H17+V5

== Morocco is located in the northern margin of the northern hemisphere desert belt, and most of its land is arid or semi-arid. I want to investigate and analyze the land cover and land-use changes, these changes associate with either actions of climate change conditions or human actions. ==



The following three maps display Morocco for fall, spring and summer of 2010, I used Normalized Difference Vegetation Index (NDVI) because is extensively use worldwide as a measure of vegetation vitality, with high NDVI values indicating vigorous vegetation. The classified NDVI values range from -1 to +1. The green areas indicate a high density of plants. The NDVI shows the agriculture areas of Morocco.







=__**Lab4**__= = ﻿ = = ** 1. Anthromes file, Samples 158 Lines 121 Bands 1, MODIS time series, Samples 263 Lines 201 Bands 144. ** = = ** 2. MODIS time series has a more lines and columns than Anthromes so they are different because they have different resolution, more line it means more overlap of each other as in MODIS tomes series. 16 days compost and 144 time steps ** = = ** 3. It would not be right to shrink the Anthroms pixels, because each pixel had a percentage so we don’t know exactly how to break it up in to small pixels and each pixels should had a one code only not multiple codes for one pixel it should be mutually exclusive. ** = ==



= ** 4. List the 5 dominant classes in your tile (in order from most dominant to least dominant). ** = = ** 1. Residential rangelands (31) ** = = ** 2. Residential rained mosaic (41) ** = = ** 3. Remote rangelands (42) ** = = ** 4. Populated rangelands (43) ** = = ** 5. Barren (63) ** =

= **Study classes: ** =


 * **//# Class //** || **//# pixels in area //** || **//% pixels in area //** ||
 * **11 Urban ** || 76 || 0.398 ||
 * **12 Dense settlements ** || 71 || 0.371 ||
 * **21 Rice villages ** || 0 || 0 ||
 * **22 Irrigated villages ** || 46 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0.241 ||
 * **<span style="font-family: 'Garamond','serif';">23 Cropped pastoral villages ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">235 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1.229 ||
 * **<span style="font-family: 'Garamond','serif';">24 Pastoral villages ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">212 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1.109 ||
 * **<span style="font-family: 'Garamond','serif';">25 Rainfed villages ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">247 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1.292 ||
 * **<span style="font-family: 'Garamond','serif';">26 Rainfed mosaic villages ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">120 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0.628 ||
 * **<span style="font-family: 'Garamond','serif';">31 Residential irrigated cropland ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">633 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">3.311 ||
 * **<span style="font-family: 'Garamond','serif';">32 Residential rainfed mosaic ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">2,279 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">11.921 ||
 * **<span style="font-family: 'Garamond','serif';">33 Populated irrigated cropland ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">81 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0.424 ||
 * **<span style="font-family: 'Garamond','serif';">34 Populated rainfed cropland ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">699 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">3.656 ||
 * **<span style="font-family: 'Garamond','serif';">35 Remote croplands ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">20 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0.105 ||
 * **<span style="font-family: 'Garamond','serif';">41 Residential rangelands ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">2,405 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">12.580 ||
 * **<span style="font-family: 'Garamond','serif';">42 Populated rangelands ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1,268 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">6.632 ||
 * **<span style="font-family: 'Garamond','serif';">43 Remote rangelands ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1,392 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">7.281 ||
 * **<span style="font-family: 'Garamond','serif';">51 Populated forests ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">247 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">1.292 ||
 * **<span style="font-family: 'Garamond','serif';">52 Remote forests ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">6 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0.031 ||
 * **<span style="font-family: 'Garamond','serif';">61 Wild forests ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0 ||
 * **<span style="font-family: 'Garamond','serif';">62 Sparse trees ** || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0 || <span style="font-family: 'Garamond','serif'; line-height: normal; margin: 0in 0in 0pt;">0 ||
 * **<span style="font-family: 'Garamond','serif';">63 Barren ** || 793 || 4.184 ||

** 4. ** **The area is covered by the 5 most dominant classes 42.598% or 2043.25 km2 of the total area**

**5. There is no much variation and different between the NDVI and coefficient expect that the coefficient nosier that the NDVI for the 5 classes.** **6. The start and the end of the growing season can be detriment from the NDVI time series the growing season starts in spring and hits highest peak in the summer and the level of NDVI over most of the agricultural areas of Morocco, as expected, and here is some examples** __ NDVI average Residential rainfed mosaic year 2003 __

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 0pt 0.5in; text-indent: -0.25in;">[[image:res_rainfed_moro_03.jpg width="264" height="203"]]
__ NDVI average Residential rainfed mosaic year 2007 __



**7. Don’t think it would matter in my study area, because there are no major land cover changes in the area.** **8. There is no major transformation in the trends, so they are normal they stayed the same.** **9.** **In my study area there is no intensive changes, the land cover classes mostly similar. (( here is brief description of the most 5 dominant classes in my study area ,Residential Rainfed Mosaic Croplands i is the most extensive of the more densely populated anthromes a mixture of cropland, forest, human settlements, pasture, and limited urban area, The Remote Rangeland anthrome essentially wild with little sign of human activity and the Residential Rangelands anthrome has two key features; its not population and a substantial portion of its area is used for pasture, so no other land cover is as dominant as pasture in rangelands. Populated Rangeland is an anthrome have a low net primary production and land use and finally the Barren anthrome represents the most extreme environments is comprised of the Sahara Desert and the annual precipitation is very low and the land use is existent in the is region)). (([|http://ecotope.org]))**

**__ Time Series NDVI __**

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;"> <span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">**__ Coefficient of Variation __**

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">

== Vegetation net primary productivity (NPP) is a key component of the terrestrial carbon cycle. The result shows that the terrestrial NPP of Morocco from 2002 to 2006. Net primary productivity (NPP) is defined as accumulative organic matters by green plants per unit of space and time NPP reflect not only the plant community productivity directly for a certain natural environment but also fixing ability for CO2 by photosynthesis. The NPP time series of Morocco terrestrial during 2002 - 2006 using MODIS, shows an obvious lower values on eastern land areas, with higher NPP value in Coastal areas. It can be drawn that from 2002-2006, the NPP value varies from 0 to 1.6 gC/m2/yr.The representation of the NPP data was poor due to lack of appropriate model. ==



<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;"> <span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">Landsat

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">Lab 7 the table below is the overview of the all Landsat images withh less than 20% cloud cover for Morocco

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 115%; margin: 0in 0in 10pt 0.5in; text-indent: -0.25in;">Lab 8

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">﻿Atmospherically correct image 6/23/2010



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">﻿Original Spectrum



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Radiance



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Atmosphere Reflectance



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Surface Reflectance

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">﻿Atmospherically correct image 6/20/2009



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">﻿Original Spectrum



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Radiance

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Atmosphere Reflectance



<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">Surface Reflectance

<span style="font-family: Calibri; font-size: 18pt; line-height: 0px; margin: 0in 0in 10pt 0.5in; overflow: hidden; text-indent: -0.25in;">

Spatial Filter
Low pass filters are used to emphasize low spatial frequency data and are designed to smooth out an image with high spatial frequency, I believe in Moroco region the low Pass Filter with less Kernal size the image appreas more smooth and less noise.

QuickLook of all six principle components
In the Principle components PC 1 contains most of the vegetationvartions, PC2,3,4,5 the vartions seems to be less and the noise increses more in PC 6.

Quicklook of the inverse PC ture color image.
In PC ture color images, it seems the colors washed ou for the image,and vegations appear to be really dark in the image, so fort hat reason I used Band 4,3,2 compnation to dispaly the vegetation areas

Band 3 shows the highest correlation and Band 4 shows least correlation


Band 1, 2, 3 contains mot of the information between 88% to 4% and band 4,5,6 containe the noise.

<span style="font-family: 'Arial','sans-serif'; font-size: 12pt; margin: 0in 0in 10pt;">**__Change Detection__**

Change Time 2



Change Time 1



PC 3

PC 2



NDVI



=Lab 11 and Lab 12=

=2009 Image Training ROI=

=2009 Image Validation ROI= ==

=2010 Image Training ROI= ==

=2010 Image Validation ROI= == =Supervised Classification 2009=

=The Legend for Supervised Classification Maps 2009 and 2010=



=Supervised Classification 2010=



I can compare both images 2009 and 2010 by tracking the changes through the year. By looking at the type of material present on the landscape for example water, sand, crops, forest, wetland, human-made materials such as roads and also depend on the features on the surface of the Earth that already measured by remote sensing. The other way to compare is by looking on what people do on the land surface for example agriculture, commerce, settlement and human activity in a Location, derived from Land-Cover. On the other hand, I can compare Land-cover and Land-use by examines deforestation, settlement growth and water resources, which what found is increasing in agriculture areas in 2010 in the south east of the country, more shurbland in 2010 but I notice the forests decreased in 2010 especially in the south west of the country. I think both of the classification is accurate but regarding which one have more accuracy assessment I will go with Landsat 2009 image because of the higher percentage which is 99.8577%. =Confusion Matrix 2009=



=Confusion Matrix 2010=



<span style="font-family: 'Arial','sans-serif';">The accuracy assessment for Landsat 2009 images is 99.8577% and Landsat 2010 image is 99.6949% shows that my both images have a really good accuracy in the classification. The lowest user accuracy in both 2009 and 2010 is Shurbland and the lowest production accuracy is Forest because of my selections especially that some of the Shurbland located near the Forest, which <span style="font-family: 'Arial','sans-serif'; line-height: 200%;">may either occur naturally or be the result of human activity.

=Change Detection=

=Water=



=Urban=



=ShurbLand=

=Forest=

=Agriculture= ==

=Pixel Count=



=Area Change=

.

=Percentage Change=

According to the change detection statistics report, which is to compile a detailed tabulation of changes between two classification images, that the largest changing class is km2 is urban with 418.12 km2 and agriculture with 94.723 % and the second class for largest change in both km2 and percentage is forest with 535.89 km2 and 55.470%. The results for changed classes are more visible with change detection statistics. This analysis focuses primarily on the initial state classification changes; that is, for each initial state class, the analysis identifies the classes into which those pixels changed in the final state image, by looking to the change detection mask I can see the changing for each class and understand it more and how change by looking to the changing mask and in the same time the change detection statistics report. However, from the previous report the majority of change detection result in the data is difficult to interpret.