The southwest monsoon arrived in Sri Lanka around May 18thwith a vengeance. According to the public reports, a storm preceding the Cyclone Mora dumped 300 mm to 500 mm in a 24-hours by May 25th causing extensive flooding and landslides in several provinces in the southwest of Sri Lanka. Heavy rains caused flooding in the river basins of Nilwala Ganga, Kalu Ganga, Gin Ganga, and the Kelani Ganga, and landslides in Kalutara, Galle, Matara, and Ratnapura Districts. A recent report indicates that the ensuing tragedy resulted in 224 deaths, 72 non-fatal injuries and 78 missing persons. Over 13,000 houses have been fully or partially damaged. By all accounts, this was a major storm event causing a large loss of life and extensive damage, similar in magnitude to that of the 2004 tsunami.
Disaster management for a major disaster such as the May 2017 flooding and landslides is a challenging task. While there are many lessons to be learned from every disaster, it is not uncommon to receive public complaints regarding the immediate response and lack of planning and preparation. Although it is a useful subject to discuss, this article will not focus on disaster response but rather identify potential inefficiencies in long-term flood management which must be addressed in future planning that is being discussed currently at all levels.
Many major rivers such as Mahaweli, Kelani, and Kalu originate in central hills, eventually discharging their waters to the ocean in large quantities. The contributing areas (also known as “basins”) of the rivers vary in size and include numerous tributaries, minor streams and creeks (often called Oya or Ela in Sinhalese) which flow into the major rivers. Adjacent to the extensive network of rivers and streams in a river basin are paddy fields and wetlands which provide an important function of storing flood waters and slowing down the rapid conveyance of floods downstream. Many areas along this network also include urban centers ranging from villages to major cities which experience flooding occasionally as was observed during the most recent event.
Flooding due to heavy storms is a natural phenomenon in river basins. However, human intervention has made extensive changes to the river basin landscape, exacerbating floods and increasing landslides. First, deforestation in the headwaters and, in some cases, other areas of the river basins have resulted in rapid flow of flood waters downstream, often accompanying large amounts of sediments which deposit in the lower reaches, altering river morphology. Second, the ancient system of conveying flood waters via creeks, streams, paddy fields, and wetlands onto major river courses has been disrupted. Paddy fields and wetlands have been encroached upon to allow and expand human settlements and in many places the roads and highways have blocked the conveyance route of the flood with only a little opening (e.g. small culvert). Third, the floodplains of the lower reaches of the rivers have been encroached similarly for urban settlements. Fourth, the construction of housing at the foothills of mountainous areas susceptible to landslides has made such areas highly vulnerable during heavy rain storms. Perhaps there are other changes in the river basin that have contributed to frequent flooding. Published flood maps of the May 2017 event show that, in many instances, flooding occurred in low lying areas upstream of the major rivers and it is likely that there was little or no capacity for the excess rain water to be stored and/or to flow downstream. To make things worse, there are no designated floodplains, no standards for setting house pad elevationsand roadway elevations are not specially designed to prevent disruption during floods,and this has resulted in unplanned encroachments in the floodplains of river networks. Clearly, an unprecedented opportunity existsfor assessing the conditions of the river basins which may have led to the recent flood and to take necessary measures to alleviate future impacts.
The need of the hour is better floodplain management. It is formally defined as the decision-making process that achieve the wise use of the nation’s floodplains (FEMA, United States). The “wise use” means both reduced flood losses and protection of the natural resources and function of floodplains. In Sri Lanka, proper planning and management of floodplains is clearly lacking, and there is no designated authority with a clear mandate to address a myriad of issues associated with floodplains of the major river basins.Clearly, it is not feasible nor economical to protect against every major flood. For planning purposes, a portion of the floodplain may be defined as the “Special Flood Hazard Area” and it is associated with a “base flood” of particular magnitude (see Figure 1). In the United States, the base flood is a one that has a 1% chance of occurrence in any given year. This is also known as the 100-year flood which unfortunately has been misinterpreted by many, including trained engineers. The correct way to interpret 100-year flood is the flood magnitude whose average arrival time is 100years. It does not refer to the flood which arrives every 100 years. During the May 2017 event, there was a similar misinterpretation of the storm magnitude by some stating that is has not occurred for 200 years. To the knowledge of the writer, the Meteorological Department does not have a 200-year dataset of rainfall to verify this claim. Most likely, this quote meant to say that it is a 200-year rain event which has a 0.5% chance of occurrence in any given year. If a flood of 1% chance of exceedance is chosen for the base flood for regulatory purposes, then the floodplain is typically defined as the area adjoining a river or stream that has been covered by this flood. Trained engineers know how to delineate a floodplain through a detailed analysis,provided thatthe necessary data is available.
Figure 1. Definitions of floodplain, floodway, and flood fringe across a river section
A special flood hazard area(or simply the floodplain) may be divided into different zones. A simple way is to divide it into two zones, one designated as the “floodway”, and the other as the”flood fringe”. The floodway means the channel or river or other watercourse, and the adjacent land areas that must be reserved in order to carry the flood waters efficiently. Typically, no encroachment is allowed in the floodway. Flood fringe is the remainder of the floodplain lying outside the floodway. The base flood elevation estimated by engineers may be used to regulate building construction or encroachments within the floodplain. All property within the floodplain should be required to buy flood insurance against damage from future flooding.
Immediately following the May 2017 floods, published reports suggest that the government is attempting to establish good practices of floodplain management through a new Flood Management Act which will replace the Flood Protection Ordinance established in 1924. Apparently, this new act provides for declaration of flood areas, flood forecasting, preparation of flood maps, infrastructure improvements, and collective flood management efforts. This act, if followed through with proper funding, empowerment, and training, will be an excellent first step. Ironically, the 1924 ordinance had provisions for repayment (perhaps a form of insurance) by the land owners. The exact language is “The cost of preparing a scheme, executing works, and payment of compensation under the provisions of this Ordinance shall be defrayed out of such funds as may be provided for the purpose by Parliament, and the amount of such cost, together with interest thereon at the rate of six per centum per annum, shall be repaid by a yearly rate on all lands within the flood area.”The flood area and the repayment sound more like the floodplain and insurance scheme in the present context. It is interesting to note that, for the nearly 100years the 1924 ordinance has been on the books andthere is no evidence of implementing its requirements.
The central government consists of a large number of ministries and many line-agencies associated with a particular discipline. In the case of water resources management, these agencies are distributed among several ministries. This poses a coordination problem, particularly during a disaster such as the one which occurred in May 2017. To the knowledge of the writer, planning for flood protection and landslides is largely handled by the Irrigation Department, the Sri Lanka Reclamation and Development Corporation (SLLRDC), the National Building Research Organization (NBRO) supported by the Department of Meteorology, Survey Department and others and clearly they are distributed across several ministries.
The Disaster Management Center has its own ministry and it is dependent on the information provided by numerous agencies across the government. Recently, many provincial councils appear to have established their own “Irrigation Department” which is a positive step toward decentralizing technical activities. Ironically, none of these line-agencies have the word “flood” in their name and their founding functionsare largely water supply and land reclamation. The decentralization of the flood management function, which appears to have a secondary status, across many ministries create inefficiencies in coordination and turf battles. One solution adopted by many countries is to align the line-agencies associated within a discipline under one cabinet minister who will be responsible for efficient coordination among them.
Another approach is to establish a separate entity for floodplain management at the central government level populated by qualified professionals across line-agencies. Such an organization must be empowered with clear guidance and training without political influence. This entity needs to coordinate planning and management for floods with professionals from other governmental agencies and provincial councils. It should take the lead role in providing data and technical assistance, and capacity building. The administration of the Mahaweli River Basin through its own ministry and line-agencies is one example where coordination within a single river basin may be facilitated. However, its function should include a broader role for floodplain management. Similar authorities may be established for other large river basins in the country. Clearly, there are multiple approaches that may be used to promote coordination efficiency and with an unselfish political will that can be achieved. Efficient coordination among agencies is clearly an issue that must be addressed.
Funding of flood management projects is another key governance issue. There are many projects underway for key areas such as the lower basin of the Kelani River but they are funded largely by international donors such as the World Bank. Without adequate funding, the line-agencies may not be successful in acquiring technology and data and the capacity-building necessary for making progress in managing the nation’s floodplains. It is time now that the nation recognizes that the economic development through major infrastructure projects cannot abandon the sustainability of our habitat and its environment without equal attention through governance and funding. Let us not jump into building dams and reservoirs without any effort to understand the overall problem and the necessary solutions. Nonstructural solutions through proper floodplain management should precede construction of large infrastructure.
Data and Technology
The data associated with a flood is crucial to its management. They include topographic elevations at a sufficiently high resolution, land use patterns, soils, and, more importantly, the rainfall, water levels and flood discharge magnitudes along rivers and streams. Governmental agencies are making strides in acquiring such data and they are being used for planning of some projects. However, the funding for data acquisition by governmental agencies is depleting and lacking. There is a notion that, without assistance from donors, adequate funding cannot be provided to governmental agencies. Once again, the powers that be must recognize the concepts of sustainability and give equal importance to sustainability of water resources by providing adequate funding for monitoring and planning. There is a saying that “take care of the land and the land will take care of its people” and the lack of attention to natural resources through proper planning, monitoring and science while focusing on large-scale economic development through major infrastructure projects will lead to an unsustainable future.
In the present context, herein lies the problem. In the opinion of the writer, there is a data-sharing problem in many developing countries including Sri Lanka. Various data such as topography, rainfall, flood flows, soils, and landuse are not “freely” available to the public. The reality is that the data are available only if you willing to pay significant sums of money. The writerunderstands the need for the governmental agencies to generate revenue for funding their data collection and archival functions, particularly in view of the fact thatthe political leadership directs them to do so. The selling of data has had a significantlynegative impact on progress being made by numerous students and researchers who are qualified and are willing to analyze the data and understand the causes of and prediction of floods. More importantly, the availability of data for computer modeling is seriously lacking and, as a result, the quality of modeling studies has suffered. As a consequence of the high cost of buying necessary data, many have chosen not to acquire them and conduct proper investigations. An unfortunate outcome of this policy is that the data are lying dormant on the shelves or in computers of many agencies without making good use of them. The long-term cost incurred due to lack of scientific studies caused by the lack of affordable data far outweighs the small sums of revenue generated by governmental agencies. It is time that the budgets allocate sufficient funding for data collection and dissemination, and direct agencies to provide the data free,particularly for research. Recently, development of the Right to Information Act should have included the geophysical data collected by various governmental agencies, particularly if such data have no concerns of national security. In most developed countries, data are available freely and often published via internet since there are laws requiring governmental agencies to do so. One such example of this is the Public Records Law in the State of Florida of the United States which states that”all state, county, and municipal records are open for personal inspection and copying by any person.” Achieving progress will require us to get beyond the current data policy and facilitate the free use of data by those who are willing to use them.
Keeping up with technology such as the Geographic Information Systems (GIS) is also essential for making progress in floodplain management. It is encouraging to read the reports that the government is interested in making technological advances in the Meteorology Department by providing high-resolution Doppler weather radar. Radar data is essential for acquiring spatial information on storm events such as the one which occurred in May 2017, particularly since the rain gauge network in Sri Lanka is not adequate for such purposes.Acquiring rainfall measurements and water levels along rivers in real time through telemetry or cellular networks is another area that may need to be enhanced. A centralized authority will need to establish a flood forecasting system using such real-time data and well established computer models of simulating floods in river basins.
One lingering issue for engineers and other professionals conducting flood studies is the lack of capacity building. The writer was involved in teaching flood modeling to staff at SLLRDC many years ago and it is rewarding to see that the staff there is using it for actual projects. More recently, another capacity building effort was conducted by the writer to over 40 engineers through the efforts of Sri Lanka Association for the Advancement of Science (SLAAS). The governmental leaders need to recognize the importance of building capacity among engineers in the country at all governmental levels. The engineers produced by the local universities are extremely talented (best in the country selected at the Advanced Level) and they will make rapid progress in flood studies once they receive proper guidance and training. Once again, we should not wait for a donor agency to provide funding and the capacity building should be supported by agency budgets.
In summary, the writer’s wish is that the May 2017 event is an eye-opener to focus on floodplain management and revisit governance and policies that may have hindered planning and management of floods in the past. Clearly, the solutions established in western countries will not be directly applicable for countries like Sri Lanka. However, we can learn many things from others. Waiting for donors to provide funding for planning will lead to reports collecting dust on shelves. The political leadership must recognize the talent of the local engineers, provide adequate funding, and listen to their home-grown solutions for efficient management of floods in the future. Perhaps next time, we will not be like the monkeys talking about building houses during the storm only to abandon such planning later.
The writer is a graduate of the Faculty of Engineering, University of Peradeniya. He has a Ph.D. degree from Colorado State University and is a professional engineer working in the United States. He received the 2015 Norman Medal of the American Society of Civil Engineers for a technical paper that makes a definitive contribution in engineering.