Resilience Scanner

IoT Flood Monitoring and Early Warning System

Solution Overview

TX Intelligence, a subsidiary of Grupo TX, joined forces with AWS and the National Civil Protection System of Panama (SINAPROC) to implement a flood monitoring and early warning solution for the Juan Díaz river basin.[1] This approach integrates data from climate prediction models to allow the issuance of early warnings in cases of potential flooding.[1] The system provides real-time information to any user while offering access to historical data stored for sensitivity analyses and monitoring of changes.[3]

Technical Components

The implementation uses LoRaWAN (Low Energy Wide Area Network) communication protocol, which operates using LoRa radio frequency technology in a free band spectrum.[1] Remote IoT sensors can run for multiple years with AA batteries while transmitting data from up to eight kilometers away.[2] SINAPROC's technical team leveraged extensive studies on the river and its basin to determine the most suitable locations for the different types of sensors, allowing analysis of how rain in the upper part of the basin affects water levels from its source and along its route to create a predictive model based on different meteorological scenarios.[2] The variety of sensors provides redundant datasets, which help identify false positives, such as when debris blocks a waterway and causes water levels to rise artificially.[2] Researchers couple the information from IoT sensors with 3D mapping imagery of Panama City, data that is readily available and stored on the AWS Cloud, enabling Grupo TX to use data analytics to understand historical trends, model, and forecast the potential impact of changing weather patterns.[2] The solution uses multiple AWS products and services including AWS IoT Core, AWS IoT Rule Engine, Amazon Kinesis Data Streams, AWS Lambda, Amazon Kinesis Data Firehose, Amazon Simple Storage Service (Amazon S3), Amazon Timestream, AWS IoT Events, Amazon Connect, Amazon Pinpoint, Amazon CloudFront, Amazon Cognito, Amazon DynamoDB, and Amazon DynamoDB Streams.[2] Predictive river analytics using machine learning can help model river levels using the data collected by IoT devices combined with historical data, making it possible to warn people ahead of time if evacuation is required via a public alert system.[2]

Implementation Details

Working together, SINAPROC, Grupo TX, and AWS developed a proof-of-concept for an early warning system for flooding in the Juan Diaz River Basin.[2] The Juan Díaz river basin covers approximately 322 square kilometers and contains seven rivers that flow through the districts of Panama and San Miguelito before emptying into Panama Bay.[1] The system had to be capable of installing devices in difficult-to-access areas, some at altitudes greater than 700 meters above sea level and lacking basic services such as electricity and cellular signal.[1] A dashboard was created for SINAPROC allowing real-time monitoring of all critical metrics and retrospective analysis of historical data.[1] The project implements a multichannel alert system that notifies various stakeholders, from government authorities to local residents, through phone calls, SMS and emails.[1] The implementation of LoRaWAN and integration with AWS IoT Core enabled efficient monitoring and early warnings in flood-prone areas.[1]

Benefits and Impacts

The dashboard facilitates real-time monitoring and allows for retrospective analysis of historical data, which is invaluable for continuous system improvement.[1] Grupo TX will be able to train an artificial intelligence model to detect variables that a human may not be able to detect, and as time passes, the model will continue to improve and notifications to citizens in danger zones will be sent faster and more effectively.[2] The implementation has had a direct impact on the safety of communities and the preservation of ecosystems.[1]

Climate Adaptation Relevance

The Juan Díaz river basin has become increasingly susceptible to catastrophic flooding.[1] A strategic flood risk assessment of the Tocumen and Tapia basins has been developed to understand the key flood mechanisms, create hazard exposure and risk maps, and determine flood risks in four future scenarios.[4] This technology addresses the urgent need for a system capable of monitoring the conditions of rivers and hydrographic basins in real time while integrating data from climate prediction models for early warning issuance.[1]

Sources