The national sentiment about the Indian summer monsoons is rapidly morphing from eager anticipation to severe trepidation, as horrifying floods brutally ravage our lands and our people. The catastrophic floods currently washing across India further underline the importance of timely warnings and the building of resilient communities. Hundreds of people have already lost their lives to floods in 2018 and more than a million have been rendered homeless in their wake. From Jammu and Kashmir in the north to Kerala in the south, and Maharashtra in the west to Assam in the north-east, our fellow citizens across the country have been reeling under the impact of widespread flooding.
While disaster preparedness has evidently improved, given that the number of fatalities
caused by floods of similar magnitudes has declined over the years, what has been
accomplished is not nearly enough to cope with the increasing intensity and frequency of weather-related disasters under a rapidly changing climate. This is evident especially in cascading disasters such as flooding, when the rainfall event often leads to landslides, cutting off transport access and communication in the affected areas. If the downstream
consequences such as waterborne diseases and the mental trauma suffered by flood-affected communities are also considered, floods can be viewed as the single most devastating natural disaster worldwide.
During the initial rescue and response operations, localised information on the whereabouts of flooding is critical in the ensuring of effective regional prioritisation and efficient resource allocation. However, one can intuitively imagine that travelling into flood-affected areas to gather such information during the event is far from safe. Satellite imagery is an attractive and cost-effective alternative to observing the inundated area synoptically. This can facilitate the planning of evacuation strategies and optimise the often limited resources that are available.
Satellite images that are acquired in the visible wavelength range of the electromagnetic spectrum are very similar to the photographs that are produced by the more familiar digital cameras, which render them easy to interpret. These sensors use a passive imaging technology and record the sunlight (visible light) that is reflected from objects, similar to our smartphone cameras. However, sunlight cannot penetrate clouds, which is a huge impediment in the monitoring of floods during the monsoons. Active imaging involves shining a beam of light on the object of interest and recording the returns, which is similar to the flash on a common digital camera. Such technology is effective in the capturing of flood inundation, because emitted microwave energy can penetrate the clouds due to its longer wavelength.
There is a vast scientific consensus that radar (active microwave) satellites are preferable in flood monitoring. The high cost of acquisition of high-resolution radar images was previously a deterrent to the widespread utilisation of the images in flood management, especially indeveloping regions. However, the International Charter, which recommends open data sharing for disaster management, has greatly improved data availability globally. In addition to this, the free data that is provided by the European Space Agency’s Sentinel satellites has enabled cutting-edge research worldwide.
While radar images are widely accepted as the most reliable resource for flood monitoring, they are notoriously difficult to interpret and are affected by a variety of uncertainties. Urban and vegetated landscapes, which present an inherently large number of potential scatterers to the radar beam, often result in complex images. Therefore, to arrive at any practicable intelligence, flood maps that are generated from radar data through automated methods often require post-processing by experts who are trained in the physical principles of radar backscattering mechanisms. Automatic image processing chains have recommended the use of supporting datasets such as distance or height above the closest river channels, and land- use and cover information to enhance the accuracy of flood mapping. However, in developing countries such as India, such ancillary information with reasonable accuracy is seldom available.
A new semi-automatic single-image flood-mapping algorithm that has been proposed by
Antara Dasgupta and others from the IITBMonash Research Academy might be a viable
solution. The algorithm explicitly utilises patterns of the radar backscatter, which are
observed in the image, in addition to the recorded backscatter itself. Specific arrangements of backscatter values in the image are first identified and then optimised by using advanced mathematical techniques to amplify the information content that is used in flood identification. Finally, a fuzzy machine learning algorithm is used to classify the image into flooded and non-flooded areas, which also expresses the level of confidence in the flood mapping at each pixel.
Validating flood maps that are generated by using this technique against aerial photographs demonstrated an improvement of almost 54 % in some areas over traditional methods. These results are encouraging as the validation zone also included a notable portion of urban land-use. Urban landforms are, perhaps, the most challenging in radar-based flood detection and, arguably, the most crucial from the perspective of flood management. In a developing country such as India, where large portions of the population live in urban settlements on flood plains that are frequently flooded, improving flood mapping accuracy from a single image could potentially save hundreds of millions of rupees worth of public money.
This research constitutes the first part of Ms Dasgupta’s PhD project titled, ‘Towards a
Comprehensive Data Assimilation Framework for Operational Hydrodynamic Flood
Forecasting’. Her research strives to integrate all the seemingly disparate sources of flood information presently available, such as satellite and crowd-sourced data, to arrive at more accurate and timely flood forecasts. She is undertaking this research at the
IITBMonash Research Academy, which is a collaboration of IIT Bombay, India and Monash University, Australia, established to strengthen their bilateral scientific relationship.
Ms Dasgupta’s research team includes A/Prof. RAAJ Ramsankaran from IIT Bombay; and
Prof. Jeffrey Walker, Dr Stefania Grimaldi, and A/Prof. Valentijn Pauwels from Monash
University. This article is based on a paper that was published earlier this year: ‘Towards operational SAR-based flood mapping using neuro-fuzzy texture-based approaches’. It was published in Remote Sensing of Environment, which is a highly reputed journal in the field of remote sensing.
This story was written by Antara Dasgupta, and it comprises original, unpublished content.
Comments