WOW. this is simply perfect for me as my research topic is "LST estimation using Landsat dataset over an urban area". just looks like i'll get it all figured out thanks to this amazing practical methodology
Two part fixed: PROPVEG = Square(("NDVI" - "NDVI_min")/("NDVI_max" - "NDVI_min")) LST = "sattempb1011" /( 1+(10.89* "sattempb1011" / 14380) * Ln("LSE")) Thanks Usman Buhari's video, wish you all will do it well.
+AJ C Sir, You can follow fromgistors.blogspot.com/2014/01/estimation-of-land-surface-temperature.html, the band 10 Center wavelength (µm) of Landsat-8 is 10.8.
Very much nice! I made here, how you teached and stayed very right, equal the temperature finded in the local station of my city in the same hour. Thanks!
i have lttle confusion here 1st= while conversion factor for kelvin to degree celsius is -273.15 but u did -272.15 2nd While calculating LSE in the at first while taking NDVI min in the numerator part you told -*-=+ you added the NDVImin but at denominator part you didnt do so what is the reason behind this. Could you pls elaborate it. With best Regards
Usman, at 10:44 of the video where you put formula for Propveg, the (ndvi_max - ndvi_min) should have been (0.60875 - (-0.587384)) = (0.60875 + 0.587384). I think you forgot to multiply two minuses. Great tutorial though.
The formula for Land Surface Temperature is missing a left-parenthesis right after the division symbol to group the whole denominator. Otherwise you are just dividing by one before adding the rest of the stuff, as ArcMap follows the normal algebraic order of operations. Same for the NDVI formula -- should be ((NDVI - NDVImin)/(NDVImax - NDVImin))^2.
For those interested in using this methodology in urban areas, in 2007 Stathopoulou published an urban emissivity equation of: 0.017Pv+0.963. www.tandfonline.com/doi/full/10.1080/01431160600993421?scroll=top&needAccess=true Integrating Corine Land Cover data and Landsat TM for surface emissivity definition: application to the urban area of Athens, Greece
Dear sir. Thanks for the clip. It is very useful Could you please tell me the number 0.004 and 0.986 coefficients for emmissivity and p=14380 is constant for any Landsat 8 band 10 & 11 Thanks for your help.
Hello Usman, Thanks for posting such wonderful video. Can you please tell me the literature for this derived formula it will help me a lot. please please. Thanks Prashant
I have stack image of all the bands of Landsat 8. unfortunately, i don't have original satellite data set. How can I get separate band layer of band 10 and 11 to perform this exercise?
thank you for such a precise and great Lecture, its a break through for my research,,, can you kindly extend it evapotranspiration analysis using remote sensing,,,,,
Hello Usman Buhari, This is Yavar, I'm PhD student at UCLA. for my thesis I need to calculate Evapotranspiration using SEBAL model, I have watched two of your videos and they are quite useful, my question is do you have any resource or may be video that shows how to calculate Surface Albedo or even Evapotranspiration using SEBAL method? I would really appreciated if you could help me. Yavar
Why are calculating the surface temperature using band 10 only? Why did you calculate the mean value of band 10 and 11? Can you also verify these formula BT/1+w*(BT/p)+ln(e)), the bracket are not matching. Your tutorial is good. These questions shows that its easy to understand.
There are errors associated with band 11 (and minor errors associated with band 10). USGS does not recommend using band 11 for quantitative studies. To your other question-- the negative sign on the other NDVImin was incorrect. Band 11 errors can be up to 8 K off from actual and Band 10 errors can be up to 4 K off from actual. landsat.usgs.gov/how-are-thermal-infrared-sensor-tirs-thermal-bands-aboard-landsat-8-used
Is there any difference in final land surface temperature in Band 10 and band 11 of landset 8? Should I calculate the land surface temperature seperately?
Its really great video Usman, actually now i was working about soil moisture, in your opinion, is it possible for us to estimate soil moisture from this LST, i mean, is there any formula or method to convert it? if u have any idea and video tutorial, i hope you can share it to me, tq.
Hello Usman, nice video!! I have one question: why you choose that version of emissivity formula... Isn't that for areas with high values of NDVI (like over 0.7)??
Hi, Thank you alot for this useful video. I have a question please, which combination of bands I can use in the estimation of land surface temperature. waiting for response in the short delays. thank you
Usman, it was for simple curiosity, I understand that k1 and k2 can be taken from an image's header, and given the fact that those constants vary from one sensor to another, i wanted to know the way of reading that header to get the constants, i've been looking up and i think matlab is a good choice
Hello Brother Usman, You did an Amizing work with those calculations, I did the same thing with the ocean temperature using landsat 8 in er-mapper, but I still had some question for the work we both had done which is how we gone use this kind of analysis since I found that no matter the result is, the result depends on the period of the time that sattelite taken the data. differend time produce differend results. I think you agreed with me :)
Hi Usman, thank you for the upload - very helpful ! Do you by any chance have a reference for the LST and Emisivity formula you have used? Would really appreciate it
+islandmonkey87 Found the references Emissivity : Land surface temperature retrieval from LANDSAT TM 5, Sobrino et al, 2004 LST : Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies, Weng, 2003
+islandmonkey87 Also note that the Land Surface Emissivity formula is only valid if no water bodies are present in the image look at this article, pp 8 : Comparison Between Fractional Vegetation Cover Retrievals from Vegetation Indices and Spectral Mixture Analysis: Case Study of PROBA/CHRIS Data Over an Agricultural Area
Hi Majid, i ended up writing a Matlab script instead of using ArcGIS to process the Landsat data. I have used the NDWI with a threshold value to remove water bodies from the satellite images. Once the water body pixels were removed i calculated LST
Hi Usman it is really a wonderful video, following your instructions I got min -124 and 45 of landsat image of Lahore (Pakistan), which is theoretically not possible. Could you please guide me what was actually wrong to result?
Howdy Usman, thank you for this tutorial. However, I would like to point out a possible mistake you have done while calculating emissivity. Since minimum NDVI values are negative you should have changed both subtractions signs to addition signs.In my trial my Percent Vegetation ranges from 0 to 1.Another question : in literature I saw that NDVI for soil which is normally 0.2 and NDVI for vegetation which is normally 0.5 is used instead of Minimum NDVI and Maximum NDVI.If you have a reference for the calculation of percent vegetation in the way you did, can you please share it with us.THANK YOU SO MUCH.
Hello, thanks for uploading this video. I would suggest to check some of the equations you used in your tutorial. I can see some small errors (for example the PV formula is missing a parenthesis) that can greatly change the results. By the way, Rho value is 1.438e-2 (0.01438).
Hi, thanks for the comment. I would again want to mention that several methods have been developed for estimating LST so somethings you term mistakes might just be correct. However, I just watched the video again and can't find where i missed a parenthesis in the PV equation, would you mind telling me the exact position i missed it? And what do you refer to "Rho"?
Hi Usman. Thanks for taking the time to answer to my comments. As far as I know the PV formula is [(NDVI - NDVImin) / (NDVImax - NDVImin)]^2, but it seems that in your explanation you missed the parenthesis that separates the upper and lower part of the formula. With "Rho" I was referring to this formula: Rho = hc/s = 1.43810e-2, s = Boltzmann constant (1.3810 23 J/K), h = Planck’s constant (6.62610 34 J s), and c = velocity of light (2.998108 m/s). I am not an expert in this topic, but I was reading a few papers and then watched your video, so those differences called my attention. By the way. I just found other paper that uses an alternative way to estimate the LST. It's an open access journal: file.scirp.org/Html/3-8401104_17055.htm. All the best!
One question. In the case of needing an atmospheric correction, what would you recommend. It is more likely to have the need of work with more than one date, so the correction is handy. Thank you for the tutorial.
Hi Usman, I am not sure its a mistake or there is a reason behind it, that the equation for the NDVI, in the denominator part you didn't changed the -*- to + while putting the value for NDVI min. I am doing this for my Research Project, so kindly if you can tell me the reason for it. Great video !
thank you very much for this , but i seem to have a challenge with it. after running the analysis for LST I got a temperature range that was very high beyond the normal temperature. what could have caused it? thanks in advance.
Mr. Usman, for the proportion of vegetation, the value of min NDVI should be written as positive values. You have wrote it as a plus for the first and then negative for the second. Secondly, to find the LST as you have averaged out the temperatures of band 10 and 11, that should be used. Why did u use band 10? or what was the purpose to average out two bands then?
Thanks very much for this video. I followed the same procedures but I'm finding the negative values of LST, which have never been occurred in our country(Burundi, that is tropical country). Note that the temperature range from 18 to 30degree Celsius, but here I find -10.90 to 10.56 degree Celsius) Can you explain me why these negative values occurs? thanks.
Hi Mr usman Buhari, thats is really helpful video, i am just wondering, do i need to do atmospheric correction for my image, specially for thermal band 10 and 11. Before calculating land surface temperature. Thanks
what is the number of images you need to get a trend for the land surface temperature that depicts a notable variation and should one take monthly images or daily? which one is more easier to work with
Thank for the wonderful presentation. I closely followed all your procedures for my study area but my minimum LST temp is negative which I'm sure can't be possible. Sir, can it be because the Landsat data were not corrected atmospherically or what?
Hi thanks for uploading this! I am a student learning remote sensing and mapping and this method looks awesome!! :D I am trying to follow this method, but I wonder where and how you got the NDVI image? Please let me know! Thanks! :)
Hi Sean and thanks for the comment. Well this video assumes that anyone watching it can produce an NDVI map. Watch this other video of mine to learn how to make an NDVI map at m.ua-cam.com/video/PknESopCwjA/v-deo.html
Usman Buhari Thank you so much for very quick and thoughtful reply! :D I am a new to GIS and remote sensing world...still have lots to learn! Thanks for the suggestion video! I will definitely have a look! Very much appreciated! :D Have a wonderful rest of the week!!
Hi, Thank you alot for this useful video. please I have two questions. Which combination of bands I can use in the estimation of land surface temperature. waiting for response in the short delays and if we can extract this values of temperature, to make a map of surface temperature ? Thank you
+Usman Buhari , Ok, Thank you Usman, I want to know if I can extract theses value from the raster format to represent it in isovalues of surface temperature map ..
Hi Usaman buhari, how did you create same subset for both thermal and multispectral band of your study area. i am trying from last three days but i am getting error in ENVI, please let me know how did you do that. Thanks much appreciated
hi usman, can i ask the link of literature from this tutorial? beside the contents of Landsat product like LSE equation, proportion of vegetation, etc. thanks berfore :)
Hello Usman. Can you please make a video on how to calculate NDSI for glacier studies? I am new to remote sensing and I haven't found people uploading videos where they clearly explain how to calculate NDSI. I would appreciate if you could make a video where you also show where to get the best data and how to put algorithms using arc gis and/or qgis? If you know about a detailed explanation published as a video or a flow chart with pictures, please share.thank you
This formula will do what you want NDSI = (Band GREEN - Band SWIR) / (Band GREEN + Band SWIR).... Depending of the software you use, put this formula into a raster calculator and you are done. Good luck.
First i would like thank you for your amazing videos. My question is that as my study area falls under multiple scenes due to large area. The sun elevation is different for different scenes. I want to know how to choose the sun elevation for different scenes as i have to mosiac them first. Regards John
Hello John, since you have multiple scenes it is best you treat each of them separately after which you can do the mosaic. This is part of beat practices when using remote sensing data, because each scene is taking at different times with different factors.
Dear Usman, firstly thank you for your nice work in this video. I am trying to do same operations on my images but I think I cannot succeed it so at stage of conversion from radiance to sat temp. my values range from -124.633 to 43.7934 for Band10 and similar results for Band11, is it possible or I am doing a mistake one of the stages. Could you help me about it.
+Sinan ARAS I had the same problem while later I found the extreme values such as -124 were for the marginal part of the segment without valid thermal infrared info.
Hi dear freand, this is a really great video. I hope you can help me: I have interest on using this method for LST for my thesis, do you have de article or bibliography that describes the methodology ? Thanks!
hii... I have encountered an error during the sattemp creation.... Temperature ranges from -1.6 to 42 which is not feasible in my study area. Can you guide me about where I would have gone wrong
hi Osman or any one know please can you explain how did you get the P value where you but 14380 and is this value is fixed can not be changed we can applied wherever we need thanks
WOW. this is simply perfect for me as my research topic is "LST estimation using Landsat dataset over an urban area". just looks like i'll get it all figured out thanks to this amazing practical methodology
Dear Mr. Usman, this video is the most helpful. Thank you very much for this!
Two part fixed:
PROPVEG = Square(("NDVI" - "NDVI_min")/("NDVI_max" - "NDVI_min"))
LST = "sattempb1011" /( 1+(10.89* "sattempb1011" / 14380) * Ln("LSE"))
Thanks Usman Buhari's video, wish you all will do it well.
+ScottC Why the 10.8? Is it he mean of wavelength of Band10? Thank you.
+AJ C Sir, You can follow fromgistors.blogspot.com/2014/01/estimation-of-land-surface-temperature.html, the band 10 Center wavelength (µm) of Landsat-8 is 10.8.
Thank you very much,your corrections were totally useful,and the link as well.
The middle Value to Use Landsat-8 Band 10 is 10.89 landsat.usgs.gov/what-are-band-designations-landsat-satellites
Shouldn't the second NDVI_min also be positive since its a negative number
Quality videos... all of them. Thanks Usman
Very much nice! I made here, how you teached and stayed very right, equal the temperature finded in the local station of my city in the same hour. Thanks!
Thanks dearest for your clear steps and explanation!! Please keep it up
@Usman Buhari thanks a great deal for such an insightful RS video. It's pretty educative.
Thank you so much for this Mr Usman, ALhamdolillah very very helpful tutorial! Jazaka'Allahu Khayr Brother!
Superb job, well -272.15 is corrected oneor -273.15
i didn't take this kind of lesson before..!! many Thanks that video it is useful ..!!!!!
GREAT !please discuss surface reflectance
I think, the conversion from Degree K to Degree C is "- 273.15". Please correct me if I am wrong.
i have lttle confusion here
1st= while conversion factor for kelvin to degree celsius is -273.15 but u did -272.15
2nd While calculating LSE in the at first while taking NDVI min in the numerator part you told -*-=+ you added the NDVImin but at denominator part you didnt do so what is the reason behind this.
Could you pls elaborate it.
With best Regards
Here noticed the same things
This was excellent I finally got it.Thank you
Great Job ,,,, Thanks
We learned too much from you ,,, God Bless you
Thank you Youssef and am glad you enjoyed the video.
thank you very much sir. this tutorial is very usefull.
im from indonesian
Thanks for your great tutorial
Very interesting and educative. Thanks
Really good job ! :) I wish my computer was as fast with the calculations as Yours :D
Toolbox and document both available
ua-cam.com/video/ZzyBlsuDQb0/v-deo.html
Very Nice. Plz discuss remote sensing and AI.
Usman, at 10:44 of the video where you put formula for Propveg, the (ndvi_max - ndvi_min) should have been (0.60875 - (-0.587384)) = (0.60875 + 0.587384). I think you forgot to multiply two minuses. Great tutorial though.
i have the same doubt. i think you did wrong.
The formula for Land Surface Temperature is missing a left-parenthesis
right after the division symbol to group the whole denominator.
Otherwise you are just dividing by one before adding the rest of the
stuff, as ArcMap follows the normal algebraic order of operations. Same
for the NDVI formula -- should be ((NDVI - NDVImin)/(NDVImax -
NDVImin))^2.
For those interested in using this methodology in urban areas, in 2007 Stathopoulou published an urban emissivity equation of: 0.017Pv+0.963.
www.tandfonline.com/doi/full/10.1080/01431160600993421?scroll=top&needAccess=true
Integrating Corine Land Cover data and Landsat TM for surface emissivity definition: application to the urban area of Athens, Greece
Indiana1616 do you have paper for estimation of emissivity in montain area ?
Dear sir.
Thanks for the clip. It is very useful
Could you please tell me the number 0.004 and 0.986 coefficients for emmissivity and p=14380 is constant for any Landsat 8 band 10 & 11
Thanks for your help.
Thanks for your informative videos, please could you do a video on air temperature retrieval
thank you for this video
but I still don't understand the last step. why you using ban d10 at satellie instead of cell statistic ?
Hello Usman,
Thanks for posting such wonderful video. Can you please tell me the literature for this derived formula it will help me a lot. please please.
Thanks
Prashant
Land Surface Temperature from Landsat 5 TM images: comparison of different
methods using airborne thermal data
Congratulations for the video ! Can you advise me about papers that support the equations of the video?
Thanks !
10:46 shouldnt there be a plus sign in the video for minus* minus makes plus- Square(("NDVI" - "NDVI_min")/("NDVI_max" - "NDVI_min")) please reply.
minus multiply minus
I have stack image of all the bands of Landsat 8. unfortunately, i don't have original satellite data set. How can I get separate band layer of band 10 and 11 to perform this exercise?
did you undertake the atmospheric correction already?
thank you for such a precise and great Lecture, its a break through for my research,,, can you kindly extend it evapotranspiration analysis using remote sensing,,,,,
thank you for this video . very helpful. Could please make a video using GIS for turbidity of a river?
Hello Usman Buhari,
This is Yavar, I'm PhD student at UCLA. for my thesis I need to calculate Evapotranspiration using SEBAL model, I have watched two of your videos and they are quite useful, my question is do you have any resource or may be video that shows how to calculate Surface Albedo or even Evapotranspiration using SEBAL method?
I would really appreciated if you could help me.
Yavar
Why are calculating the surface temperature using band 10 only? Why did you calculate the mean value of band 10 and 11? Can you also verify these formula BT/1+w*(BT/p)+ln(e)), the bracket are not matching. Your tutorial is good. These questions shows that its easy to understand.
There are errors associated with band 11 (and minor errors associated with band 10). USGS does not recommend using band 11 for quantitative studies. To your other question-- the negative sign on the other NDVImin was incorrect. Band 11 errors can be up to 8 K off from actual and Band 10 errors can be up to 4 K off from actual.
landsat.usgs.gov/how-are-thermal-infrared-sensor-tirs-thermal-bands-aboard-landsat-8-used
Hi Usman!
Which paper did you read to obtain this results?
I'm very interested to know all the mathematical background of this
Is there any difference in final land surface temperature in Band 10 and band 11 of landset 8?
Should I calculate the land surface temperature seperately?
Why did you put the negative sign on the last NDVImin when calculating PV.
I love you. This is incredible!
Toolbox and documents available in description
ua-cam.com/video/ZzyBlsuDQb0/v-deo.html
Great Job...... Thank You Very Much
Hi Usman, thank you for sharing this very useful video! I was wondering if the algorithms can also used in TM and ETM+ data for calculating the LST?
Its really great video Usman, actually now i was working about soil moisture, in your opinion, is it possible for us to estimate soil moisture from this LST, i mean, is there any formula or method to convert it? if u have any idea and video tutorial, i hope you can share it to me, tq.
Hello Usman, Can you make a video on the Split-Window Algorithm as well?
THANK YOU SIR FROM INDONESIA
Dear Usman Buhari,
I have one question, for parameter P= 1.438*10^(-2) but why you put it 14380?
have you already made the necessary radiometric and atmospheric corrections or have you worked directly on the Landsat images?
Hello Usman, nice video!!
I have one question: why you choose that version of emissivity formula... Isn't that for areas with high values of NDVI (like over 0.7)??
I think more visualised technique of doing drainage density map is by creating grid using fishnet tool. Please elaborrate the process thanks
Dear Usaman, thanks a lot for your tutorial. whats the difference for calculation LST in Landsat 7 TM and ETM?
Hi, Thank you alot for this useful video. I have a question please, which combination of bands I can use in the estimation of land surface temperature. waiting for response in the short delays. thank you
Usman, it was for simple curiosity, I understand that k1 and k2 can be taken from an image's header, and given the fact that those constants vary from one sensor to another, i wanted to know the way of reading that header to get the constants, i've been looking up and i think matlab is a good choice
Hello Brother Usman, You did an Amizing work with those calculations, I did the same thing with the ocean temperature using landsat 8 in er-mapper, but I still had some question for the work we both had done which is how we gone use this kind of analysis since I found that no matter the result is, the result depends on the period of the time that sattelite taken the data. differend time produce differend results. I think you agreed with me :)
Hi Usman, thank you for the upload - very helpful ! Do you by any chance have a reference for the LST and Emisivity formula you have used? Would really appreciate it
+islandmonkey87 Found the references
Emissivity : Land surface temperature retrieval from LANDSAT TM 5, Sobrino et al, 2004
LST : Estimation of land surface temperature-vegetation abundance relationship
for urban heat island studies, Weng, 2003
+islandmonkey87 Also note that the Land Surface Emissivity formula is only valid if no water bodies are present in the image
look at this article, pp 8 : Comparison Between Fractional Vegetation Cover Retrievals
from Vegetation Indices and Spectral Mixture Analysis: Case
Study of PROBA/CHRIS Data Over an Agricultural Area
thanks for your note. what should we do in this case? there is few small water bodies in my study area! please email me at: amani.majid@gmail.com
Hi Majid, i ended up writing a Matlab script instead of using ArcGIS to process the Landsat data. I have used the NDWI with a threshold value to remove water bodies from the satellite images. Once the water body pixels were removed i calculated LST
thank you
Hi Usman it is really a wonderful video, following your instructions I got min -124 and 45 of landsat image of Lahore (Pakistan), which is theoretically not possible. Could you please guide me what was actually wrong to result?
I have to ask you one question. Landsat 7 with 2 channel 6. temperature properties , the properties in which channel. tks!
you can do the same think like usman did in his tutorial, landsat 7 etm has band 6a-6b thermal cannal same as landsat 8 with band 10-11
do you mean this formula?
TR = Ɛ ¼ Ts where :
TR= kinetic objects Temperature
Ɛ=emissivity
Ts=Surface temperature
Howdy Usman, thank you for this tutorial. However, I would like to point out a possible mistake you have done while calculating emissivity. Since minimum NDVI values are negative you should have changed both subtractions signs to addition signs.In my trial my Percent Vegetation ranges from 0 to 1.Another question : in literature I saw that NDVI for soil which is normally 0.2 and NDVI for vegetation which is normally 0.5 is used instead of Minimum NDVI and Maximum NDVI.If you have a reference for the calculation of percent vegetation in the way you did, can you please share it with us.THANK YOU SO MUCH.
Do you have such information?
@ Comparison of Three Algorithms for the Retrieval of Land Surface Temperature from Landsat 8 Images
@@rebinuthis ur thing is not English!
@@MohammedAbdulkadir-e9k search for this paper, has the values
hi answer me pleaseeee
1. what is the unit TOA radiance?? (watt?)
2. landsat 7 Like landsat 8
Hello, thanks for uploading this video. I would suggest to check some of the equations you used in your tutorial. I can see some small errors (for example the PV formula is missing a parenthesis) that can greatly change the results. By the way, Rho value is 1.438e-2 (0.01438).
Hi, thanks for the comment. I would again want to mention that several methods have been developed for estimating LST so somethings you term mistakes might just be correct. However, I just watched the video again and can't find where i missed a parenthesis in the PV equation, would you mind telling me the exact position i missed it? And what do you refer to "Rho"?
Hi Usman. Thanks for taking the time to answer to my comments. As far as I know the PV formula is [(NDVI - NDVImin) / (NDVImax - NDVImin)]^2, but it seems that in your explanation you missed the parenthesis that separates the upper and lower part of the formula. With "Rho" I was referring to this formula: Rho = hc/s = 1.43810e-2, s = Boltzmann constant (1.3810 23 J/K), h = Planck’s constant (6.62610 34 J s), and c = velocity of light (2.998108 m/s). I am not an expert in this topic, but I was reading a few papers and then watched your video, so those differences called my attention. By the way. I just found other paper that uses an alternative way to estimate the LST. It's an open access journal: file.scirp.org/Html/3-8401104_17055.htm. All the best!
Thanks Ignacio, as for the PV I would further me research on that and for Rho its just basic physics and maths and am certain about the value (14380).
One question. In the case of needing an atmospheric correction, what would you recommend. It is more likely to have the need of work with more than one date, so the correction is handy. Thank you for the tutorial.
Use FLAASH or 6S... I think are the best actually models.. but If you dont have experience use DOS directly
Hi Usman,
I am not sure its a mistake or there is a reason behind it, that the equation for the NDVI, in the denominator part you didn't changed the -*- to + while putting the value for NDVI min. I am doing this for my Research Project, so kindly if you can tell me the reason for it.
Great video !
it was a little mistake in conversion only, So as Akram siad just use (14380) for P and it will be fine
thank you very much for this , but i seem to have a challenge with it. after running the analysis for LST I got a temperature range that was very high beyond the normal temperature. what could have caused it? thanks in advance.
Mr. Usman, for the proportion of vegetation, the value of min NDVI should be written as positive values. You have wrote it as a plus for the first and then negative for the second. Secondly, to find the LST as you have averaged out the temperatures of band 10 and 11, that should be used. Why did u use band 10? or what was the purpose to average out two bands then?
Hi Usman, can you please upload a video on how to calculate soil moisture index in arcgis
Nice job Man. Where did you get 0.004 and 0.986 coefficients for emmisivity. Thank you.
agroredMX thanks....those are constant given in the formula
Thanks very much for this video. I followed the same procedures but I'm finding the negative values of LST, which have never been occurred in our country(Burundi, that is tropical country). Note that the temperature range from 18 to 30degree Celsius, but here I find -10.90 to 10.56 degree Celsius) Can you explain me why these negative values occurs? thanks.
Thank you!
When you calculate Pv value, denominator ( ie, NDVImax - NDVImin ) will be ( 0.608759+0.058784), but you still use -ve in-between.
Yes, this was a mistake realised after the video has been recorded. But I am sure the whole idea is understood by all. Thanks.
The tutorial is awesome. Simple and succinct.
Where did you find ndvi? I couldn't find it in earthExplorer. how did you create it?
thansk for the tutorial
Hi Mr usman Buhari, thats is really helpful video, i am just wondering, do i need to do atmospheric correction for my image, specially for thermal band 10 and 11. Before calculating land surface temperature. Thanks
I would recommend you do an atmospheric correction for all bands as this will take care of most atmospheric errors with image.
what is the number of images you need to get a trend for the land surface temperature that depicts a notable variation and should one take monthly images or daily? which one is more easier to work with
Thank for the wonderful presentation. I closely followed all your procedures for my study area but my minimum LST temp is negative which I'm sure can't be possible. Sir, can it be because the Landsat data were not corrected atmospherically or what?
it is due to cloud contaminated pixels..USE gross technique to detect cloud contaminated pixels and exclude by using set Null function in Arc GIS
Hi usman,
please answer
how to make air humidity analysis using Landsat 8 in Arcgis ?
Hai usman,
Can the emissivity equation be applied in every site place or there is other aquation?
In the equation of emission, can I use the same numbers for the Landsat 7 ETM sensor
Hi thanks for uploading this! I am a student learning remote sensing and mapping and this method looks awesome!! :D I am trying to follow this method, but I wonder where and how you got the NDVI image? Please let me know! Thanks! :)
Hi Sean and thanks for the comment. Well this video assumes that anyone watching it can produce an NDVI map. Watch this other video of mine to learn how to make an NDVI map at m.ua-cam.com/video/PknESopCwjA/v-deo.html
Usman Buhari
Thank you so much for very quick and thoughtful reply! :D
I am a new to GIS and remote sensing world...still have lots to learn! Thanks for the suggestion video! I will definitely have a look!
Very much appreciated! :D
Have a wonderful rest of the week!!
Hi, Thank you alot for this useful video. please I have two questions.
Which combination of bands I can use in the estimation of land surface temperature. waiting for response in the short delays and if we can extract this values of temperature, to make a map of surface temperature ? Thank you
This videos explains exactly your question. The therm bands (10&11) are what is required.
+Usman Buhari , Ok, Thank you Usman, I want to know if I can extract theses value from the raster format to represent it in isovalues of surface temperature map ..
Thanks a lot sir..
hello Usman Why you use a p value of 14380? when the result of p = h * c/s is equal to 0.01439474
hi Husam is this value (14380) can be fixed value so it can be used for all Landsat data
Hi Usaman buhari, how did you create same subset for both thermal and multispectral band of your study area. i am trying from last three days but i am getting error in ENVI, please let me know how did you do that. Thanks much appreciated
when you change the BT into Celcius, is the P already in Celcius too? thank you
My NDVI range shows a min of -1 and a max of +1, is there some setting I need to change to get the same in decimals?
hi usman, can i ask the link of literature from this tutorial? beside the contents of Landsat product like LSE equation, proportion of vegetation, etc. thanks berfore :)
Good video thank !
Hello Usman. Can you please
make a video on how to calculate NDSI for glacier studies? I am new to
remote sensing and I haven't found people uploading videos where they
clearly explain how to calculate NDSI. I would appreciate if you could
make a video where you also show where to get the best data and how to
put algorithms using arc gis and/or qgis? If you know about a detailed explanation published as a video or a flow chart with pictures, please share.thank you
This formula will do what you want NDSI = (Band GREEN - Band SWIR) / (Band GREEN + Band SWIR).... Depending of the software you use, put this formula into a raster calculator and you are done. Good luck.
First i would like thank you for your amazing videos.
My question is that as my study area falls under multiple scenes due to large area. The sun elevation is different for different scenes. I want to know how to choose the sun elevation for different scenes as i have to mosiac them first.
Regards
John
Hello John, since you have multiple scenes it is best you treat each of them separately after which you can do the mosaic. This is part of beat practices when using remote sensing data, because each scene is taking at different times with different factors.
Sir can you kindly guide how to calculate mean LST for each class of LULC
Can we use this method just for Landsat 8? How about Landsat 5 and 7?
EXCUSE ME, Why did you put the value of radiance_MULT like 0.000342 not as it put it in fill?
Hello, Could you tell me how to construct the temperature distribution and concentration of chlorophyll a of water bodies?
Hi thks for the videoBy chance have you the procedure on paper already such as word?really appreciated
Ooh sorry I don't have it on paper.
The best
Thanks a lot
Sir how to calculate Air temperatures from LST
Dear Usman, firstly thank you for your nice work in this video. I am trying to do same operations on my images but I think I cannot succeed it so at stage of conversion from radiance to sat temp. my values range from -124.633 to 43.7934 for Band10 and similar results for Band11, is it possible or I am doing a mistake one of the stages. Could you help me about it.
+Sinan ARAS I had the same problem while later I found the extreme values such as -124 were for the marginal part of the segment without valid thermal infrared info.
Hey, I know this is an old post, but how did you end up fixing that problem?
Cut out only one area that you will use. The edges of the image have no ability to express themselves negatively.
Hi dear freand, this is a really great video. I hope you can help me: I have interest on using this method for LST for my thesis, do you have de article or bibliography that describes the methodology ?
Thanks!
How to do correction on band 10 and band 11 landsat?
Hello Usman i wana ask you how we can change the min and the max of the at satellite radiance temperature. Thanks
hii... I have encountered an error during the sattemp creation.... Temperature ranges from -1.6 to 42 which is not feasible in my study area. Can you guide me about where I would have gone wrong
hi Osman or any one know please can you explain how did you get the P value where you but 14380 and is this value is fixed can not be changed we can applied wherever we need thanks
Hello usman Bhari pls bro I want to know about LST prediction using ANN model or other