[spacer height="20px"]

Event detection for water and food security in Mali

Mali is fighting a large number of economic and ecologic challenges. However data is needed to support solutions to these problems. In this project we help the Ministère de L'Energie et L'Eau in Mali with an analysis  of the longer term drivers and effects of floods and droughts for IWRM, by using and combining new online media and remote sensing data sources, The ultimate aim is to improve the sustainable and equitable water allocation, as well as the equitable and efficient water use in urban and peri-urban agriculture in the Inner Niger Delta. The project is supported by the Dutch ViaWater programme.

Realtime (impact) monitoring at the Tanzanian Red Cross Society

For the Tanzanian Red Cross Society, Deltares and FloodTags analysed publicly available online media (including but not restricting to Twitter) for early action and flood response at the Tanzanian Red Cross Society. With the results they now can 1) map and monitor flood related events in real-time (sign-up and request access here) and 2) use historic event data for future impact forecasting. By analysing historic flood and water scarcity data together with data about hydrologic variables such as precipitation and discharge, relations between the two are established and hydrologic variables can be directly translated to expected impact. The results are integrated in a mobile website here.

[spacer height="20px"]Online dashboard to connect to the API

Global Flood MonitorFloodTags developed a Dashboard to access FloodTags’ data easily on the web (also for mobile devices). Via the Dashboard you can online monitor new developments related to floods by searching through the online media content. The Dashboard alerts you about new floods happening (in many countries Twitter is the first platform to hold information on new flood events) and you are able to review new information about those flood events (situational awareness) in real-time.

Please note that all data and functions that you will find in the Dashboard, are also accessible from the API. There you can find the full suite of Floodtags' functions, and connect to them from within your already existing applications or Dashboards (if you have them). Read more.

Flood Observations for Red Cross Philippines

Citizens and comFlood_damage_in_Manila,_Philippines_2012._Photo-_AusAID_(10695722693)munities in Philippines are affected by floods each year. For 2014 alone, the Philipines Red Cross registered 21 floods in 80 different areas in their country. To respond to the floods effectively, disaster managers need to know how the people are affected. What FloodTags and partners will do is collect Twitter data and filter out the most relevant flood observations. Next, we combine it with various relevant (water) information so that we get a clear overview of what is happening on-the-ground during a flood. The results will be used as addition to the volunteer reporting system that is already in place, so that response activities of the Philippine Red Cross can be planned even better. The project is winner of the prestigious Challenge Fund of the Worldbank. Partners in the project are the Red Cross Climate Centre (RCCC), Deltares, VU University Amsterdam and Radboud University Nijmegen. Read more.

Global Flood Monitor on the basis of Twitter

Twitter contains information about new flood events, as observed andKaldari_Nashville_flood_08 reported by citizens, news agencies and otherorganisations. In this project, FloodTags and Deltares developed a flood detection on the basis of Twitter and connected it to an event server at the ID-lab. As a result, disaster management agencies obtain timing and location of new floods as observed from Twitter and take appropriate measures (or obtain additional information from satellite products). To obtain a first global flood overview using Twitter, we were looking for a ‘certain flood’ approach. Meaning that we aimed to minimise false positives while allowing for a fair amount of false negatives... Read more.

[spacer height="20px"]Dengue and Diarrhea forecast for Indonesia

c229beaf-7554-47f9-8886-d7519f76aed2In Indonesia there is little reliable data on waterborne diseases such as dengue and diarrhea. In this project we developed a relation to forecast new outbreaks on the basis of reported floods. For the research we used three datasets: Flood tweets, health tweets and field data from hospitals. After validation, we implemented the resulting forecast in FloodTags' data service so that it can be used by the Ministry of Health of Indonesia to support health measures. The project was part of the "Data Innovation Grant" organised by UN Global Pulse. FloodTags and partners Radboud University and Universitas Bandung (UNPAD) are the winners of this contest together with three other innovative data projects for Indonesia. Read more.

Local newspapers show where Uganda was flooded in the past

Capture2Most extreme weather events are recorded in local newspapers. However it is labor intensive to review all the archives. FloodTags supported RCRCCC by downloading and analysing newspaper-aggregated information about historic floods events. The result is a historical flood map with dates of flood occurrences and links to the relevant news articles. This can now be used by RCRCCC to prepare for future floods through Finance Based Forecasting in Uganda. Read more.

Real-time Flood Maps for Jakarta

FloodTajakarta roundaboutgs and Deltares developed an automated procedure to use the thousands of observations generated by the social media for real-time flood mapping. By applying statistics and using hydrodynamic corrected Digital Elevation Maps, they piloted the method for Jakarta where with 75% accuracy they could simulate flood extent. The real-time flood-extent maps provided a good comparison with ground-truth photographs in most neighbourhoods in Jakarta. Read more.

Tool to build "Flood Classifiers"

Relevancer2How do you know which text messages are relevant and which are not? The Centre of Language Studies of Radboud University and FloodTags developed a tool that classifies text messages using a K-means algorithm. The tool is named "Relevancer". On the basis of initial user input (annotating clusters of content in several rounds of user input), Relevancer estimates the relevance of the text messages and classifies the observations on the basis of the words and language that is used.

Develop with us

FloodTags works together with a growing research community that develops open software for specific tasks within data analysis for water management. If you would like to participate in the research or otherwise be connected to its results, please contact us!

Develop with Us

[spacer height="20px"]