This document presents a collection of selected papers produced for and discussed at the Second International Conference on Early Warning, in Bonn, Germany, in 2003, and four regional conferences in Africa, Asia/Pacific, America and Europe. The conferences focused on integrating early warning into sustainable development policy. The document notes the failure of scientists, policymakers, local authorities and other relevant actors to use early warning systems efficiently, and makes suggestions for improvement. It highlights challenges, lessons learnt and possible trajectories for further development of early warning systems, and outlines key steps towards strengthening frameworks, finding resources, and designating responsibilities. Key areas for action include improving data collection, ensuring that warning systems focus on people and achieve gender balance, and creating platforms for early warning dialogue.

This article describes how a network of sensors, linked by software and the Internet, can provide an automatic satellite-based surveillance system for disasters such as volcanoes, wildfires and flooding.

The system, or 'Sensorweb', uses data from low resolution, high frequency sensors to trigger imagery from high resolution instruments. The low-resolution data is collected regularly (twice a day or more) from instruments such as NASA's Moderate Resolution Imaging Spectroradiometer (MODIS).

Anomalies, such as hotspots in the case of fires and volcanoes or surface water for floods, are automatically detected. The SensorWeb then sends a request to a higher resolution satellite such as Hyperion, which is very sensitive in the infrared spectrum, to request data over the area of interest. These data can then provide disaster managers with early warnings of adverse events.

This article, written by Chris Funk of the US GS, outlines how the Famine Early Warning Systems Network (FEWS NET) uses satellite data and statistical forecasts to provide early warnings of potential droughts in sub-Saharan Africa.

Funk outlines the role of satellite data in FEWS NET at all stages of the crop-growing season — from scenario building before the season to calculating the water balance during it and assessing yields at the end. Focusing on food security outlooks for East Africa in late 2008, Funk describes how data from NASA's Aqua and QuikSCAT satellites can be used to track moisture and wind conditions over the Indian Ocean and Africa, and how these help anticipate hydrologic conditions in the future to predict shortfalls.

Funk emphasises the need to combine such data with socio-economic analyses of, for example, crop prices, grain stores, political conditions and agricultural inputs. This will help maximise the accuracy and effectiveness of early warnings of drought and food shortages.

This review article describes how remote sensing data can and are being used to map and monitor disasters such as earthquakes, volcanic eruptions, landslides, flooding and wildfire.

The authors summarise the main satellites and sensors used in disaster monitoring and their characteristics. They also discuss in more detail the data and techniques used for individual types of disaster, outlining the advantages and drawbacks to each. In particular, they describe the methods most commonly used to analyse optical, thermal, radar and LiDAR (Light Detection and Ranging) data.

The authors summarise ongoing initiatives using remote sensing data for disaster management, including Sentinel Asia and the International Charter on Space and Major Disasters. Potentially useful emerging systems such as the Disaster Monitoring Constellation are briefly discussed.

The authors of this article analysed simulation results from a regional climate model for the northern Indian Ocean to predict likely changes in the strength and frequency of tropical cyclones in the Bay of Bengal from 2041–2060.

They find that rising concentrations of greenhouse gases will lead to more frequent cyclones in the region, particularly during the post-monsoon period. In addition, the number of intense cyclones and storm surges will increase. These results are consistent with other trend analyses that show intensification of cyclones in the bay during the last century.

But the research described in this paper only deals with simulations from one future climate scenario. To obtain better regional climate projections, the authors suggest it is necessary to examine simulations from more scenarios.