NDVI: Normalised Difference Vegetation Index a.k.a Vegetation From Space

Exercise 1

1. Describe the two NDVI images.

The first image, which is the image taken from SPOT satellite shows vegetation in January 2000 while the second image is taken from SPOT satellite in August 2000. Addition two images are the precipitation map in Latin America, taken in January and August 2000. Based on the first map given, it shows only a small part of dark green area observed, which is located on the North side of the map. The second map shows a larger area with dark green, which is located in the middle of the map. Dark green region shown in both maps indicate that the vegetation in that region is very healthy. With comparison of both maps, it shows that vegetation is very healthy during August instead of January.

With the reference to the addition two images, which are the precipitation maps, it is observed that there are more precipitation in the middle of the continent during August compared to January. This further explains the reason why vegetation is found to be healthier is the region where there are abundant of precipitation during August compared to January.

2. Where do you find the most abundant vegetation?

Most abundant vegetation can be found in dark green region shown in the map, which is Peru, Bolivia and Brazil during August, where there are abundant of precipitation.

3. Where do you find the least abundant vegetation?

In the region of Chile and Argentina, where precipitation in that region is in a range of 20-50mm/quarter.

4. Compare the precipitation maps for the two seasons and try to explain the changes in vegetation from January to August. 

Based on the precipitation map for December to February, it is observed that there is abundant of precipitation on the North of Latin America, as the wind direction is shown to be blowing towards North. Meanwhile, according to the precipitation map for June to August, it shows that there is abundant of precipitation in the middle region of Latin America. Changes in vegetation from January to August is seen by the changes of the health of vegetation. Dark green region in the map indicates that the vegetation is very healthy. Thus, to explain the changes, precipitation brings healthy vegetation. This is due to the basic need of vegetation, water. Abundant amount of precipitation and sunlight helps in the growth of vegetation.

5. Find some areas where there is much variation in the NDVI. By comparing the NDVI with the precipitation maps can you explain the variations?

Brazil and Argentina. In Brazil, it shows a high value of NDVI, while in the Southern part of Argentina, it shows a very low value of NDVI. Both values of NDVI is indicated by dark green and baige colour respectively. The darker the colour, the healthier is the vegetation, and vice versa. Comparing the the precipitation maps, Brazil shows a high precipitation compared to the Southern part of Argentina, which precipitation received is minimum. This variations happened due to the meeting of the needs of vegetation. NDVI shows the health condition of vegetation by obtaining the colour of leaves of vegetation. Green leaves shows healthy vegetation, while baige or light yellow shows unhealthy vegetation. In order for healthy vegetation, water is one of the basic needs. Thus, in Brazil, it receives the maximum precipitation during June to August, which leads to the healthy vegetation conditions. Compared to Southern of Argentina which receives minimum precipitation, vegetation there is very poor.

6. Find some desert areas. Can you explain why precipitation is so low in these areas?

Nothern Peru, Western Boliva, Northern Chile and Southwestern Argentina.

Precipitation is so low in these areas throughout the year is due to the wind patterns. These areas experience changes of wind direction as monsoon season changes throughout the year. Wind in these areas is warmer and drier. The dry air blows through desert regions, which absorbs the moisture and reduce the likelihood of cloud formation and rainfall. High temperatures causes rapid evaporation to take place. Besides, a landscape consisting of mountains around stops moist air from approaching, which then decrease the precipitation.

7. There are some areas with high NDVI values in both seasons. Can you explain why?

Guyana and Paraguay shows high NDVI values in both seasons. This may due to the topography of the area. Both areas are located in low areas, and are not surrounded by any mountains or highlands. Therefore, monsoon which brings rainfall, coming in both ways in different seasons is not blocked by mountains or highlands. Thus, both areas receives sufficient precipitation required for vegetation growth. Thus, both areas have high NDVI values which indicates healthy vegetation.

Exercise 2

8. Observe and describe the variation in the NDVI value during the year 2000.

NDVI value variate from January to December in the year 2000 significantly. From the map given, it is observed that during January, there is only a small region of dark green. Dark green indicates the area where vegetation grows healthily, while blue shows the most healthy vegetation. In the month of February, it is observed that there is a small region with blue spots. This shows a high value of NDVI. Meanwhile, in the month of March to May, we can see that NDVI value increases as the blue region shows an increase in size. In the month of June and July, the blue region spotted increases rapidly, which indicates that the value of NDVI increases too. However, in the month of August, it shows a slow decrease of blue region or dark green, and continue to decrease rapidly in the month of September and in the month of October, it shows no sign of blue or dark green spot. This shows that the value of NDVI increases as blue region increases in size and decreases as blue region or dark green decreases. In the month of October, it shows a very low value of NDVI. In November, NDVI value increases slowly as the darkness of green colour increases in several regions. During December, NDVI value increases as blue region is spotted in the map.

9. In what months is the NDVI at its maximum in the Amazon basin? Can you explain why?

NDVI shows its maximum value in the Amazon basin in the month of July. NDVI value shows the health of vegetation on Earth surface. The higher the value of NDVI, the healthier is the vegetation. Blue colour in the map shows the highest value of NDVI. Based on the precipitation map provided, this is because during June, July and August, there is abundant of precipitation observed. As water is one of the basic needs of vegetation growth, vegetation shows its maximum health during these three months.

10. Observe the December image for the northeastern part of Brazil along the Atlantic coast and inland. Can you explain the vegetation pattern?

The vegetation in the northeastern part of Brazil along the Atlantic coast and inland shows a moderate health. The light green shown in December image indicates that vegetation there is not in an excellent condition. As it goes along the Atlantic coast and inland, until it almost reaches the Eastern part of Brazil, it shows a lighter green compared to the upper part. This shows a lower health condition of vegetation. This may due to the low precipitation of the Eastern part of Brazil.

11. Can you explain the variation in vegetation along the Pacific coast?

Pacific coast have a very low variation in vegetation. This is shown through the baige colour in the map. Along the Pacific coast, it is observed that it is almost desert area, which only have maximum precipitation of 20-50mm/quarter every month. This leads to unsuitable condition for the growth of vegetation, as vegetation requires precipitation for growth. Thus, almost zero vegetation is observed throughout the year.

12. Compare the information on climate as described in the maps in a normal school atlas with the NDVI images. How are the yearly changes of the location of the Inter-Tropical Convergence Zone (ITCZ) reflected in the NDVI images?

Information on climate as described in the maps in a normal school atlas is just the basic information. It shows how climate variate, however, it does not have detailed information as NDVI images do. NDVI images are able to show the region with the maximum or minimum precipitation. In addition, an animation can be played in order to see the variation of climate at once, without needing to have many images on hand at once. Besides, normal school atlas climate information is hard copy, while NDVI is digitised information. This enables user to manipulate the data with applications in order to get the information that he desired.

Inter-tropical convergence Zone (ITCZ) is the area encircling the Earth near the equator, where the northeast and southeast trade winds come together. It is also known as the monsoon trough, when it is drawn into and merges with a monsoonal circulation. ITCZ is formed by vertical motion largely appearing as convective activity of thunderstorms driven by solar heating, which effectively draw air in, which is also known as trade winds. The rising air produces high cloudiness, frequent thunderstorms, and heavy rainfall within the zone. ITCZ changes north and south seasonly with the Sun. As ITCZ changes its location yearly, it is reflected in NDVI images. The growth and health of vegetation is the direct reflection of ITCZ’s effect. ITCZ causes a part of the continent to be extremely dry, while the other part of the continent to be very wet, with thunderstorms. Regions which are extremely dry will have poor vegetation, which is indicated in NDVI images with light green or baige colour of the map. Meanwhile, in regions with wet weather, vegetation shows a healthy condition, which is indicated as dark green in NDVI images.

Figure 1. NDVI Image for January 2000

Figure 2. NDVI Image for August 2000

Figure 3. Case study in South America

Figure 4. Precipitation Map of Latin America

Lineament Mapping.

1. What is greyscale image?

Greyscale is a range of monochromatic shades from black to white, and thus, a greyscale image contains only shades of grey and no other colour. The displayed brightness of a particular pixel is proportional to the pixel digital number, which is related to the intensity of solar radiation reflected by the targets in the pixel and detected by the detector. Panchromatic images is one of the examples of greyscale images, where it consists of only one band.

2. What is false colour composite image?

False colour composite image is the colour of target in the displayed image which does not have any resemblance to its actual colour.

3. Which colour composite you think is the best for lineament mapping? Justify it.

In my opinion, I think that the true colour composite is the best for lineament mapping. The true colour composite consists of three primary colours, which includes red, green and blue. I think that this is the best for lineament mapping because when a multispectral image consists of the three visual primary colour bands, the three bands can be combined to produce a “true colour” image. In a multispectral image, it can be assigned respectively to the red (R), green (G), and blue (B) colours for display. This enables the colours of the resulting colour composite image to resemble closely to what would be observed by the human eyes.

4. What is the difference between .ers and .alg?

.ers files are data files which is just the header file for a “band interleved by line” binary data file. .alg giles are “algorithms” which shows how to manipulate the data. .alg files can be opened as if they were data, or known as virtual dataset.

5. Compare between the two imagery. Describe the landuse change that had occurred between 1990 to 2002.

In the year 1990, there is not much development being carried out, while in the year 2002, there is an obvious development seen from the map. In year 2002, a road linking two places can be observed, and in year 1990, road is not seen. This also indicates that year 2002 shows more landuse changes compared to year 1990.

6. Zoom in and out of the image. Right click on the Intensity Layer. Turn off. Zoom in and out of the image again. Describe the difference/s that you notice in terms of spatial resolution.

With intensity, the image is dark in colour and the road could not be seen clearly. Without intensity, the image is bright in colour, and the road looks clearer and brighter.

7. Which one is the best, use polyline, polygon or rectangle in the Toolbox to digitise the road?

Polyline.

8. Describe the procedure of changing the colour and thickness of the poly line.

Select/Edit points mode is selected and is a box including the whole image is drawn. Many small squares appear and double click on one of the small squares. A table written “Line Style” will appear. Colour of the contour can be change by selecting “Set colour” at the bottom left of the table. Select colour desired and click “ok”. The thickness of the contour lines can be adjusted by selecting column next to “Width”. Select the width desired and close the table. Changes can be seen right away.

9. What is DEM?

DEM is a shortform for Digital Elevation Model. DEM file is a simple, regularly spaced grid of elevation points. It represents terrain elevations for ground positions at regularly spaced horizontal intervals. DEM file is also a middle-state format used in the generation of three-dimensional graphics displaying terrain slope, aspect (direction of slope), and terrain profiles between selected points.

10. What is the lowest and highest elevation in the study area?

The lowest elevation in the study area is 800m and the highest elevation in the study area is 1800m.

Process of Completing This Lab

Figure 1. Processed 1990 in .alg format.

Figure 2. Processed 2002 in .alg format.

Figure 3. Edited River Lines. Changed colour and thickness of river lines.

Figure 4. Edited Road. Changed thickness and colour of the road.

Figure 5. Edited Contour. Changed thickness and colour of contour.

Figure 6. DEM created from Gridding process.

Figure 7. DEM created from Gridding process.

Figure 8. 3-Dimensional layer.

Pop Quiz

Did popquiz today! :D

Pictures.

Figure 1. Mosaic.tiff

Figure 2. Mosaic Aerial Photo

Figure 3. Study Area Polygon

Figure 4. Study Area

Figure 5. Elevation

Figure 6. Slope failed to be produced

Things You Need To Know

1. What is rectifying process?
Rectifying process is the process of shifting pixel locations to remove distortions, and very often the rectification includes georeferencing, because one can both shift the pixels to remove distortion and assign coordinates to those pixels at the same time.

2. What is Ground Control Points (GCP) in rectifying process?

Ground Control Point (GCP) in rectifying process is a specific pixel on an image or location on the surface of the Earth.

3. What is map projection?

Map projection is the attempt to portray the surface of the Earth or a portion of the Earth on a flat surface. Map projection causes distortions of distance, direction, scale and area. In some projection, it minimises distortions in some of the properties mentioned above, by expending maximum errors in other properties. Different map projections will result in different spatial relationships between regions, which also mean that different projections cause different types of distortions.

4. What is geodetic datum?

Geodetic datum is defined as the size and shape of the Earth, the origin and orientation of the coordinate systems (latitude and longitude lines) used to map the Earth. Or, to further describe geodetic datum, it provides a frame of reference for measuring the location on the surface of the Earth. Datums have evolved from a spherical Earth to ellipsoidal models derived from years of satellites measurements. One of the modern geodetic datums is the flat-earth models, which is used for plane surveying, while the complex model is used for international applications. Some characteristics which are completely described by the complex model are the size, shape, orientation, gravity field, and also the angular velocity of the Earth.

**For your information, the first estimation of Earth’s size was made by Aristotle.

5. Can two maps of different map projection can be overlay onto each other? Provide the reason/s.

Two maps of different projection are best not to be overlay onto each other. This is because the Earth’s surface cannot be flattened without distorting geometrical properties. Geographical properties include area, shape, distance and direction. These spatial properties can be preserved individually and in certain combinations on map projections, however, these four basic properties cannot be held true simultaneously. Besides, different projections will cause different distortions, and a suitable projection will be chosen to suit the mapping situation. Therefore, if maps of different projections are to be overlay onto each other, many distortions and errors will occur, which directly decrease the accuracy of the maps to be produced.

6. How many GCP that you had utilised in georeferencing the aerial photos? Provide the table of the GCPs in your report.
I had utilised 7 Ground Control Points (GCP) in georeferencing the aerial photos.

GCPs for File 124

GCPs for File 125

GCPs for 126

7. What are the criteria of a good GCP?

The criteria of a good Ground Control Points (GCP) is when the points chosen are:

i) uniformly distributed

ii) respect the terrain variations in the scene (selecting of points at both highest and lowest elevations)

iii) clearly identifiable points in the image

8. Provide three examples of good features for GCP.

Examples of good features for GCPs:

i) cultural feature

- roads and railroads intersections

- river bridges

- large low buildings (industrial buildings)

- airports

ii) line feature

- well defined edges

- two line features forming intersection with an angle larger than 60 degrees

iii) natural features

- not preferred due to their irregular shapes

- used in areas lacking of suitable cultural features

- examples: forest boundaries, forest paths, forest clearings, river confluence

9. What is the min value of RMSE for each point that was used that you manage to achieve?

The minimum value of Root Mean Square Error (RMSE) for each points that was used that I managed to achieve is 0.03.

10. Why do you think it is difficult to achieve the RMSE of less than 1 in this aerial exercise?

I think that it is very difficult to achieve the RMSE of less than 1 in this aerial exercise due to human errors, such as inaccuracy during selection of GCPs and lack of observations during selection of GCPs.

11. Why it is important to achieve RMSE of less than 1?

RMSE is the square root of the mean or average of the square of all of the error. In Geographic Information System (GIS) dictionary, the RMSE is a measure of the difference between locations that are known and locations that have been interpolated or digitised. It is derived by squaring the differences between known and unknown points, adding those together, dividing that by the number of test points, and then taking the square root of that result. Therefore, it is important to achieve RMSE of less than 1, for it indicates the error occurs during the selections of GCPs on different maps. The smaller is the value of RMSE, the smaller is the error, the higher is the accuracy of the map, provided that the value is less than 1.

12. What is/are the difference/s between polygon, polyline and point in feature type?
Point feature is a geometric element defined by a pair of x-coordinate and y-coordinate. Polyline is a shape defined by one or more paths. Paths here referred to a series of connected segments. If a polyline has more than one part, or known as a multipart polyline, the paths may either branch out or discontinuous. Meanwhile, a polygon is a closed shape defined by a connected sequence of x and y coordinate pairs, where the first and last coordinate pair are the same and all other pairs are unique on a map, while in ArcGIS software, a shape is defined by one or more rings, where a ring is a path that starts and ends at the same point. If more than one ring exist in a polygon, the rings may be separated from one another, or might be inside one another, but they will never overlap one another. Difference between point, polyline and polygon is point does not connect each other, it is just a point of x and y-coordinates, while polyline is an open series of points, from the first till the last point, and lastly, polygon is always closed. There is a line from the first point, which will also be the last point, which makes it a closed shape. A polygon can either be filled by default, or can not be filled. However, a point and polyline can never be filled.

Remote Sensing. Cell Value. Traverse Profile. Scattegrams.

1. How can remote sensing help in searching of MH370?

          Remote sensing can help to search for MH370 over the air and sea due to the application of satellites. Satellites can be used to capture images of the surface of the Earth of where the plane is estimated to have passed by. Images can also be taken in fixed intervals at certain places so that the place is well observed. Images taken can be processed using the application of GIS, which is able to provide the exact coordinates, given that the map of the Earth is present. Besides, remote sensing helps in searching efficiently because it can search at a large area at once and take only a short time. Therefore, remote sensing can save searching time as it can get images of places where people reported to have seen the place without having the rescue team to go in person, which will take more time.

2. Cell value profiles windows. How it works and what it shows.

          Cell value is the value of a particular data in each cell or grid, normally known as pixel as well, of an image. It shows also values of each cell or grid of each band in a multi-bands image. Cell values can be of high resolution or of low resolutions depending on the satellites. By selecting a particular area desired, values of each cells according to bands will be shown. Below is attached with the figure of the values of cells by randomly selected point.

Figure 1 shows the cell values of a randomly selected point.

3. Traverse profiles. How it works and what it shows.

          Traverse shows the profile of different lines. Different lines can be formed according to the desire of the user on what he or she wants to interpret of the image. When a line is formed, a graph will be shown and it can be changed into different  bands, and graphs will change accordingly. Many lines can be formed and line profiles can be viewed and compared easily. Below attached are images of some steps of viewing traverse profiles.

Figure 2 shows map before the line is formed.

To show the line profiles, poly line icon from the tool bar must be selected, then click on two desired points. Once two points are selected and line is formed, a graph will be shown together with the details below it. And it can be changed to different bands.

Figure 3 shows the lines and its profiles.

4. Scattegrams. How it works and what it shows.

          Scattegram image is a profile of data which shows the correlation between different bands. Correlation between two bands can be shown by the plotting of X-axis and Y-axis. The region with red colour in the image represents the correlation of two selected bands. The larger the red region, the more is the correlation between the two bands. Below attached are pictures showing how scattegram image works.

Figure 4 shows the scattegram image which shows the correlation between the data of band 1 and band 2.

Bands to be compared can be changed by clicking on the set up button, and change the X-axis and Y-axis. The red region shows the correlation between two bands.

Figure 5 shows the correlation between band 4 and band 5.

Importance of Algorithm in Remote Sensing

Algorithm is one of the functions in remote sensing. It helps to process raw data obtained into data that can be read and understandable to all. It uses Einstein's equation to convert or process data. It can be converted into 2D or 3D data, which helps users to further understand and analyse more of the data. 

In 3D data, image can be rotated front and back, left or right, in order to assist analysis. Several function can be selected and randomised in order to change the structure of the 3D data. Colour of the data can be changed according to user's choice. In algorithm, there are seven bands, from band 1 to band 7. The higher the band, the clearer is the image. However, it only applies until band 5. For band 6 and 7, image will remain unclear.

I felt excited and happy through this learning process because it is very interesting. I am eager to learn more on this application so that I can further analyse any data to then create maps for many users in all forms. ^^

My Expectations for ESC 4511 Environmental Remote Sensing

I hope that by the end of this semester, I would be able to use remote sensing application, along with GIS to further improve my FYP. Also to sharpen my skills on both remote sensing and GIS so that I am able to apply them in my career in the future.