Department of Sustainable Development


Evaluation of cases

Case 3: An Early Warning System


In the context of natural disasters, Early Warning Systems (EWS) are playing an important role by reducing losses and the number of casualties and, therefore, they are being designed and implemented all over the world. In general, EWS can be considered measures that are part of three components: institutions, technology and civil society.

Scientific institutions usually evaluate natural phenomena using technology and can issue an alert about a possible event. Civil defense or civil protection agencies such as national emergency offices or boards then react by alerting the population and, in turn, the population responds by evacuating toward safe havens and by implementing measures to minimize losses. 

From a technical standpoint, the implementation of Early Warning Systems requires sufficient knowledge about the phenomena causing the

disasters and of the risks societies must face as a result of those disasters. This implies acquiring knowledge of the temporal and spatial behavior of such phenomena by modeling physical quantities and by developing procedures to produce the forecasts that are used as part of the EWS. In this context, we look for precursors that can be correlated with the most probable magnitude of the event.

The previous chart illustrates the traditional procedure to develop and implement forecasting mechanisms that are an integral part of any EWS.  In the case of floods, for example, we should consider the intensity and amount of precipitation and river flows as precursors.  Their measurement can provide information with regard to the possible area the flood would cover once hydrologic models have been developed that outline the characteristics of the basin and the respective flood plains.

While developed countries expand the list of precursors through new techniques designed to measure them, developing countries are finding more basic precursors that fit their capacity and limitations.  Continuing with floods as an example, developed countries and some developing countries are implementing Doppler Radar to estimate possible amounts of precipitation and the geographic area a storm may affect. In contrast, some communities in Central America are implementing simple techniques to measure the amount of precipitation using simple rain gauges and staff gauges to measure river levels. However, it is important to acknowledge that advanced techniques provide wider and more precise forecasts than simple techniques but, obviously, at a higher cost.  In this case, the support element becomes critical when deciding which precursors will be monitored as part of the forecasting system.

Dr. Juan Carlos Villagrán did not present a specific case of a Flood Warning System but, instead, made available to users of the Virtual Forum a series of documents of particular interest to those individuals working in this area.  One of the documents made available was The Sinking of the San Antonio Neighborhood: an early warning that was not understood.

This report describes that on the evening of February 22, 2007, a section of the ground collapsed in one sector of the San Antonio neighborhood in Guatemala City, taking by surprise not only the residents of this colorful neighborhood but also the authorities of various municipal agencies, the Municipal Water Company –Empagua-, the National Coordinator for Disaster Mitigation-CONRED, the Institute of Seismology, Volcanology, Meteorology and Hydrology –INSIVUMEH- and others.

Weeks before the event, residents of this neighborhood began to notice tremors and noises caused by the process of underground collapse. Those tremors and noises increased in magnitude and frequency as the day of the tragic event approached. In addition, residents of areas near the sinkhole began to notice the unusual movement and presence of insects. Cockroaches, ants, sompopo ants and earthworms began to migrate from below ground to the surface, entering homes and, in some cases, climbing up the walls. Several people interviewed at the temporary shelter set up by the National Civil Police to house hundreds of residents evacuated as a precaution, said that, one day before the event, all insects in the sinkhole area had already evacuated  to safe ground. As could be expected, dogs and other pets also displayed a sense of foreboding about the situation, becoming more anxious or scared and, in some cases, more aggressive.

Although this is not a flood warning report, it clearly reveals that there were environmental precursors or warning signs that were not understood by the specialists at several agencies responsible for managing this type of situations or by the residents, and that led to the loss of human life that could have been avoided with an appropriate early warning system. In this document, Dr. Villagrán introduces several precursors that have been identified by ethnic groups in Guatemala as associated with various natural threats.

In conclusion, the author says that, in Central America, community based early warning systems operated by volunteers in several basins in the isthmus use two precursors to forecast floods:

  1. Measuring rain amounts makes it possible to determine whether floods are a possible.
  2. Measuring river flow or level at the mid-basin point supplements the rain data.

Combined, the use of these precursors will make it possible to forecast whether the flood plains in the basins where these systems have been implemented will be flooded or not. 

Dr. Villagrán also made available the report, Early Warning Systems to Mitigate the Impact of Natural Phenomena within the Framework of Early Warning Systems in Central America:  an integral vision.
In this document, the author points out that there are two approaches in the development of EWSs:  The first, the traditional approach, has three phases:

  1. Measurement of precursors;
  2. Forecasting events
  3. Warning

The second approach, the approach most in use currently and the approach being promoted by national emergency agencies and by risk management agencies, has four phases:

  1. Measurement of precursors
  2. Forecasting events
  3. Warning
  4. Response in case of emergency


Juan Carlos Villagrán


Presentation on an Early Warning System

Basic notes EWSs

Antigua Declaration Early Warning

San Antonio Neighborhood



Case 1: Floods in the Río Cabra Basin

Case 2: Susquehanna River (Pennsylvania) and San Antonio River (Texas)

Case 3: An Early Warning System