An environmental crisis: science has failed; let us send in the machines

Abstract

The ecological condition of the world's waterways continues to decline under increasing pollution, human land use intensification, and/or demand for water abstraction. This is occurring despite the fact that freshwater ecologists and other water scientists have been investigating these environmental concerns for many years. Freshwater science has made considerable advances understanding the causes of this ecological decline, but we still appear to be further from halting that decline than ever before. Perhaps the scientific solutions are clear but political, social, legal or economic constraints intervene? Irrespective of the reasons, in my opinion freshwater science is failing to deal effectively with this environmental crisis. I believe that artificial intelligence devices and machine learning software may offer potential for dealing with the environmental crisis facing the world's freshwater. There are numerous, free and easy to use software packages that would enable freshwater ecologists to better understand some of the complex, nonlinear relationships in their data, and to potentially make better predictions about the effects of stressors and/or how best to mitigate them. I see a not too distant future where these devices will take over direct management of river reaches to hopefully provide more effective balancing of economic and environmental needs for water. I would like to encourage more scientists to embrace the ease and power of machine learning as a way to better interpret collected data, or at least provide an alternative perspective that may prove useful. WIREs Water 2015, 2:595–600. doi: 10.1002/wat2.1102This article is categorized under: Water and Life > Methods