Data Science for Water
The world is facing several problems related to water: flooding, drought, access to potable water, availability of water for agricultural production, pollution of water resources, and an absence of basic sanitation. The public and private sector, profit and non-profit alike, are all working tirelessly to develop tools to improve the decision-making process to better forecast, monitor, control, and prevent (FMCP) the problems mentioned above.
Traditional approaches to FMCP largely coalesce around physical-based modeling techniques and applications. Advances in data science, machine learning, artificial intelligence, on the back of access to large and complete data sets enables the implementation of new approaches that complement traditional efforts to water conservation and sustainability. In short, Data science in water has the potential to bridge gaps between monitoring and physically based forecasting models.
Moreover, we can leverage the power of data visualization and storytelling to facilitate communication between decision-makers and stakeholders. Data analysis enables the effective evaluation of preventive and control measures targeting water-related problems. Finally, physically based models can be enhanced with powerful machine learning models for both classification and clustering analysis to define the current state of water resources and for forecasting the future availability and quality of our water resources.
An example use case of “data science for water” can be found in agriculture. Data science can be used to enhance the development of efficient irrigation and water management techniques to produce more sustainable food crops. Another example is the use of machine learning algorithms to predict wind, rain, evaporation, and soil moisture thereby assisting farmers to determine the ideal time to plant, irrigate and harvest.
By: Tatiana Garcia