Principles of Remote Sensing - Airborne and Satellite Remote Sensing

Dr. Anwar Ali
Space Research and Remote Sensing Organization (SPARRSO)
Agargaon, Sher-e-Bangla Nagar
Dhaka, Bangladesh

Introduction

Remote sensing is the technique of gathering information about an object without coming into physical contact with the object.

The first and the immediate examples of remote sensing are the human remote sensing like smelling, hearing and seeing. The most important of them all is the visual observation in which case the eye and the brain are the key components. Visible light reflected or emitted by an object is detected by the eye (a sensor) and transmitted to brain (a high speed real time data processor or computer) to give a visual image of the object.

But human remote sensing has limitations. Although it can see to infinity, it cannot distinguish between things beyond a certain distance. It also cannot see in other ranges of the electromagnetic spectrum like infrared and microwave. It is limited to visible ranges of the spectrum only. However artificial sensors can see through much larger distances and a few other ranges of the spectrum. Technically the term remote sensing, which was coined in the early 1960s by geographers in the office of Naval Research of the USA, is used to see an object from a distance with artificial ‘eyes’ or sensors. In the present day context, the term is most commonly used to observe the earth from platforms in space (aircraft, satellites, etc.).

However, there are a number of ways of gathering information through remote sensing even without the use of satellites. In facts, these practices have been in use even long before the invention of satellites. A few examples are:

Astronomy: Almost the whole subject of astronomy is based on remote sensing.

Cameras: They are in use from long before the satellite remote sensing came into existence.

Marine acoustics: Ultrasonic waves which do not travel far in the atmosphere but travel a large distance in water are useful for bathymetric work, detection of submerged installations, fish, etc. and in underwater communication.

Photography from balloons: Earliest known photograph from a balloon was taken as early as in 1859 in Paris.

Aerial photographs: Wide development and applications were made during World War I and extensive development was done during World War II. Remote sensing in its earliest stages was dealing mainly with aerial photography and photo interpretation.

The advent of satellite in 1957 gave birth to the beginning of satellite remote sensing.

In satellite remote sensing, sensors on board satellites collect data about the earth’s surface and the atmosphere below it and transmit the data/information to satellite ground station for subsequent processing, analysis, interpretation and applications. Remote sensing is comparatively a new field. It encompasses a wide spectrum of subjects like natural sciences, environmental sciences, engineering and technology, social sciences, space science, and planning and management. It is a technique, a science, a technology and an art. It has multidisciplinary involvement. Being a new field, remote sensing has created a great impact throughout the world. The subject is fast developing and is having an ever increasing importance.

The Basic Remote Sensing Principles

The observation of the earth’s surface and the atmosphere below the aircraft or satellite is made by some sensors. These sensors take pictures of the objects below and/or collects information about them and relay the data to the ground station. The taking of pictures is done through the electromagnetic radiation. The whole spectrum of EMR is not used. Some bands of wavelengths are used.

As is well known is physics, energy is transmitted from one place to another in three different ways: conduction, convection and radiation. Every object not at absolute zero temperature radiates energy. This radiation propagates in a wave like fashion. The important characteristics of a wave is its wave length (or frequency), amplitude, direction of propagation and the polarization. The wave length may vary from zero to infinity. On the basis of the wave length (or frequency), radiation is divided into different spectra or bands. Some of these bands are:

Gamma rays X-rays, ultra-violet, visible, infrared, microwave, radio and audio (sound) waves.

Sensors are made responsive to certain wave bands in the visible, infrared and microwave ranges. These bands are mostly used in satellite remote sensing.

The visible radiation extends from a wavelength of about 0.4 micrometre (blue light) to about 0.75 micrometre (red light). Sun is the main source of visible radiation. This radiation is reflected by an object and the reflected radiation travels back to space and is captured by the sensors.

Infrared radiation is not detectable by human eye but by specially designed sensors. Infrared radiation is divided into two regions-near infrared (NIR) and thermal infrared (TIR). Near infrared follows the red light of the visible spectrum of the higher wavelength side. It ranges from approximately 0.75 micrometre to about 1.5 micrometres. Thermal infrared range varies from 3 or 4 micrometres to 12 or 13 micrometres. The NIR is important for vegetation mapping because vegetation reflects strongly in this wavelength. Water is almost a perfect absorber of NIR.

There are three principal reasons why remote sensing is restricted to visible, infrared and microwave regions: (i) amount of radiation, (ii) less atmospheric attenuation and (iii) easy recovery. A substantial amount of radiation should be received by the sensor, otherwise it may not be possible to have a noticeable characteristic of the radiation. Secondly, the emitted/reflected radiation should not be appreciably attenuated by the atmosphere in its journey from the object to the sensor. The attenuation by the atmosphere is a function of wavelength. Some waves are less attenuated and some are strongly attenuated. If any radiation of a particular wavelength is not attenuated, then this wavelength is called a window wavelength (atmospheric window). The third condition is that the data generated by the sensors should be recoverable. The visible, infrared and microwave regions have favourable response to the above-mentioned three conditions.

EMR as means of RS

The satellite sensors receive radiation emitted from or reflected by an object. The measurement of emitted radiation provides the estimation of the temperature of the emitting body. The reflected radiation provides an identification of the nature of the reflecting object, the reflection being a characteristic of the body. It is described as a spectral signature of the body, by which nature of the body can be identified. The reflectance characteristics of a body or its spectral signature is a function of wavelength. By studying the spectral signatures, the nature of a body can be identified. And this is the philosophy behind using EMR in remote sensing.

Passive and active remote sensing

Sensors are of two different types: passive and active. Passive sensors respond to the radiation that is incident upon them. Reflected sunlight and emitted radiation from the earth are among these waves. Active sensors generate their own radiation like microwave and send them downward to the earth. This radiation is reflected back to the sensors. Visible, infrared and microwave radiation are used by passive systems and microwaves by active systems. Active systems operating in the visible and IR region are not flown on satellites, but on aircrafts.

Imaging and non-imaging systems

Sensors can be divided into two categories: imaging and non-imaging. The reflected or emitted radiation can produce images through the sensors of the part of the surface of the earth or the clouds. If the sky is overcast with clouds and images are taken in the visible and infrared wave bands, then the images are of the clouds. Microwaves can see through clouds and take pictures of the earth’s surface. The non-imaging systems in which case no images are formed give information about satellite height, integrated value of certain parameters like surface roughness, wind vector (speed and direction), etc.

Satellite system and sensors

In collecting information about the earth from space, two types of space platforms are mainly used-aircrafts and satellites, although balloons, rockets, etc. are also occasionally used for specific purposes. Aircrafts and satellites have particular advantages and disadvantages which are discussed later. It should be sufficient for the time being to note that aircrafts fly below the satellite heights.

We shall concentrate here on satellites and sensors on board the satellites. The Moon is the earth’s only natural satellite. But there are innumerable artificial satellites that are presently moving around the globe. Each of these satellites has been designed for special purposes. Depending on their uses, the satellites are named differently, e.g.

Weather satellites
Resource satellites
Communication satellites
Defence satellites

The first two are the remote sensing satellites and are open to civilian applications.

Manned & unmanned satellites

Manned satellites are those in which men were on board the satellites. They are usually of short duration and amount of data received from them are very little compared to the demand. Manned satellite programmes were more of prestige concern. They are not worth satellite remote sensing. Unmanned satellites are the ones in which there is no man inside the satellite and they are the ones about which all the discussions relating of satellite remote sensing will be made here.

Period of revolution

Satellites move around the earth along certain orbit above the earth’s surface. For a circular orbit, the period of revolution, T, depends only on the radius r provided the satellite altitude is high enough to be almost free from atmospheric drag.

Satellite orbits and altitudes

Satellites, for a better life time, should be put into a minimum operational heights. Otherwise, because of frictional drag of the atmosphere and gravitational pull of the earth on the satellite, the satellite will re-enter the earth’s sphere of influence and burn out or fall onto the earth. In Table 1 is given the average life of a satellite before re-entry corresponding to the satellite altitudes.

It will be seen that there is a lower limit below which the satellite cannot be operated. Usual designed life time of a satellite is from 1-5 years, which requires an operational altitude of about 450 km.

Polar orbiting and geostationary satellites

Satellite orbits are generally of two types: near-polar and geostationary. The near polar orbiting satellites move in the north-south direction with some inclination with the earth’s equatorial plane. Their usual heights above the earth’s surface is about 800-900 km. At this height, the period of revolution is about 90-100 minutes. If the radius of revolution is increased by increasing the altitude, the period will naturally increase. The moon, which has a period of revolution of about 28 days, has its radius of revolution of about 384,400 km. Obviously somewhere in between these two radii (800-900 km and 384,400 km) there is a radius for which the period will be exactly 24 hrs or a day. This radius is approximately 42,250 km with the corresponding altitude of about 36,000 km. If a satellite is launched to move at such a height over the equator in the same direction of the rotation of the earth, then the satellite’s position will be fixed relative to a position on the earth’s surface. That is, the satellite will be geostationary with respect to the earth. All the communication satellites are geostationary. The near-polar orbiting satellites are made sun-synchronous so that satellites view the same point on the earth’s surface at the same local time.

Orbit/swath/pixel

The polar orbiting satellite moves in a north-south plane making on angle with the earth’s equatorial plan. For half of an orbit, the satellite is in the day period and for another half it is in the night period. The north bound movement is called ascending mode and the south bound trip is called the descending mode. The period of revolution determines the number of orbits required to make one complete coverage of the earth. It is to be mentioned here that a polar orbiting satellite does not move towards east or west, it is made to move in its orbital plane which is fixed in position. The earth beneath the satellite (moving from west to east) exposes its surface to the satellite.

While the satellite moves (north-south, say) the instruments on board scan the earth below across the track (either from left to right or from right to left). The scanning is done by rotating the instrument across the track along a line called scan line. Thus the instrument can see a certain distance across the track. The maximum distance the instrument can see is called swath width. Swath width is measured by the total angular swing of the instrument. It can also be measured by the distance the instrument covers across the scan line. Swath may also be called the general field of view of the instrument.

Instruments on board the satellite record pictures of the earth’s surface by small surface elemental areas called picture element (pixel). The pixel gives the average radiation received from the pixel area. The pixel centre is located on the scan line, covering certain distance along and across it.

Weather Satellites

Polar-orbiting weather satellites

The first artificial satellite Sputnik I was launched on October 4, 1957. About 2.5 years later the United States launched the first operational meteorological satellite TIROS-1 (Television and Infrared Observation Satellite) on April 1, 1960, for weather monitoring. Prior to the space age, meteorologists were collecting upper air data using balloons and rockets, although balloons are still in use. TIROS-1 ushered in a new era in meteorological observations from space. Ten TIROS satellites were launched in the series from 1960-65. This first generation of meteorological satellites was followed by the second generation of ESSA (Environmental Science Services Administration) series from 1966-69. The third generation, TIROS-M series started with the launching of ITOS-1 (ITOS-Improved TIROS Operational system) in January 1970 after which the satellites were termed NOAA (National Oceanic and Atmospheric Administration) satellites. The fourth generation, TIROS-N was introduced into service in 1978 and this series is still continuing. Presently NOAA-11 and NOAA-12 are in orbit.

Bangladesh has been receiving data from the meteorological satellites of USA since 1968 and using them for different purposes particularly for detecting, monitoring and tracking the cyclones forming in the Bay of Bengal. The data received are-also-useful for estimating the wind speed in a cyclone and forecasting its movement or likely place of hitting.

The NOAA satellites operate in a near polar sun-synchronous orbit. The orbital period (time taken to complete one revolution around the globe) is about 102 minutes which produces about 14.1 revolutions per day. Since the number of orbits per day is not an integer, the tracks on the earth’s surface do not repeat itself, although the local solar time of the satellite passage remains essentially unchanged for any latitude.

The polar orbiting TIROS-N/NOAA satellite series provide twice-daily coverage of a region per satellite. Two satellites permit four times coverage with an interval of about 6 hours. The satellite altitude is at about 833 + 50 km.

A number of instruments are there on board a NOAA satellite. They are:

* AVHRR, the advanced very high resolution radiometer;
* HIRS/2, the high resolution infrared radiation sounder;
* SSU, the stratospheric sounding unit;
* MSU, the microwave sounding unit;
* SEM, the space environment monitor; and
* DCS, the ARGOS data collection and platform location system.

Of these the most well known is the AVHRR. It has a ground resolution of 1.1 km at nadir and 4.5 km at the far end of the swath. The AVHRR has a scan angle of 110.8 (± 55.4 from nadir on either side). With this scan angle, it scans about 2700 km across the track, allowing the users to look at the environmental conditions and the temporal/spatial changes over a large area.

The AVHRR has five spectral channels which are given in Table-2 along with their detection functions and primary uses. The AVHRR data are globally and widely, used for a variety of purposes particularly in meteorology (for cyclone detection and forecasting), agriculture (world vegetation mapping, crop condition determination, crop yield estimation), flood and oceanography.

HIRS/2 SSU, and MSU (collectively known as TOVS - TIROS Operational Vertical Sounder) are used for atmospheric data collection. The SEM is used for measuring solar proton alpha particle, electron flux density, etc.

The ARGOS system allows collection of data from land-based and sea-based platforms for hydro-meteorological and oceanographic purposes. These platforms are equipped with necessary instruments to collect the data and transmit them to the NOAA satellite whenever it pases over the platforms. The NOAA satellite ARGOS system receives the data from the platforms and transmits them to the nearby satellite ground station.

Geostationary weather satellites

A geostationary satellite can see about one-third of the globe. So for the whole global coverage, at least three geostationary satellites are needed. About six geostationary satellites are now in orbit. They are given in Table 3.

In addition to NOAA, Bangladesh is receiving data from GMS (Geostationary Meteorological Satellite) of Japan. The currently operated satellite is GMS-4, which was launched in September 1989. The meteorological instrument on board the GMS-4 is VISSR (visible and Infrared Spin Scan Radiometer) with wavelength 0.50-0.75 micrometre (visible) and 10.5-12.5 micrometre (infrared) corresponding to ground resolution of 1.25 km and 5.00 km respectively. Earth images can be taken every 30 minutes.

Landsat Satellites

After seeing the successful operation of the meteorological satellites, the United States initiated a programme in 1967 (Earth Resources Technology Satellite-ERTS programme) to launch a series of satellites for resource monitoring and surveying of the earth. The first ERTS-1 was launched on July 23, 1972 and it operated up to January 6, 1978. Just before the launching of ERTS-B on January 22, 1975, ERTS programme was renamed as Landsat (Land satellite, to distinguish it from the planned seasat oceanographic satellite programme) programme. Then onward ERTS-1 was retroactively named as Landsat-1 and ERTS-B became Landsat-2 at launch. Landsat-2 was retired on January 22, 1980. It may be mentioned here that before launching satellites were designated by alphabets ERTS-A, B, C, D and after successful launching into orbits, they were designated as ERTS-1, 2.

There have so far been five satellites in the Landsat series since 1972: Landsat 1, 2, 3, 4 and 5. Landsat 4 and 5 are currently operational and Landsat 6 which was launched recently ended in a failure. The general characteristics of the Landsat satellite are in Table 4.

The Landsat satellite is placed in a near-polar sun-synchronous orbit at a height of about 918 km for Landsat 1-3 and 687 km for Landsat 4-5. It circles the earth every 103 minutes resulting in 14 orbits per day. The satellite passes within 9° of the earth’s polar axis. It passes overhead at about 1000 hrs local solar time. The satellite is southbound at day time and north bound at night time. The instruments on board the satellite cover a distance of 185 km across the track. This distance between the tracks varies with latitude. It is about 2760 km at the equator and 2100 km at about 40° NS latitudes. As a result, there is a large gap in image coverage between successive orbits in a day. However on every new day, the orbit progresses slightly westward (because of rotation of the earth below the satellite) and the images overlap (degree of overlapping depending on latitudes) with the previous day. It takes about 16 days (18 days for Landsat 1-3) to repeat the orbit.

Landsat sensors

The following are/were the Landsat sensors:

i) RBV (Return Beam Vidicon)
ii) MSS (Multispectral Scanner)
iii) TM (Thematic Mapper)

The RBV system on Landsat 1 and 2 had three television-like cameras to take an image of a ground area of 185 km x 185 km. Landsat 1-3 had 18-days-coverage cycle while Landsat 4 & 5 have 16 days cycle. The first three had RBV and MSS sensors and the last two have MSS and TM as sensors.

The sensor characteristics of RBV and MSS is given in Table 5 and that of TM in Table 6.

RBV sensors

Return Beam Vidicon (RBV) had three cameras on Landsat 1 and 2. The three cameras took pictures of the same area with three different filters to select the spectral bands. Ground resolution was 80 m. RBV on Landsat 3 had two cameras (panchromatic) mounted to view side by side and aligned to cover the same swath width, as the RBV three camera system, with a ground resolution of 40 m.

MSS sensors

Multispectral Scanner (MSS) has four bands-two in the visible and two in the infrared. It scans from west to east and takes about 33 milliseconds to scan 185 km. Six lines are scanned by each band requiring 24 detectors.

Thematic Mapper

Landsat 4 and 5 have an advanced scanner called Thematic Mapper (TM), in addition to MSS. The TM has more spectral, radiometric and geometric sensitivity than its predecessors (RBV and MSS). The spatial resolution of TM is 30 m compared to about 80 m of MSS. The band characteristics and the principal application areas of TM are given in Table 6.

WRS (Worldwide Reference System)

A reference system has been developed to refer to any Landsat scene of the earth’s surface. This is kind of coordinate system. For Landsat imagery this is called Worldwide Reference System (WRS).

Each orbit (with a cycle of 18 days or 16 days) is designated as a path. Along these paths the centres of the individual nominal scene (185 km x 185 km) designate the rows. There are about 119 potential day light scenes per orbit. The scene whose centre is on the equator has been designated as row number 060. The row number decreases northward and increases southward. The row number 001 corresponds to latitude 80 deg 01 min 12 sec north and row number 119 corresponds to south latitude of 80 deg 10 min 12 sees. The WRS consists of 251 paths for Landsat 1, 2 and 3 and 233 paths for Landsat 4 and 5. The less number of paths for Landsat 4/5 compared to Landsat 1/2/3 is because of the lower orbital altitude of the former. There are a total of 119 rows.

The orbital paths are numbered westward (because of west to east rotation of the earth below the satellite) with path number 001 passing through eastern Greenland and South America. The last path number is 251 for Landsat 1/2/3 or 233 for Landsat 4/5. Path 252 (or 234) will correspond to path 001 and cycle will start to repeat again and again.

The paths and rows are designated by three-dight numbers. The intersections between paths and rows correspond to geographic positions at which Landsat scene centres are located. Thus a path and a row number will identify the geographical location of the scene. Example: WRS Landsat 4, 140-050 will indicate a Landsat frame/scene with path number 140 row number 050.

SPOT Satellite

The SPOT (System Probatoire l’Observation de la Terre) satellite, owned by France, is another resource satellite with better spatial resolution than Landsat satellite. It completes 14+5/26 revolutions per day at an altitude of about 822 km at the equator. In 26 days it completes a whole number of revolutions with one complete ground track cycle. The same pattern is repeated again and again like Landsat. The maximum distance between SPOT ground tracks is 108 km at the equator. The SPOT satellite works in the visible and infrared region. It has two identical HRV (High Resolution Visible) instruments featuring high spatial resolution (10 m and 20 m) with four spectral bands corresponding to two spectral modes. The multispectral mode (xs) has three spectral bands in the green (0.50-0.59 micrometre), red (0.61-0.68 micrometre) and NIR (0.79-0.89 micrometre) and resolution of 20 m. The panchromatic mode (P) corresponds to spectral band extending from 0.52-0.73 micrometre with a ground resolution of 10 m. The SPOT has the major advantage of viewing through ± 27 deg relative to the nadir and to the orbital plane resulting in (i) excellent revisit capability and (ii) possibility of acquiring stereopairs. The width of the strip (or swath width) varies between about 60 km for nadir imagery and 80 km for extreme oblique imagery. This capability offers added advantage of more frequent coverage of a particular site. As for example, at the equator, a given region may be accessible on 9 separate occasions during the 26 day cycle with an average of 2.6 days coverage period. At a latitude of 45 degree, a region may be imaged 12 times per orbital cycle, giving an average frequency of coverage of 2.1 days.

IRS (Indian Remote-sensing Satellite)

The Indian Remote-sensing Satellite (IRS) is a polar orbiting Sun-synchronous resource satellite. IRS-1A was launched on March 17, 1988 and IRS-1B on August 29, 1991.

The satellites are placed in a 904 km sun-synchronous polar orbit with an orbital period of 103 minutes. Each of them has a repeat cycle of 22 days, but they are placed in such a way that a repeat cycle of 11 days is obtained.

The IRS satellites include two types of advanced sensors: LISS-I (Linear Imaging Self Scanners) with a resolution of 72.5 m and LISS-IIA and LISS-IIB with a resolution of 36.5 m. LISS-I has a swath of 148 km while the combined swath of both LISS-II is 145 km. The cameras operate in 4 spectral bands in the range 0.45-0.86 micrometres (band 1 0.45-0.52, band 2 0.52-0.59, band 3 0.62-0.68 and band 4 0.77-0.86).

MOS (Marine Observation Satellite)

Marine Observation Satellite (MOS) is a Japanese satellite series for earth observation with particular emphasis on oceanography. MOS-1 was launched on February 19, 1987 and MOS-1b February 7, 1990. Instruments on board MOS include.

(i) MESSR (Multi-spectral Electronic Self-Scanning Radiometer)
(ii) VTIR (Visible and Thermal Infrared Radiometer)
(iii) MSR (Microwave Scanning Radiometer)

Spatial resolution (pixel size)

Spatial resolution, often simply called resolution, of a sensor is the minimum size it can view and take images of. This, as noted earlier, is also called a pixel (picture element). In Landsat satellites, MSS takes about 3240 pixels per line with 2340 lines per scene (of 185 km x 185 km). This resolution is also called IFOV (Instantaneous Field of View) of the sensor. The selection of resolution depends on particular purpose of use. Landsat MSS is adequate but not sufficient for many purposes. For land based analysis with finer scale MSS is not that good. For such a purpose, TM and SPOT are better. For oceanographic work such as temperature mapping and ocean current identification, AVHRR is more than sufficient.

Spectral resolution

The spectral resolution of a sensor is given by its bandwidth. The narrower the band-width, the higher the spectral resolution. In the limit, the highest spectral resolution converges to a single wavelength. But it is not possible to obtain measurements of reflection/emission as a continuous function of wavelength (response) signal-to-noise ratio (S/N), lowering the sensors’ radiometric resolution. The radiometric resolution is determined by the number of discrete levels into which the received-signal value can be divided. The maximum number of quantized levels for a sensor depends on the S/N.

Coverage frequency

There has to be trade-off between the spatial resolution and the frequency of coverage. For the AVHRR which has a resolution of 1000 m, the satellite coverage over an area is at least twice a day. One the contrary, for MSS, the coverage frequency is once or twice in 16 days. For SPOT, it is about 26 days.

Airborne Remote Sensing

Airborne remote sensing uses aircrafts as the space platforms. Aerial photography has been in practice since much before the satellite remote sensing. Even now aerial photographs have not found their total replacement by satellite remote sensing in some particular fields and for some special purposes.

Airborne remote sensing uses both photographic and scanning devices. Cameras are mounted on aircrafts and flown over the area for which the photographs are required. Photographs may be in

black and white
colour
colour infrared

Black & white photography

In the simple black and white photography, a two phase negative-to-positive sequence is employed. The negative and positive materials are typically films and paper prints respectively. Both have a light sensitive photographic emulsion coated into a base or support. The emulsion consists of a thin layer of light sensitive silver iodide crystals or grains held in place by a solidified gelatin. Plastics are used as base materials for films and paper is the material for paper prints. When exposed to light the silver iodide crystals undergo a photochemical reaction process forming an invisible image. After development, these exposed silver salts are reduced to silver grains which appear black, forming a visible image. From this image positive images are printed on paper. If, instead of paper, positives are printed on transparencies, then these images are called diapositives or transparencies.

Colour aerial photography

Many remote sensing applications are currently using colour photography. Colour is a combination of varying degrees of blue, green and red light. The main advantage of colour photography is that human eyes can discriminate many shades of colour compared to black and white. Details of colour photography is not given here. It can be found in any standard text book on remote sensing and aerial photography. In very brief: the colour film has different layers each sensitive to different wavelengths (blue, green and red) and the ultimate product is a colour photo depicting the original colour of the object. Black and white aerial photographs are usually made with either panchromatic films or infrared-sensitive films. The spectral sensitivity of panchromatic films extends over the ultra-violet (0.3-0.4 micrometre) and the visible (0.4-0.7 micrometre) portions of the spectrum. Infrared-sensitive films are sensitive not only to UV and visible but also to reflected infrared region (0.7-0.9 micrometre). The limit to 0.9 micrometre is restricted by the reason that emulsions sensitive to wavelengths beyond this are photochemically unstable. The lower limit is imposed by the fact that energy having wavelengths shorter than 0.4 micrometre is (i) absorbed or scattered by the atmosphere and (ii) glass camera lenses absorb such energy. However photography in UV range and beyond reflected infrared is made only in a very limited case and for special purposes.

Colour infrared photography

In colour infrared photography the film is manufactured in such a way that is records green, red and infrared (to about 0.9 micrometre) energy in contrast to blue, green and red in simple colour photographs. In such a case, the image does not reflect the true colour of the object but a false colour in which blue results from objects reflecting primarily green energy and red colour results from objects reflecting primarily in the infrared. Colour infrared photography was developed during the World War II to detect green painted military targets that were camouflaged to look like vegetation. Healthy vegetation reflects infrared more strongly than the green energy. And objects painted green generally have low infrared reflectance. Thus infrared colour photography can distinguish between natural green and painted green. Now a days colour infrared aerial photography has found extensive applications in almost all the fields. Aerial photography is very expensive, and hence they are not generally taken unless specifically needed.

Scale of aerial photographs

Like scales in maps, aerial photographs have scales. The scale of an aerial photograph is given by

This is a good relation for flat terrain. For an irregular terrain, the scale varies and it is given by

The ground coverage of aerial photographs is dependent on focal length of the camera and the height of the aircraft.

Microwave Remote Sensing

Microwave range lies approximately between wavelengths 1 mm and 1 metre. The microwave remote sensing has the principal advantage that it is a all-weather remote sensing technique. Microwaves can see through clouds, haze or rains.

Radar (Radio Detection and Ranging) is an active microwave sensor. Radar was developed to detect objects and determine their range (position). The principle involves transmitting short bursts or pulses of microwave energy in the direction of interest and recording of the reflected energy from the object. By measuring the return time of the signal after reflection, the distance or range of an object can be determined. Radar system may be both image forming and non-image forming. They may be ground based or space based (aircraft, satellite). The spatial resolution of a radar system is dependent on, among other things, the size of the antenna. For a given wavelength, the larger the antenna, the ligher the resolution.

SLR/SLAR

Most airborne radar remote sensing systems use antennas fixed below the aircraft and pointed to the side. These systems are termed Side Looking Radar (SLR) or Side Looking Airborne Radar (SLAR). They usually have large antenna systems and produce continuous strips of imagery of very large ground areas located adjacent to the aircraft flight line. SLAR was first developed for military purposes in the early 1950s.

SAR

As note earlier, the resolution of a microwave system will improve if a longer antenna is used. The Synthetic Aperture Radar (SAR) principle is based on the principle of ‘synthetically’ generating an effectively long antenna without any physical increase.

The first satellite SAR for civilian applications was on board the Seasat (Sea Satellite) satellite which was launched in June 1978. The satellite had a life time of about 100 days. Seasat had three active and one passive microwave sensors:

i) Radar altimeter, for measuring ocean topography

ii) wind scatterometer, for global wind mapping

iii) SAR, for regional & large scale high resolution surface imaging.

iv) SMMR (Scanning Multichannel Microwave Radiometer) for measuring sea surface temperature, ocean surface wind and extraction of ice age.

SIR-A/SIR-B

The SAR of Seasat was followed by SIR-A (Shuttle Imaging Radar), the first of the three shuttle missions. SIR-B was launched in 1984. Bangladesh participated in SIR-B data calibration. The SIR-C is scheduled for 1994.

ERS-1

The European Space Agency (ESA) launched the ERS-1 (European Remote-sensing Satellite) on July 17, 1991. This satellite has SAR. ERS-2 is to follow in 1994.

JERS-1

Japanese Earth Resources Satellite (JERS-1) launched on February 11, 1992 also contains a microwave instrument SAR.

MOS-1

As already mentioned, MOS satellite also has a microwave instrument.

RADARSAT

The Radar Satellite (RADARSAT) of Canada, scheduled for launch in 1994 will also be a microwave satellite.

ADEOS

Advanced Earth Observing Satellite (ADEOS) of Japan will follow the MOS and JERS in 1996. This will also be a microwave satellite.

Unlike visible and thermal sensors, the microwave remote sensing application may be said to be still in the R & D stage. But it can be said almost unequivocally that the future of remote sensing will be the microwave remote sensing.

Table 1

Satellite altitude (km)	Life before re-entry
250	12 days
500	10 years
600	50 years
1,000	1,000 years
10,000	indefinite

Table 2. Uses of AVHRR Channel

Channel	Spectral interval (micrometre)	Detection Function (radiation deteced)	Primary use
1.	0.58-0.68	Reflected visible solar	Day-time cloud and surface mapping
2.	0.725-1.10	Reflected near-infrared solar	Day-time cloud & surface mapping, land-water
3.	3.55-3.93	Emitted thermal IR Reflected IR	Surface temperatures extreme heat sources and night cloud maps
4.	10.50-11.50	Emitted thermal infrared	SST, day/night cloud mapping
5.	11.50-12.50	Emitted thermal infrared	SST, day/night cloud mapping

Table 3. Geostationary Weather Satellites

Name	Operating country	Longitude
GOES-E	USA	75 deg. W
GOES-W	USA	133 deg. W
Meteosat	European countries	0 deg.
GOMS	Russia	70 deg. E
GMS	Japan	140 deg. E
INSAT-1D	India	83 deg. E

Table 4. GENERAL CHARACTERISTICS OF THE LANDSAT SATELLITES

	Launch date	End of activity	Orbit altitude	Orbital cycle (days)	Sensor
LANDSAT 1	1972	1978	918 km	18	MSS+RBV
LANDSAT 2	1975	1983	918 km	18	MSS+RBV
LANDSAT 3	1978	1983	918 km	18	MSS+RBV
LANDSAT 4	1982	-	687 km	16	MSS+TM
LANDSAT 5	1984	-	687 km	16	MSS+TM

Table 5. The spectral wavelength intervals for RBV and MSS

System	Wavelength (micrometre)	NASA code
RBV	0.475-0.57	Band 1
RBV	0.580-0.680	Band 2
RBV	0.690-0.830	Band 3
RBV	0.505-0.750	panchromatic
MSS	0.5-0.6	Band 4
MSS	0.6-0.7	Band 5
MSS	0.7-0.8	Band 6
MSS	0.8-1.1	Band 7