-
Notifications
You must be signed in to change notification settings - Fork 3
SitkaNet
Mechanical, Mentor: Emily Pannell
Computer Science: Will Richards
Project on Hiatus
The SitkaNet project is a collaboration with Dr. Josh Roering and Dr. Annette Patton from the Department of Earth Sciences at the University of Oregon to gather and study soil and weather data to better understand the conditions that cause landslides. The goal of the project is to create a customized in-situ sensor system that allows for multiple potential landslide initiation points to be monitored. Existing solutions are costly, bulky, and hard to install, limiting the number of sites that can be instrumented. Each sensor node measures soil moisture and temperature, atmospheric pressure and temperature, humidity, well pressure, and rainfall. The node transmits the data wirelessly over LoRa to a internet connected hub which uploads the data in real time.
Sensors in each Node:
Brand | Model | Quantity | Purpose | Accuracy |
---|---|---|---|---|
METER Group | Teros 11 | 3 | Industry Standard Soil Moisture Sensor (Volumetric Water Content) | Accuracy of ±0.03 m^3/m^3, Range up to 0.70 m^3/m^3 |
Adafruit | STEMMA Soil Sensor | 3 | Low Cost Soil Moisture Sensor (Capacitive Measurement) | returns readings from 200 (very dry) to 2000 (very wet) |
TE Connectivity | MS5803-02ba | 2 | Pressure and Temperature Sensor (mbar and °C) | Pressure accuracy of ±2.5 mbar (from -25 to +80 °C), Temperature accuracy of ±0.8 °C |
Adafruit | SHT-30 | 1 | Humidity Sensor (Relative Humidity) | Accuracy of ±2% relative humidity (RH) |
Adafruit | MMA8451 | 1 | Accelerometer (Gravitational Acceleration) | Range of ±2g to ±8g, Sensitivity of up 1026 to 4096 count/g |
Davis Instruments | Aerocone | 1 | Rainfall Tipping Bucket | 0.01 in/0.2 mm per tip, 33.2 in^2 collection area, Accuracy of ±3% of total during rainfall of up to 10 in/hr (250 mm/hr) |
All Components for the Node/Hub found in Bill of Materials (See Additional Logistics)
The SitkaNet system is expected to operate year-round in the relatively wet and cold climate of SE Alaska. Currently, most nodes are placed in the wilderness outside Sitka, AK. There is a possibility for increased transmits during the fall and winter seasons when landslides are more frequent due to wet soil conditions.
Currently working on troubleshooting the deployed V2 nodes and development of V3. Six nodes and two hubs have been deployed in Sitka and are currently being monitored.
Bill of Materials
Power Budget
Inventory Spreadsheet
Each node houses three METER Group Teros 11 soil moisture sensors, three Adafruit STEMMA soil moisture sensors (discontinued on nodes made after June 2021), two TE MS580302BA01 pressure and temperature sensors, an Adafruit SHT30 humidity sensor, a MMA8451 accelerometer, a Davis Instruments rainfall tipping bucket, and the control electronics and the batteries to power everything. The six soil moisture sensors are used to measure the saturation of the soil at different depths. The combination of industry proven Teros sensors and the hobbyist-grade STEMMA sensors allows for a reduction of cost without sacrificing accurate data. The STEMMA sensors were ruggedized and calibrated to the Teros sensors. One of the two pressure sensors and the humidity sensor measure the atmospheric conditions around the node. The other pressure sensor is used as a piezometer at the bottom of a one meter deep well pipe to measure the water table. A tipping bucket provides localized rainfall measurements. The accelerometer is constantly monitoring for any sudden movement to the node, which would indicate the area around the node is involved in a landslide. A microSD card is used to back up data and a real time clock provides timestamps for the data. The electronics are enclosed in a modified Pelican case which provides protection from wildlife and weather and is mounted to the top of a well pipe. The nodes transmit data wirelessly over the LoRa communication protocol and are designed to last over six months on battery power. Every five minutes, the node wakes up from a deep sleep mode, measurements from the sensors are taken, recorded to the microSD card, and transmitted to the hub which uploads the data to this spreadsheet
The node is based around the Arduino-compatible Adafruit Feather M0 with RFM95 LoRa Radio. A Hypnos Board allows sensors to be fully powered down when not in use. Code for the SitkaNet is built using the LOOM project.
Six V2 nodes have been deployed in the field and are collecting data. We have run into some issues with the transmission and upload of data, but the nodes are still measuring and recording data locally.
The team is working on building an educational node for the Sitka Sound Science Center as a tool for teaching how the system works to both SSSC employees and locals interested in landslide science. This educational node is the first of V2.5 which has a redesigned node interior, removes the STEMMA sensors, and allows for easier replacement of batteries. V2.5 will also be run on Loom V3 rather than Loom V2 as with previous versions.
Goals of the SitkaNet project include the following:
Short Term:
- Determine cause of LoRa signaling issues in Sitka
- Update all nodes and hubs in Sitka to Loom V3
- Complete, send, and install educational V2.5 node at the Sitka Sound Science Center
Long Term:
- Integrate 4G or satellite communications
- Establish long term, stable data collection and data sending
- Integrate the SitkaNet system into a greater SE Alaska landslide warning and monitoring database/system
- Deploy the SitkaNet system outside of Alaska
- Project Planning
- Prototyping
- Synthesis
- Lab Testing
- Field Testing
- Finalization/Production
- Poster/Presentation
- Publication
- CRES
- Dendrometer
- Djinn
- eDNA Sampler
- eGreenhouse
- Evaporometer
- FloDar
- HyperRail
- Hypnos
- Isotopic Sampler
- Lilypad
- Loom
- Micro-Aggregating Sewer Sampler
- Mooraca
- OPEnSampler
- Pied Piper
- Rag Guard
- Rain Gauge Calibrator
- RainSavor
- RFID Moisture
- Sap Flow Meter
- SitkaNet
- Slide Sentinel
- Smart Rock
- Spool
- WeatherChimes
- Weed Warden
- Wisp
- Archived Project Blogs