_war_by_api("1771049130")
# Get WAR DataFrame
war_df = well_activity.to_war_dataframe()
print(f"Total activities: {len(war_df)}")
print(f"Activity types: {war_df['activity_type'].unique()}")
# Calculate drilling and completion durations
print(f"\nTotal drilling days: {well_activity.total_drilling_days}")
print(f"Total completion days: {well_activity.total_completion_days}")
# Detailed drilling history
drilling_df = war_df[war_df['activity_type'] == 'drilling']
print(f"\nDrilling Operations:")
print(drilling_df[['start_date', 'end_date', 'duration_days', 'rig_name', 'total_depth_md']])
Example 5: Block-Level Drilling Analysis
# Query all well activities in Walker Ridge Block 758
activities = client.query_war_by_block("WR", "758")
print(f"Found {len(activities)} wells with WAR records in WR 758")
# Aggregate activities
activity_agg = ActivityAggregator(activities)
# Drilling timeline for rig scheduling
drilling_timeline = activity_agg.drilling_timeline()
print("\nDrilling Timeline:")
print(drilling_timeline[['api_number', 'start_date', 'rig_name', 'duration_days']])
# Rig utilization analysis
rig_util = activity_agg.rig_utilization()
print("\nRig Utilization:")
print(rig_util)
# Operator activity summary
operator_summary = activity_agg.operator_activity()
print("\nOperator Activity:")
print(operator_summary)
# Depth statistics
depth_stats = activity_agg.depth_statistics()
print(f"\nWater Depth Range: {depth_stats['water_depth']['min']:.0f} - {depth_stats['water_depth']['max']:.0f} ft")
print(f"Avg Total Depth: {depth_stats['total_depth_md']['mean']:.0f} ft MD")
Example 6: APD Tracking and Permit Analysis
# Query APD records for a well
apd_records = client.query_apd_by_api("1771049130")
print(f"Found {len(apd_records)} APD records")
for apd in apd_records:
print(f"\nPermit: {apd.permit_number}")
print(f" Application Date: {apd.application_date}")
print(f" Status: {apd.status.value}")
print(f" Well Type: {apd.well_type}")
if apd.approval_date:
print(f" Approval Date: {apd.approval_date}")
days_to_approve = (apd.approval_date - apd.application_date).days
print(f" Days to Approval: {days_to_approve}")
Example 7: Combined Production + Activity Analysis
# Get both production and activity data for a well
well_production = client.query_by_api("1771049130", start_year=2015)
well_activity = client.query_war_by_api("1771049130")
# Combine for comprehensive well analysis
print(f"Well: {well_production.well_id.api_number}")
print(f"\n--- Production Summary ---")
print(f"First Production: {well_production.first_production}")
print(f"Cumulative Oil: {well_production.cumulative_oil:,.0f} bbls")
print(f"Cumulative Gas: {well_production.cumulative_gas:,.0f} MCF")
print(f"\n--- Activity Summary ---")
print(f"First Activity: {well_activity.first_activity}")
print(f"Drilling Days: {well_activity.total_drilling_days}")
print(f"Completion Days: {well_activity.total_completion_days}")
print(f"Total WAR Records: {len(well_activity.war_records)}")
# Calculate time from spud to first production
war_df = well_activity.to_war_dataframe()
drilling_records = war_df[war_df['activity_type'] == 'drilling']
if not drilling_records.empty and well_production.first_production:
spud_date = drilling_records['spud_date'].min()
if spud_date:
days_to_production = (well_production.first_production - spud_date).days
print(f"\nDays from Spud to First Production: {days_to_production}")
Best Practices
Data Caching
- Enable caching to avoid repeated downloads
- Set appropriate expiry for frequently updated data
- Use local cache for development, clear for production runs
Query Optimization
- Query by year ranges to limit data volume
- Use block/area queries for field-level analysis
- Filter early in pipeline to reduce memory usage
Error Handling
- Handle missing data periods gracefully
- Log warnings for data quality issues
- Validate API numbers before querying
File Organization
project/
├── config/
│ ├── bsee_query.yaml
│ └── field_analysis.yaml
├── data/
│ ├── bsee_cache/ # Downloaded BSEE files
│ ├── raw/ # Raw extracted data
│ ├── processed/ # Cleaned data
│ └── results/ # Analysis outputs
├── reports/
│ ├── wells/ # Individual well reports
│ └── field_production.html
└── src/
└── bsee_extractor/
├── client.py
├── models.py
├── aggregator.py
└── reports.py
Related Skills
