Effectiveness Vs Power Consumption

Some of the predator monitoring and trapping solutions we are developing require more power than traditional tools. This article explores why we think the effectiveness of tools is more important than power use for most of the applications we are interested in.

Why can't you have low power and high effectiveness/sensitivity?

In all designs there are trade-offs. Scissors are inevitably lighter than a lawn mower. They both have their place but you can’t make a lawn mower as light as scissors so it’s a matter of different trade-offs for different uses. The reason that our thermal camera uses more power is that it has an always-on computer that uses software to analyse motion, making it a more sensitive detector (more on this in previous blog post). We use this design because it is more effective. The next most sensitive tool is a trail camera - designed for large animals - and we have shown that trail cameras fail to detect 60 to 95% of the animals that appear within range.

There are no doubt designs for tools that fall somewhere in between a trail camera and our thermal camera on the power/effectiveness scale, but it’s highly unlikely there will be a low power tool that is also the most sensitive and effective. The following discussion of this trade-off will be relevant for any new predator management tool.

Different use cases require different strategies around power and effectiveness

To explain why we have chosen tool effectiveness and sensitivity over power consumption it is worth considering the various challenges around predator control in New Zealand.

High density predator monitoring

Much of New Zealand has a high density of introduced predators and detection is quite easy with a number of relatively crude tools. There are lots of low power tools that are fine for this use case so high sensitivity doesn’t really matter and low cost/low power makes sense.

Low density predator monitoring

Trying to monitor the last hard to detect predators is quite a different problem and a highly sensitive tool is essential if we want to be sure we have eliminated all predators. Our testing shows the best low power tools miss 60 to 95% of predators in the area. With the addition of sound lures to cover a larger area, the Cacophony Thermal Camera has a much higher detection rate. Even though it is more expensive and uses more power, the cost of detection can actually be lower. A device with a 15 times higher detection rate is essentially equivalent to 15 times better battery life.

Lure and trap experiments

Understanding predators and how different lures and tools perform is a specialised use case where seeing everything that's going on is essential. A tool that misses parts of predator behaviour won’t allow you to optimise the lures and traps for maximum effectiveness. In terms of seeing and understanding predator behaviour, our thermal camera captures full animal behaviour and provides new insights.

Predator suppression (traps)

Most trapping in NZ is about predator suppression rather than total elimination with traps designed to stay out in the bush for years with very low maintenance. With this approach, low power consumption is paramount. The inherent assumption is that the traps will never get rid of all predators so they need to stay in place in perpetuity.

If we had a tool that could remove every predator in an area then there is no need to leave it permanently in place. It can be regularly moved on, at which time batteries can be swapped. It's possible that a modest number of highly effective devices which are moved around will have a lower overall cost than thousands of low power but ineffective tools. At the Cacophony Project we consider total elimination a higher priority than suppression and this is where our focus lies.

Last predator elimination

The last predators in an area are always the hardest to remove. The two most commonly used approaches for doing this are intensive poisoning or active hunting with dogs. If you are looking at a tool to reduce the cost of last predator elimination, then effectiveness is a much more important consideration than power consumption because the last predator elimination approaches currently in use are very expensive. Power is not a significant factor for this use case when compared to effectiveness.

We note that there are claims that current traps have achieved total predator elimination in some areas but this seems unlikely given the relatively insensitive predator detection tools being used.

Reinvasion management tool

For areas that are already predator free, managing reinvasion is a constant battle and a high sensitivity device which can be placed at sites of likely reinvasion (e.g. a beach or near a wharf) would be a powerful tool for these situations. A single reinvasion can cost hundreds of thousands of dollars to deal with so a highly effective detector running on solar power has the potential to dramatically reduce costs. Powering a device using solar means that it can stay in one location long term.

Virtual fence

The goal of a fence is to keep everything out so effectiveness is the most important factor. We envisage two approaches to virtual fences.

The first is a mobile line of devices that is swept through an area to make it 100% predator free (we are still a way off achieving this). In this case, batteries can be swapped when devices are moved.

The other approach is a permanent virtual fence where a site can be chosen based on geographical features and access to solar power. It is then a matter of economics to choose between a virtual fence and a physical barrier. A fence that leaks 60 to 95% of predators is not a fence - effectiveness is more important than power for this application.

Total elimination urban/rural/bush

The ultimate goal of this project is total elimination of all introduced predators. The reason for considering urban, rural and bush separately is that power is much more accessible in urban and rural areas. We can do a lot of testing and development in these areas before focussing on large scale deployment in more remote areas.

Comparison

The graph below shows each of these parts of predator control and how important power is compared to sensitivity/effectiveness. The two parts of the problem where power is most important are high density predator detection and predator suppression. These are the two parts of the NZ predator control challenge that are not currently a focus for us. For all the other parts of the problem, sensitivity is more important than power.

Trends in cost and power

The tools that the Cacophony Project are developing are based on digital technology and the long term trend is digital technology tends to use less power as it develops. At the same time, battery performance and solar power options are getting better and less expensive. This means that power concerns will become less of an issue and that choosing designs with high sensitivity/effectiveness makes even more sense.

Why such a long blog article on power consumption?

We are constantly puzzled by people working to remove introduced predators in New Zealand who are focused on the power consumption of tools. For us, the total cost of achieving an outcome is the most important thing. If we have a more expensive device that is more effective and uses less labour then it can have a lower overall cost. We have shown this to be the case for possum monitoring where a $3000 camera can have a lower cost for possum detection than a 50c chew card.

The ultimate cost advantage of removing all introduced predators is that there is no ongoing need for predator control. We don’t see how New Zealand is going to get to zero predators with ineffective tools so our focus is on creating highly effective tools.

Summary

If your main focus is monitoring areas with lots of predators or you are only interested only predator suppression, then power consumption is probably more important than sensitivity/effectiveness. If your goal is to eliminate all predators then effectiveness is more important than power.