Mounting the hardware for the Honeysuckle Hill weather gateway. Credit: The Quoin team.
This LoRaWAN gateway has a range of up to 10km. Credit: The Quoin team.
This bushfire sensor combines an atmospheric CO2 sensor with a fuel flammability sensor. Credit: The Quoin team.
Honeysuckle Hill gateway complete. Credit: The Quoin team.
This weather station tells us Stocker's Bottom's temperature, humidity, barometric pressure, cumulative rainfall and wind direction. Credit: The Quoin team.
Technology
Weather stations
Soil sensors
Bushfire sensors
Fuel flammability sensors
Partner
Locations
Honeysuckle flats
Big Keys Hill
Stocker’s Bottom
The Quoin
Plateau
Penny Hill
Aim
Deploying environmental sensors across The Quoin to raise alerts for immediate action, inform the property’s ongoing management, understand whether our interventions are successful over time, and predict and prepare for changes in and to the landscape.
Context
The best way to understand a landscape is through regular observation: checking the soil, measuring tree growth, watching for wildlife and gauging the temperature. In this ideal world, one person is responsible for this work, enabling them to become a single source of truth, building knowledge, spotting patterns and figuring out how changes in one ecosystem influence another.
Of course, this is not practical or efficient, especially when your property is 5,000 hectares in size. So, the next-best option is deploying a range of sensors to act as your digital proxies.
Approach
To date, our deployed sensor network includes weather stations, soil moisture sensors, fuel flammability sensors and CO2 sensors. This network is enabled by three LoRaWAN gateways, each with a range of up to 10kms, receiving data from the solar-powered sensors and relaying the data to remote servers via the mobile phone network. We also have feral cat traps, with LoRaWAN sensors, so we know if they’ve been triggered.
Zooming in, our weather stations tell us the localised temperature, humidity, barometric pressure, cumulative rainfall and wind direction at indicative locations throughout the property. We co-locate a soil moisture and soil temperature sensor with each of our weather stations to create a more complete view of the environmental conditions. Localised climate monitoring informs the property’s management, and provides important context for research conducted on the property. For example, if our team is going to tackle an emerging weed challenge at Stocker’s Bottom, they can check if the wind speed is low enough for effective herbicide spraying without driving the 7km from the machinery shed to the field.
Our three bushfire sensor sites combine an atmospheric CO2 sensor (with temperature, barometric pressure, and humidity), paired with a fuel flammability sensor. The fuel flammability sensors help us understand how much moisture remains in the woody debris of the forest floor: we plan to pair this understanding with the vapour pressure deficit (the difference between the amount of moisture that could be in the air, given barometric pressure and temperature, and the amount of moisture that is in the air on a given day) to predict the likelihood and severity of bushfires.
Significance
The significance of all these sensors becomes obvious when you start to compare the data to identify localised weather systems throughout the landscape and map those in our digital twin. For example, wind is a huge driver of moisture loss and plant stress in this landscape, which is why no two of our weather stations are among similar landforms. Seeing how temperature, humidity, wind, rainfall, and soil conditions correlate across dissimilar areas of the property informs our understanding of environmental stresses on flora and fauna.
Equally, these sensors will be invaluable in helping us to understand the context of the research results we’re seeing. We currently have 20 camera traps on rotation, and a research partner has a further 40. These cameras show the where, but overlaying highly localised weather patterns helps us consider the why. Our sensors aren’t just for historical analysis, they’re also used to raise alerts when action is required. Has a new restoration planting reached a dangerously low soil moisture level compared to the speed and temperature of the prevailing wind? If so, it might be time to fill up the water tanker and go give them a drink.
In the long term, we’ll be able to learn from this data too, by pattern-matching and creating predictions. We’ll know that a certain amount of localised rainfall near Stocker’s Bottom, when combined with certain levels of soil moisture, will likely cause flooding, just as we’ll be able to predict when will be the best time each year to manage weeds, build fences and conduct ecological burns.
We can’t be everywhere all at once, but with our ever-increasing network of sensors, we can be where we’re needed.