Towards a robust and affordable Automatic Weather Station

Abstract
The frequency and severity of extreme weather events have increased over the last 30 years, making predictability of weather a challenge. Weather extreme events often cause adverse impacts to lives and property. Thus, accurate and timely provision of weather data is becoming crucial to improve the skill of weather prediction and to strengthen resilience to the impacts of the adverse weather conditions. Uganda and many developing countries have challenges in acquiring accurate and timely weather data due to their sparse weather observation networks. The sparse weather observation networks are in part attributed to the high cost of acquiring an Automatic Weather Station (AWS) and limited funding to national meteorological services of the respective countries. The inability of developing countries to manufacture their own AWSs leads to high recurring costs accruing from importation and maintenance. In this study, we propose an AWS based on Wireless Sensor Networks. We plan to design three generations of the AWS prototype, the first being the subject of this paper. The purpose of this paper is therefore to evaluate the first-generation AWS prototype and to propose improvements for the second-generation, based on needs and requirements. Results from the AWS prototype data suggest improving non-functional requirements such as reliability, data accuracy, power consumption and data transmission in order to have an operational AWS. The non-functional requirements combined with cost reduction produces a robust and affordable AWS. Therefore, developing countries like Uganda will be able to acquire the AWSs in reasonable quantities, hence improvement in weather forecasts.
Description
Keywords
Automatic Weather Station, Wireless Sensor Network, Robust and affordable AWS, AWS prototype
Citation
Nsabagwa, M., Byamukama, M., Kondela, E., & Otim, J. S. (2019). Towards a robust and affordable Automatic Weather Station. Development Engineering, 4, 100040. https://doi.org/10.1016/j.deveng.2018.100040