Some very preliminary notes for an: Open-Source Climate Logging Array (OSCLA ?)
Please comment below, or join the Arduino forum discussion on this topic…
Objective: cost-effective logging of critical environmental data in remote locations and under canopy or in buffered environments.
Design challenges: The system must be cost effective, to allow roll-out over broad environmental gradients and in developing communities globally. System must be maximally automated, to allow deployment in remote locations where data retrieval must be remotely accessed. System must be flexible to allow deployment of a range of sensors. System must be flexible to allow inclusion of mini-sensor arrays into buffered locations where a full array is not possible.
Standardisation issues: see eg.conservation international’s TEAM climate monitoring protocol…
> air temperature
> soil temperature
> relative humidity
> barometric pressure
> windspeed via anemometer
> photosynthetically active radiation (via LED light sensor)
> leaf phenology (via web cam)
> rainfall (via tipping bucket)
(A Stephenson screen housing for shade temp and humidity will be a must)
Sample rates: System should be capable of logging moderately high sample-rate of sample, eg. every minute or so, and storing data between remote-uploads, eg six-hourly. (4 times per day) more rapidly would be better? Eg hourly?
Power: System will draw power from small compact battery arrays connected to solar panels.
Control: Arduino microprocessors could form the backbone of the system. Arduino has dozens of low cost boards that connect together in a modular framework, allowing the user to quickly create devices for a wide range of applications. On board memory will provide short term data storage. 8 gig USB or SD cards will provide medium-term data storage capabilities.
Sensors: There are some High-quality commercial sensors that are relatively inexpensive (?) and easily available. Most sensors have standard outputs such as analog voltage or digital serial signals that are easily interfaced with the Arduino hardware.
User customization of the Arduino node interface hardware and software means virtually any sensor can be used.
One Arduino logging node can interface with many different types of sensors at the same time by using multiplexers.
• A 16-channel, solar-powered Arduino datalogger node with self-meshing wireless communications and waterproof enclosure costs $150.
(From CRB-CZO Arduino Project webpage: http://www.stroudcenter.org/research/projects/czo/arduino.shtm)
Calibration: Sensors will need to be extensively calibrate over the full range of conditions likely to be encountered.
Loggers will be modular, and the base-stations configurable in three main types:
> Local: downloadable from USB or SD connection
> Regional: Communicate with data repository (server) via 3G cellphone network
> Remote: Communicate via satellite telephone
“Hub and node” system: will also be configurable in a system consisting of a “hub” logger base station and multiple “nodes” that can be deployed into nearby environments unsuitable for a base-station, such as beneath forest canopy, into boulder-fields, into freshwater or marine environments, ice environments (eg glacial crevasses). similar to that deployed here at the Stroud watershed project.
Design obstacles to be overcome here include data retrieval and power supply. Data retrieval from nodes could be achieved via radio links using “Xbee” units, through a “mesh” system. Power could be provided to these nodes using smaller photovoltaic arrays connected to smaller batteries, though this would limit their deployability, and may require extensive cabling, for example, beneath forest canopy.
Outputs: Main interface will be web based, showing loggers individually, plus generating realtime weather layers including.
What are the operational limits of the components?: can they withstand (and more importantly, log accurately) the extremes of temperature and humidity likely to be encountered in deployment?
Data failsafe: Can a system be designed with a long term data storage capacity, for example if the remote data retrieval link fails, allowing data to be retrieved manually after an extended period, for example after a winter in the himalayas? Answer: YES: Each node can be equipped with a removable memory card to store all local sensor data in the unlikely case there is a failure in the mesh network
Webcam capability. Can a system be designed with a camera component to record visual data, for example on forest leaf phenology, snow cover, glacial retreat/advance? Would this need to be time lapse??? (data storage and transfer rates limiting, so not video streaming).
Communication Range: What is the expected range of the mesh system between loggerlets? If it is tens or hundreds of metres, then the density of base-station “mother” logger loggers across an environmentl gradient, eg. a mountain, will need to be quite high. If the communication range is in the order of kilometres, then a lower cost alternative may be possible using limited “mothers” and many loggerlets, with high capability, deployed more densely. The expensive remote data retrieval capabilities could then be limited to the “mothers” freeing up funds for more distributed sensor arrays. Answer GOOD: Self-healingmeshwirelessnetworksarerobustand reliable, with nodes spaced hundreds of meters to several kilometers apart.
Battery life: The logging nodes conserve battery power by sleeping most of the time, then waking periodically to take measurements from the sensors, and then transmitting their data through the mesh back to the base before going back to sleep.
Assembly: What skills are needed to assemble the units? Are all parts freely available worldwide?
Recalibration: What skills, equipment and training are need to re-calibrate the sensors in situ? Are these tasks suitable for local people to be involved?
Maintenance: How much maintenance does such a system need? What is the expected lifespan of such a system. How simple is it to isolate problems with the system?. Which components need regular replacement? Solar panels, batteries, sensors, microprocessor?. What is the coast and transportability of these components?
Printability: Can a system be designed that is largely printable using 3d printing technology like the reprap?
> GPS? In case a node gets lost?
> Seismic sensors?
From Stroud wireless logger array project:
“The Open-Source nature of Arduino means the cost is extremely low compared to similar commercial options. The large user community provides support along with constant innovation and development of new hardware and applications.
There are many different types of Arduino-compatible hardware variations, with several being very well suited for wireless sensor networks and dataloggers.
Low powered simple logging nodes collect and transmit basic sensor data short distances to data radio hubs, and the hubs relay the data to a base station.
High-powered radio units connected to a versatile microprocessor board collect sensor data and transmit long distances via a self-healing mesh network back to base stations
Standalone loggers with just a memory card can be used in areas when there is no radio mesh network.”
(From CRB-CZO Arduino Project webpage: