ONC Pressure Data Access

Jupyter notebooks for accessing and processing bottom pressure recorder (BPR) data from Ocean Networks Canada's (ONC) cabled observatory at Barkley Canyon via the ONC Open API. Covers two cabled ONC instrument types: the RBR BPR|Zero and the Sonardyne Fetch AZA .

---

Repository Structure

BarkleyCanyon_Cabled/
├── BPR_Zero_Cabled.ipynb            # Main BPR|Zero workflow: download, drift analysis, AZA calibration
├── BPR_Zero_Cabled_getLogFiles.ipynb # Download raw archive log files for BPR|Zero
├── SonardyneFetch.ipynb                # Download and parse Sonardyne Fetch serial data
├── ParseFetches.py                     # Reusable class for fetching and parsing Sonardyne Fetch data
└── README.md

---

Prerequisites

ONC API Token

All notebooks require an ONC API token (see here for information about ONCs OpenAPI:https://oceannetworkscanada.github.io/api-python-client/). Store it in a .env file in the notebook directory:

ONC_TOKEN=your_token_here

Tokens are available from your ONC account page.

Python Dependencies

onc          # ONC Python client library
requests
requests_cache
pandas
numpy
scipy
matplotlib
python-dotenv

Install with:

pip install onc requests requests-cache pandas numpy scipy matplotlib python-dotenv

---

Notebooks

1. BPR_Zero_Cabled.ipynb — BPR|Zero Analysis

Downloads scalar and raw data for the cabled RBR Quartz BPR|Zero at Barkley Canyon, characterises sensor drift, detects AZA in-situ calibration cycles, and compares against a reference BPR.

Device: RBRQUARTZ3BPRZERO207223
Locations: NCBC.P1 (BPR|Zero), NCBC (reference non-AZA BPR)
API methods: scalardata/getByDevice, scalardata/getByLocation, onc.getDirectRawByDevice

Workflow:

1. Fetch 900-second averaged scalar pressure (pressure1) data via the ONC REST API
2. Plot pressure time series; filter out calibration pressure drops (outside 410–412 dbar).
3. Fit exponential + linear drift models using scipy.optimize.curve_fit and numpy.polyfit.
4. Detect AZA calibration events from systemstatus sensor data (step-down transitions).
5. For each calibration event, fetch a 900 s window of raw scalar data and compute the BPR|Zero − Reference pressure difference.
6. Overlay all calibration cycles and apply a 16-sample rolling mean.
7. Compare 15-min averaged pressure from NCBC.P1 and NCBC; compute and plot residuals with a 25-hour rolling mean.
8. Retrieve raw archived data via onc.getDirectRawByDevice; split CSV-encoded lines into SF pressure, SF temperature, barometric pressure, and barometric temperature channels.

---

2. BPR_Zero_Cabled_getLogFiles.ipynb — Download Archive Log Files

Downloads raw binary/ASCII log files from the ONC archive for the cabled BPR|Zero at Barkley Canyon (locationCode: NCBC.P1). Useful for offline or batch processing of the raw instrument data.

Device: RBRQUARTZ3BPRZERO207223
API methods: onc.getArchivefileByDevice, onc.downloadArchivefile

Workflow:

1. Query available archive files for a given date range (getArchivefileByDevice).
2. Download files to a local output/ directory, skipping files already present.

---

3. SonardyneFetch.ipynb — Sonardyne Fetch Data

Downloads, parses, and plots raw serial ASCII data from Sonardyne Fetch AZA and non-AZA autonomous pressure loggers at Barkley Canyon. Also documents the Sonardyne Fetch serial message format.

Devices (three Fetch units):

Device Code	Unit
SONARDYNEFETCH8306SN329744001	Fetch AZA
SONARDYNEFETCH8306SN329742001	Fetch North
SONARDYNEFETCH8306SN329742002	Fetch East

API method: onc.getDirectRawByDevice (via ParseFetches)

Workflow:

1. Instantiate ParseFetches(SN, deviceCode, dateFrom, dateTo) and call getONCRawData() to retrieve raw ASCII lines.
2. Call parseData() to decode message types and extract sensor columns (see table below).
3. Plot Digiquartz (DQZ) and Keller (KLR) pressure and temperature as 4-panel subplots.
4. Resample to 15-minute means and re-plot.
5. Optionally save/load parsed data as Parquet.

Parsed message types and output columns:

Message	Columns
DQZ	DQZ_Press (dbar), DQZ_Temp
KLR	KLR_PRESS (dbar), KLR_Temp
TMP	TMP_Temp
INC	INC_Pitch, INC_Roll
PIES	PIES_PRESS (dbar), PIES_TOF, PIES_Mag, and others
AZA	Primary/secondary/low-port pressures and temperatures, RMS error, settling flag
AZS	AZA status summary (start/end of calibration cycle)

Pressure values are converted from kPa to dbar (×0.1) during parsing.

---

4. ParseFetches.py — Sonardyne Fetch Parser Module

Reusable Python class ParseFetches that encapsulates all ONC API calls and serial message parsing for Sonardyne Fetch instruments. Used by SonardyneFetch.ipynb.

from ParseFetches import ParseFetches

fetcher = ParseFetches(SN, deviceCode, dateFrom, dateTo)
df_raw  = fetcher.getONCRawData()   # raw ASCII lines with timestamp index
df      = fetcher.parseData()       # parsed sensor columns

---

Notes

• All API calls use allPages=True so the ONC client handles pagination automatically.
• HTTP requests in BPR_Zero_Cabled_AS.ipynb are cached with requests_cache to avoid redundant downloads during re-runs.
• CRC checksums in Sonardyne serial messages are stripped during parsing.