WATER SCIENCE AND PUBLIC POLICY

IWP’s work in this area focuses on the interaction of science, engineering and public policy. Projects include:

  • Flood hydrology and the effectiveness of engineering interventions  
  • River catchment water, sediment, nutrients and pollutant flows 
  • Extreme floods and droughts: probability, frequency and trends. 
extreme-flood---water-science-&-public-policy

Summary

Publicly Available and Publicly Sourced Data Analytics Solutions for Water Challenges (WATCHA) is a $180,000 grant from the Institute of Data Science (NUS) supporting research on policy-relevant analysis of publicly available data. The project brings together a team from the Institute of Water Policy and NUS Faculty of Engineering, led by Professor Stephane Bressan of the NUS School of Computing. The research involves the design of a mobile app and data analytics that will allow for citizens to interpret and share water quality data generated using a portable testing kit.

The app will provide citizens with guidance on safety risks while anonymising and mapping the data for use by public authorities. Senior research fellow Olivia Jensen is a co-investigator on the project. Her focus is the design and implementation of the crowd-sourcing experiment to test for the validity and reliability of citizen-generated water quality data. The experiment is being conducted in Singapore and Jakarta.


Research Cluster

Water Science and Public Policy


Researchers

Olivia Jensen

Summary

Design of water infrastructure and water policies are usually based on projected needs and measured historical data that span typically only a few decades. The chance of extreme events being included in these data is rare. Assuming the measured data is representative for the system and extrapolating these data to get values for extremes can result in inefficient systems and even amplify extreme events. To develop robust, resilient and disaster-proof water infrastructure and water policies different types of data, assumptions and models are needed. One source of alternative data are paleo-climatological studies that investigate the past climate using alternative methods such as tree-ring measurements, analysis of sediment depositions, and investigation of historical data.


Objectives

The overarching aim of this study is to investigate if paleo-climatic information on floods and droughts would make a difference in short-term water management decisions and long-term water policy decisions.

A hydro-economic simulation model of the Ping/Chao Praya river basin in Thailand will be developed to analyse dam operations, water policies, and emergency management using only measured data records, and using both measured and paleo-climatological data. The results will indicate the value of including alternative sources of data and if changes in dam operations or water policies would be needed.


Research Cluster

Water Economics, Water Science and Public Policy


Researchers

Joost Buurman

Summary

In August 2008 the embankments of the Koshi river in Bihar State, India, broke causing massive flooding of a large part of northern Bihar. Embankments provide flood protection in many river catchments. However, river embankments have been controversial as rivers may have high sediment loads that shallow the channels between embankments. Constraining the river by embankments could actually increase flood risks.

In addition, some benefits of periodical floods, such as fertilisation of agricultural lands, are lost when embankments are in place. Furthermore, the presence of embankments could make people more complacent towards flood risks, which in itself can increase flood risks. Once a flood occurs, political processes, such as public pressure for action and compensation, could lead to ever increasing investments in embankment projects, thereby increasing risks and flood losses.


Objectives

Are embankments the best possible solution for flood risk management? Key research questions include the following aspects:

  1. What impacts have embankments had on river dynamics?
  2. Do embankments reduce perceived flood risk? What is the narrative content of the “public pressure for action” that has already been noted above?
  3. What are the costs and benefits of river embankments from the perspectives of different stakeholders?
  4. How can non-traditional sources of flood data help flood risk estimation?
  5. What are the reforms and institutional changes required to strengthen embankment governance?

Research Cluster

Water Economics, Water Governance, Water Science and Public Policy


Researchers

Robert Wasson

Summary

This is a collaborative research between the Institute of Water Policy and Assistant Professor Winston Chow from the Department of Geography (NUS). In this study, we (i.) quantified and identified drought episodes using the Palmer Drought Severity Index (PDSI) in the neighbouring regions of Singapore and Johor, Malaysia, and (ii.) qualitatively examined each region's drought impacts and consequent responses through archival research over the past fifty years.

The data indicate that both frequencies and intensities of drought episodes in Singapore and Johor have increased over time, suggesting greater exposure to this hazard. However, there are notable variations in drought impacts in Singapore and Johor, and how each region addresses water resource management to drought with varying degrees of success.

Despite the geographical proximity, significant variations in regional adaptive capacities suggest that different drought vulnerabilities exist. The efficacy of drought responses over different time scales was discussed. Finally, a combination of demand- and supply-side policies was suggested, especially in the long-term, to reduce vulnerability to this droughts.


Objectives

  1. Quantified and identified drought episodes using the Palmer Drought Severity Index (PDSI) in the neighbouring regions of Singapore and Johor, Malaysia.
  2. Qualitatively examined each region's drought impacts and consequent responses through archival research over the past fifty years.

Research Cluster

Water Science and Public Policy


Researchers

Winston Chow, Joon Chuah

Summary

Droughts are complex natural hazards that are difficult to identify, measure and manage, as they are context-specific phenomena and different water users are affected differently. However, drought is a hazard for humans only if there are socio-economic impacts: a meteorological drought can exist without socio-economic impacts. This is because socio-economic droughts are not only caused by lack of rain, but also by water management (e.g. use of reservoirs) and the ability of people to cope and recover from it. Yet socio-economic drought is hard to measure because secondary impacts, for instance increases in food prices, may be more important than primary impacts and impacts could continue for a long time, even after the drought has ended. As policy-makers will need to deal with water demands and possibly demands for financial compensation during droughts, they would need to have accurate information on the socio-economic drought.


Objectives

This study aims to contribute to better management of droughts risks. Specifically, it aims to determine when there is a drought from socio-economic perspective. It will do so by comparing hydro-meteorological drought indicators with drought experiences of different types of water users in the Vu Gia-Thu Bon river basin in terms of frequency/duration and severity as described in the research methodology below.

The study will use semi-structured interviews and a survey to find the degree of correspondence between several hydro-meteorological drought indicators and drought experiences of water users in the Vu Gia – Thu Bon river basin in Central Vietnam.


Research Cluster

Water Economics, Water Science and Policy


Researchers

Joost Buurman

Summary

This is the overarching theme for Dr Robert Wasson’s research.

Environmental disasters result from a combination of hazards (e.g. floods, tsunamis, earthquakes, landslides, cyclones, geomagnetic storms) and human vulnerability (e.g. location of human settlements, preparedness for hazards including warning systems, human health, gender). The descriptor ‘natural disasters’ is still used (see the figure below from the Centre for Research on the Epidemiology of Disasters, CRED) even though all environmental disasters have both an environmental component, which is increasingly human-induced (e.g. extreme rainfall as a result of global warming), and a human component. Environmental disasters are globally the most damaging in Asia, with more deaths and people affected but about the same amount of economic damage as in the USA between 2001 and 2010 (CRED), and the total number of reported environmental disasters increased faster than elsewhere until 2000 when worldwide the number appears to have plateaued or slightly declined. The same source shows that between 1990 and 2012, floods in Asia caused damage of USD330 billion (30% of total environmental disaster damage in Asia) compared with USD520 billion (49%) from earthquakes.

In many Western countries, disaster management agencies attempt to identify potential hazards as a basis for preventative policy and action; in Asia the emphasis is mostly focused on response to disasters rather than prevention. The Sendai Framework for Disaster Risk Reduction (SFDRR), which in 2015 replaced the Hyogo Framework for Action (HFA) (2005-2015), stresses that prevention is better and more cost-effective than response alone, although response will always be required as prevention rarely provides sufficient protection against the largest disasters. The SFDRR also gives much more emphasis to knowledge as an indispensible component of Disaster Risk Reduction (DRR).

Given the environmental and human components of disasters it is essential that a cross-disciplinary approach be adopted to produce the knowledge for DRR. Prevention or response designed by only considering vulnerability may fail if hazards are poorly identified or increasing. Equally, DRR designed by only considering hazards is likely to be a waste of resources. In addition, including only one hazard in an analysis can miss the key feature of many disasters, namely that they often involve more than one hazard and often cascade: one disaster triggers others with consequences far into the future. For example, earthquakes can damage nearby settlements and their populations by ground motion but also can trigger landslides that block rivers, producing lakes that burst and flood areas downstream, damaging settlements, human health and components of the economy distant from the earthquake source.

Adopting a cascading multi-hazard approach to analysis is a forward-looking view, yet disasters are often incubated over decades and even centuries as people move to dangerous locations driven by social forces that take no account of possible future disasters. Therefore a backward-looking view is also necessary to understand how vulnerability has evolved. History is also required to determine hazard frequencies, magnitudes, trends and causes. Without both forward-looking and backward-looking views, the knowledge input to DRR is likely to be inadequate and the causes of disasters misunderstood.

This approach can be illustrated in the case of floods in Asia where river floods have about the same economic impact as earthquakes and the damage is apparently increasing (EM-DAT). But it is not clear if the increase is a result of increasing flood hazard frequency and/or magnitude, increasing vulnerability or a combination of these phenomena. Without a better understanding of the causes of the apparent increase of this type of disaster, the design of policy will be severely hampered. In addition, the cascading nature of floods is rarely analyzed.

Even with better knowledge, however, uncertainty will remain. This is particularly the case for the most extreme events, best described as events of high consequence and low probability (HILP). Such events raise questions about both the kind of knowledge that is most useful for policy formulation and the optimal policy response. Policymakers can take a probabilistic approach coupled with a cost-benefit analysis, a worst case approach possibly moderated by identification of the maximum credible event(s), or do nothing. Where the uncertainty of frequency is very high but the hazard is reasonably well known, the best approach may be to build resilience rather than attempt probability-based planning.


Objectives

  1. Reconstruction of long records of floods using measured flows, documentary evidence, and geomorphic information such as sedimentary archives to determine causes and trends.
  2. Reconstruction of histories of vulnerability, particularly identifying the time scales and nature of disaster incubation, to determine causes and trends.
  3. Building system dynamics models of the interactions of these histories to include cascading disasters and account for non-linear and feedback relationships as a knowledge input to policymaking.

Research Cluster

Water Science and Public Policy


Researchers

Robert Wasson