Our Challenge

As an answer to the 6th of the 17 Sustainable Development Goals of United Nations, Clean Water and Sanitation, SUCARAPP addresses the problem of detecting one of the most deadly of natural calamities - floods.

In the Philippines, clogged canal-related flooding has notable impact on environmental and economical aspects. Most of these were concentrated on the capital city of Manila which is the focus of this project.

Looking at the situation, our team decided to address this problem by Detecting Clogged Canals, Forecasting Flood Probability, and enhancing Community Involvement through the SUCARAPP.

Our Solution

To build an application that can locate probable clogged canals, forecast flood possibilities and bridge citizens to their respective local authorities. Taking clogging and rate of precipitation into account, we constructed our solution:

1. Locate high moisture areas using satellite data
2. Overlay canal network blueprints
3. Analyze the clogged area if it is located beneath canal lines
4. Project exact coordinates of the clogged canals.
5. Plot the analyzed data to a map
6. Integrate rainfall data to the target areas
7. Simulate analysis through machine learning
8. Integrate processed data into the application
9. Design user-friendly application that anyone can easily enjoy
10. Bridge the local citizens to their local authorities by having a reporting system

How it works

The application relies on this 2 data sets:

Space data: Satellite data form Landsat 8, Sentinel-2 Copernicus and Jaxa Global Rainfall Watch (GSMap)

Local Data: Provided Canal Network Blueprints

Locating

Using the spectral data that Landsat 8 and Sentinel 2:Copernicus provides. We Calculated the FAPAR, NDVI and NDWI. Using this calculations and QGIS we are able locate water to the surface. But we need to consider that water can be anywhere so with the help of Canal Blueprints we can separate different water bodies to the water inside the canal lines.

Predicting

Using JAXA Global Rainfall Watch (GSMap) data and the data of the clogged canals we can predict the probability of a flood on the area.

Reporting

Using reporting system of the application that we made, an ordinary citizen can now take a photo, GeoTag the concerned clogged canal on his vicinity. The report will now be sent into our database then the application will update real-time how many times that a clogged canal is reported.

Space Data

In locating the possible clogged canals we use Landsat 8 (NASA) and Sentinel-2 Copernicus (ESA). We use FAPAR, NDVI and NDWI to know the stagnant water, soil moisture and water vapor. Vegetation indices are calculated using FAPAR and NDVI calculations then water indices is derived from NDWI. When determined the readings was done long-lat is extracted through QGIS.

Spectral Bands used:

• Landsat 8 - Green
• Landsat 8 - Red
• Landsat 8 - Near-Infrared(NIR)
• Sentinel - Green
• Sentinel- Red
• Sentinel - Near-Infrared(NIR)

Calculation Indices:

• NDVI,FAPAR = (NIR - Red)/(NIR + Red)
• NDWI = (Green - NIR)/(Green + NIR)

Software

QGIS for determining the location, Figma for Prototype, Google Earth Engine for satellite data, Mapbox used for mapping.

SUCARApp Prototype

References

DETECTION OF CLOGGED CANALS:

AEDES Project, The researcher propose to forecast mosquitos breeding places using Space Data, Nasa 2018 Global Winner, Dominic Vincent D. Ligot, Mark Toledo, Frances Claire Tayco, Jansen Lopez here

NDWI - http://ceeserver.cee.cornell.edu/wdp2/cee6150/Readings/Gao_1996_RSE_58_257-266_NDWI.pdf

NDVI- https://earthobservatory.nasa.gov/features/MeasuringVegetation/measuring_vegetation_2.php

FLOOD FACTS, PREDICTION AND MONITORING:

Mosavi, A., Ozturk, P., & Chau, K. (2018). Flood Prediction Using Machine Learning Models: Literature Review. Water, 10(11), 1536. doi:10.3390/w10111536 

Arnaud, P., Bouvier, C., Cisneros, L., & Dominguez, R. (2002). Influence of rainfall spatial variability on flood prediction. Journal of Hydrology, 260(1-4), 216–230. doi:10.1016/s0022-1694(01)00611-4

Lagmay, A. M., Mendoza, J., Cipriano, F., Delmendo, P. A., Lacsamana, M. N., Moises, M. A., … Tingin, N. E. (2017). Street floods in Metro Manila and possible solutions. Journal of Environmental Sciences, 59, 39–47. doi:10.1016/j.jes.2017.03.004 

https://sharaku.eorc.jaxa.jp/GSMaP/

https://www.untvweb.com/news/clogged-drainage-systems-delay-flood-pumping-operations-mmda/

https://center.noah.up.edu.ph/list-of-flood-prone-areas-in-metro-manila/

https://www.nssl.noaa.gov/education/svrwx101/floods/detection/

https://www.nssl.noaa.gov/education/svrwx101/floods/forecasting/

https://www.nationalgeographic.com/environment/natural-disasters/floods/

https://www.un.org/sustainabledevelopment/water-and-sanitation/

https://www.officialgazette.gov.ph/how-to-make-sense-of-pagasas-color-coded-warning-signals/

http://bagong.pagasa.dost.gov.ph/learnings/legend

https://www.gmanetwork.com/news/scitech/science/267638/the-color-of-danger-pagasa-s-new-rainfall-and-flood-warning-system/story/