Towards a robust, open-source logging platform for environmental monitoring in challenging environments: the Riverlabs toolbox

Hydrology is still one of the most data scarce natural sciences. The large number of variables to measure, their extreme spatiotemporal gradients, and the often harsh and hostile environmental conditions all contribute to this issue. This challenge is even more pronounced in remote and extreme environments such as the tropics, and mountain regions, where the need for robust data is most acute.

Many new and emergent technologies can help with building more cost-effective, robust, and versatile hydrological monitoring systems. However, the speed at which these new technologies are being incorporated in commercially available systems is slow and dictated by commercial interests and bottlenecks.

An alternative solution is for scientists to build their own systems using off the shelf components. Open-source hardware and software, such as the Arduino and Raspberry Pi ecosystems, make this increasingly feasible. As a result, a plethora of global initiatives for open-source sensing and logging solutions have emerged.

But despite these new technologies, it remains a major challenge to build open-source solutions that equal the reliability and robustness of the high-end commercial systems that are available on the market. Sharing experiences, best practices, and evidence on the real-world performance of different designs may help with overcoming this bottleneck.

In this contribution, I summarize the experience gained from developing and operating over 300 open-source data loggers, built around the Riverlabs platform. This platform is mostly a compilation of existing open-source hardware and software components and solutions, which were refined further and tweaked for robustness and reliability in extreme environments. Our loggers have been installed in locations as diverse as Arctic Norway, the high Andes of Peru and Chile, the Nepalese and Indian Himalayas, the Somali desert, and the Malaysian rainforest, providing a wide range of real-world test-cases and performances.