| | Kathmandu Valley with optical images - General overview
|  | Kathmandu Valley as seen from Landsat ETM on 27 December 2001 | |
The Valley
You will certainly have heard about Kathmandu as the capital of Nepal. However, to know more about its geographical situation, it is important to know exactly where it is located. It is like drawing a virtual map in your mind.
Therefore, if you have not already done so, have a look at your school atlas. Search the index of placenames for Kathmandu and study the geographical situation. Note the mountains which surround the city, the rivers which flow through the city, the lakes, deserts, forests and glaciers near the city, the nearest towns, and the location of Kathmandu within Nepal. You should also note the neighbouring states.
All this information will help you understand and review the information provided in the case studies.
The Landsat image shows the Kathmandu Valley with the homonymous capital of Nepal in the centre. The valley is about 25 km long and 15 km wide, with an average elevation of 1,200 m above sea level. There are more than 60 towns and villages within the valley. The largest one, Nepal's capital Kathmandu, is clearly visible in the centre of the image. To the East of the city the runways of Tribhuvan International Airport can been seen, and the city of Patan is visible to the South-East. The city is surrounded by agricultural land. The high mountains of the Himalayas are situated in the North. Numerous rivers and streams like the Bagmati, clearly visible in the city centre, cross the valley, arising in the mountains and flowing South into the Ganges.
Locate the image in an atlas or a regional map of Kathmandu and the valley. Name the geographical features, such as the cities, rivers, and mountains.
Study the different colours of the image and allocate them to geographical entities (surface coverage classes).
Try to divide the image into different elevation zones.
Observe the forested regions. What is the physical characteristic of all forested regions in relation to the topography?
Where are other areas of lush vegetation (almost linear)? What geographical feature causes this vegetation?
What are the reasons for the distribution of forested areas in the image?
| | | Landsat 7 ETM |
Description of Landsat data
The US satellite Landsat 7 ETM is NASA's 6th operational Earth observation satellite of a series (the launch of Landsat 6 failed in 1993) dating back to 1972. The Landsat system therefore constitutes the longest continuous record of the Earth's surface. The main tasks of all Landsat satellites are environmental monitoring, disaster assessment, land use and regional planning, cartography, range management, and oil and mineral exploration.
Its mode of operation has been continuously improved. Today, the Landsat ETM features 8 channels, from visible light (channels 1,2,3) to near and middle infrared (channels 4,5,7) and thermal radiation (channel 6). The ground resolution is 15 metres for the panchromatic channel (8) and 30 metres for channels 1 to 5 and 7. The thermal channel 6 provides a resolution of 80 metres. The Landsat 7 satellite operates in a circular, sun-synchronous, near polar orbit.
See Technical data of Landsat bands (PDF)
| | Natural colour combination of Kathmandu using bands 3,2,1 | |
Multispectral Image Combination
Before starting the exercises you have to download the images required.
Save the images to your hard disk in a folder named Kathmandu.
In this exercise we will use Landsat 7 ETM images acquired on 27 December 2001, during the dry season. The ground resolution of the images is 30 m/pixel.
Human vision is optimised to interpret a coloured world. This makes it difficult for the human eye to select different features on the Earth´s surface in a greyscale image. It is therefore useful to combine 3 greyscale Landsat images into one RGB colour image.
RGB in a natural-colour combination means red (band 3), green (band 2), and blue (band 1), and uses the physical principles of the additive colour system.
The colours of the different features depend on the bands selected for the combination, because every object has its own spectral signature. Different combinations allow different renderings of the same feature. We will try out some of these in the next exercise.
| |
1
