Weather forecasting


“Know yourself and know your enemy, and victory is guaranteed. Know the terrain and know the weather, and you will have total victory.”

Chinese philosopher Sun Tzu, 4th century BC

Today, as throughout history, many aspects of our lives are governed by the weather. Our well-being and prosperity depends on it, civilisations have prospered or struggled because of it and its direct impact on many social and economic sectors – including public health, agriculture, energy, construction, transportation, tourism, recreation, ecosystems and biodiversity. Of particular consequence are severe weather events, such as hurricanes, tornadoes, flash floods, blizzards, heat waves, droughts, and poor air quality episodes – which impact every nation on Earth and which lead to the loss of tens of thousands of lives annually.

The economic and social benefits of accurate weather forecasts are immense and include improved efficiencies in agricultural systems, optimized planning of energy supply and distribution, and of course ensuring public safety.

The art of weather forecasting began with early civilisations using reoccurring astronomical and meteorological events to help them monitor seasonal changes in the weather. With the formation of regional and global meteorological observation networks in the nineteenth and twentieth centuries, more data became available for observation-based weather forecasting.

As weather and climate know no national boundaries, international cooperation at a global scale is essential for weather forecasting and for sharing its benefits. Established in 1950, the World Meteorological Organization (WMO) became the specialised agency of the United Nations for meteorology (weather and climate), operational hydrology and related geophysical sciences. WMO provides the framework for such international cooperation – with a membership of 187 Member States and Territories.

Numerical weather prediction and observations

Modern weather forecasting techniques use sophisticated, computer-intensive models of the atmosphere. These models obtain an objective forecast of the future state of the atmosphere by solving a set of equations that describe the evolution of variables (such as temperature, wind speed, humidity and pressure) that define its state. The process begins by analysing the current state of the atmosphere – taking a previous short range forecast and using observations to amend this forecast so that the best guess of the current true state of the atmosphere is obtained. A computer model is then run to produce a forecast.

All numerical models of the atmosphere are based upon the same set of governing equations, but may differ in the approximations and assumptions made in the application of these equations, how they are solved and also in the representation of physical processes.

Observations are crucial inputs to these Numerical Weather Prediction (NWP) models. Weather bureaus receive many thousands of observations each day and these are processed, quality controlled, and monitored. Sources include surface data from weather stations, ships, and buoys, as well as radiosondes, aircraft, and satellites. These observations need to be incorporated, or assimilated, into the numerical models so that the information that they contain can be usefully exploited in making the forecasts.

The essential role of Earth observation satellites

Modern technology, particularly computers and weather satellites, and the availability of data provided by coordinated meteorological observing networks, has resulted in enormous improvements in the accuracy of weather forecasting. Satellites, in particular, have given forecasters routine access to observations and data from remote areas of the globe. On 1st April 1960, the polar-orbiting satellite TIROS-1 was launched. Although the spacecraft operated for only 78 days, meteorologists worldwide were ecstatic over the pictures of the Earth and its cloud cover that TIROS relayed back to the ground.

The first picture of Earth from a weather satellite, taken by the TIROS-1 satellite on April 1, 1960. Although primitive in comparison with the images we now receive from satellites, this first picture was a major advance.

Over the past 40 years, satellite sensor technology has advanced enormously. In addition to providing visual images, satellites can also provide data that allow calculation of numerous atmospheric and environmental variables – including temperature and moisture profiles. This is done using a variety of instruments, among them atmospheric sounders, which measure quantities at various levels in atmospheric columns. The data retrieved from spaceborne sounder measurements can be utilised in a similar way as that from radiosonde observations – with the major advantage that the satellite data are more complete spatially, filling in gaps between weather ground stations, which often are hundreds or even thousands of kilometres apart.

Geostationary and polar-orbiting orbits for weather satellites

USA, Europe, Russia, China, Japan amongst others contribute satellites to the Global Observing System of WMO

2-D cloud height maps generated by MISR on the Terra spacecraft

QuikSCAT image of winds on the surface of the Pacific Ocean on 8th January 2004

The impact of satellite observations on weather forecast accuracies during the 1980’s and 1990’s is evident in this chart – particularly for the Southern hemisphere where weather stations are relatively sparse (% scale is the evolution of annual mean forecast skill for the European Centre or Medium Range Weather Forecasting)

Satellite imagery is an everyday feature of TV weather forecasts. The early warnings they provide of hurricanes and severe storms can help save lives and property (images show track of Hurricane Isabel, Sept 2003)

The Precipitation Radar (PR) on TRMM observed the heavy rain in Fukui, Japan on 18th July 2004 which claimed several lives. The 4-D data gave insight into the storm structure and showed that the rain fell from as high as 13 km.

Today, the global system of operational meteorological satellites includes a constellation which is evenly spaced around the equator in geostationary orbit, and at least two further satellites in near-polar orbits. The geostationary satellites fly at an altitude of about 36,000km and each has the capability to provide almost continuous imagery and communications support over a wide region of the planet. Each satellite can generate full earth disc images covering nearly one quarter of the Earth’s surface, day and night.

The polar orbiting satellites fly in much lower orbits, typically at around 850 km. Each polar satellite can typically observe the entire planet twice in each 24-hour period, with better resolution than the geostationary satellites.

These missions provide a wide range of valuable data used for weather forecasting and warnings, including:

  • visible and infra-red imagery of the Earth’s surface and atmosphere: such imagery is usually the only means of obtaining continually updated quantitative information about cloud over a wide area, and is used to determine the movement of the atmosphere – providing the main source of wind information for NWP models; infrared imagery can also be used to determine cloud top temperature, and from this estimates derived of rainfall in tropical regions;
  • atmospheric humidity and temperature profiles: atmospheric sounder instruments generally make passive measurements of the IR or microwave radiation emitted by the atmosphere – from which vertical profiles of humidity and temperature may be obtained – which are at the heart of daily weather forecasts using NWP models and have improved forecasts significantly;
  • ozone concentration profiles: developments are under way to add satellite measurements of ozone (from sounder instruments) as a new NWP model variable – primarily for use as a tracer for information on wind;
  • sea surface temperature estimates: which are required for low level cloud diagnosis (and their discrimination from the sea background) as well as for seasonal to inter-annual forecasts – especially in the tropics;
  • precipitation and liquid water: microwave imagers and sounders on satellite missions provide information on precipitation, and represent the only potential information source of their kind over the oceans – although accurate measurements are difficult to achieve and validation is difficult;
  • sea surface winds: the strength, direction and circulation patterns of the surface wind is of great importance for meteorology and climate studies – including for helping to define storm centres, and for detecting patterns associated with inter-seasonal climate variations such as the El Niño/Southern Oscillation (ENSO), which affects the weather over large parts of the planet; until recently the only source of surface wind data over the oceans were reports from ships, mostly concentrated in a few shipping lanes; since the launch of satellite wind scatterometers a huge quantity of high resolution wind data over the oceans has been available, proving of great value to NWP models and other applications.
In addition to the data from operational satellite series – which are planned, operated, and funded on a continuous basis as essential public services – the world’s space and weather agencies are cooperating to explore the introduction of new satellite-derived information streams into the weather forecasting systems. For example, scientists have used the sensors on the joint US-Japan TRMM mission to peer inside the tropical thunderstorms associated with hurricanes in an attempt to understand which parts of a hurricane produce rainfall and why. Most importantly to people endangered by hurricanes, such satellite missions are adding to the knowledge required to precisely predict the path and intensity of these storms.

Future developments

Planners of the Global Earth Observation System of Systems (GEOSS) described in Part I, have outlined a vision where every country will have the weather information needed to virtually eliminate loss of life and to reduce property damage from severe weather events. The aim is to have a society where weather forecasts are fully used in decision support systems to improve economic efficiency and productivity, as well as environmental protection, through improved longer-range predictions available in probabilistic terms.

The GEOSS Implementation Plan calls for an end-to-end weather information system that provides, to decision makers around the world, timely, reliable and actionable information. This system will have improved in-situ and space-based observations of critical parameters, coordinated and exchanged globally. These will provide input to improved NWP models, with advanced physics capabilities, providing accurate (in location and time) forecasts of severe weather events to new or strengthened regional and local warning centres, allowing rapid and tailored notification to local authorities responsible for protecting people and property.
Given the importance of weather observations, a significant number of the future missions planned by CEOS agencies over the next decade have the objective of providing either improved operational observations for meteorology or new research capabilities.

These will include:

  • finer spatial, temporal, and spectral measurements of atmospheric parameters, allowing more accurate determinations of parameters such as temperature and moisture; this data – from missions such as Aqua (NASA), METOP (EUMETSAT), and NPOESS (NOAA) is expected to lead to substantial improvements in the accuracy of mid- and long-range weather forecasts;
  • the introduction of new series of polar orbiting missions to complement the current NOAA series: the METOP series (EUMETSAT) will be launched from late 2005 and the FY-3 series (China) soon afterwards;
  • new capabilities for monitoring precipitation and cloud properties: by 2007, a constellation of satellites will be in place (comprising Aqua, Aura, CALIPSO, Cloudsat, PARASOL, and OCO) and will fly in orbital formation to gather data needed to evaluate and improve the way clouds are represented in global models; the Global Precipitation Mission (GPM) will provide global observations of precipitation every three hours; and Megha-Tropique (a French-Indian cooperation) will collect data on rain over the tropical oceans;
  • soil moisture measurements: which are important for initialisation of NWP models will be provided by SMOS (ESA) and HYDROS (NASA);
  • global three-dimensional wind-fields: direct measurements will be made from space for the first time in 2007 by the ADM Aeolus mission (ESA) – with the aim of improving weather forecasting and climate research.
    There will be significant challenges in developing models and assimilation techniques to make full use of these new capabilities in operational forecasting systems.