Landsat 8 Data & Reflectance Tutorial
Includes Landsat 8 DOS Method, COST, DOS, and TOA reflectance
Data download includes blue, green, red, near-infrared (NIR), short-wave infrared (SWIR), and NaturalLook imagery; guide and values to calculate reflectance are included below. See the information below about the easier Landsat 8 DOS surface reflectance method (the correct image-based atmospheric correction method for Landsat 8).
Data upon opening in free Quantum GIS; LandsatLook imagery is visible while band layers are turned off.
The Landsat 8 imagery is for 7/11/13, path 22 / row 33 (located in Southern half of Illinois in Corn Belt, USA). Imagery includes LandsatLook natural color (shown above), and DNs for band 2 (blue), band 3 (green), band 4 (red), band 5 (near infrared), and band 6 (short-wave infrared). The images can be seen in Landsat 8 Color and Band Graphics page and is the same used for the example in the Landsat 8 Atmospheric Correction page. Data is saved as a free Quantum GIS map project. If opening in Quantum GIS the map will appear as above, with a LandsatLook natural color image visible. Band layers (turned off) are all grayscale with lighter shades representing higher values scaled ± 2 standard deviations from the mean. If you are using other GIS software, such as ArcGIS, the imagery can be opened as individual layers (.tif files). Surfaces represent a wide ranges of reflectance in each band and include crop and non-crop vegetation, water, clouds, shadows, and urban areas.
Landsat 8 DOS Method
Important: Read the following two paragraphs about the easier DOS atmospheric correction surface reflectance method that should be used for Landsat 8 prior to viewing the tutorial below.
ESUN values needed to be established for Landsat 8 in order to determine if the COST or DOS method should be used to calculate surface reflectance (TOA is not surface reflectance). Correct ESUN values were calculated by GIS Ag Maps and are listed and described in the Landsat 8 ESUN Values section of the Landsat 8 ESUN, Radiance, and TOA Reflectance page. It has been determined by GIS Ag Maps, based on information shown in this website, that the DOS method is best for Landsat 8 surface reflectance; this makes atmospheric correction for Landsat 8 easier as is explained in the next paragraph. Data that show DOS should be used can be viewed on the following pages of this website: Landsat 8 DOS vs. COST NIR Reflectance Landsat 8 vs. Landsat 7 LEDAPS Reflectance Landsat 8 vs. Landsat 7 LEDAPS Reflectance for Larger Area.
The fact that DOS is best for Landsat 8 reduces the steps necessary in the atmospheric correction process; simply apply the steps that follow in this paragraph to convert to Landsat 8 DOS Method surface reflectance (Landsat 8 DOS Method was published by GIS Ag Maps at www.gisagmaps.com on 9/5/2013 and is based on the DOS method from Chavez ): 1) Convert the digital number (DN) values to TOA reflectance with the USGS Landsat 8 Conversion to TOA Reflectance Equation (USGS, 2013b; included near bottom of page), making sure to apply the sun angle correction factor. As of now, the following equation applies to all bands: ([DN * .00002) - .1] / [cosine of the solar zenith]). This will compute TOA reflectance for all bands. 2) Establish one-percent haze reflectance. First, establish the haze DN for each applicable band as shown in Step 2 of the tutorial below (this step involves the most time and there are different methods to determine haze amount). This is described in more detail in the Landsat 8 Atmospheric Correction page and in the Haze Selection Methods section of the Atmospheric Correction Guide. Apply the Conversion to TOA Reflectance Equation (same reflectance equation as above) to calculate haze reflectance from the DN (making sure to apply the sun angle correction factor). Subtract .01 from the haze amount to calculate one-percent haze reflectance. Subtracting .01 reflectance units will result in the established dark object subtraction pixel having one-percent reflectance and should be applied because of " the fact that very few targets on the Earth's surface are absolute black, so an assumed one-percent minimum reflectance is better than zero percent" (Chavez, 1996). 3) Subtract the one-percent haze amount (haze reflectance - .01) from all TOA reflectance values to calculate surface reflectance. Because DOS should be applied for Landsat 8 image-based atmospheric correction instead of COST, this much simpler method can be used to retrieve surface reflectance as opposed to using all the DOS steps shown in the Landsat 8 Atmospheric Correction page or in the tutorial below. The tutorial below describes and shows how to derive values of the above steps in more detail.
Based on the information in the tutorial below, the Landsat 8 imagery can be converted to COST, DOS, and TOA reflectance:
It is important to understand how the ESUN values were derived for this data (which are applied in the atmospheric correction process); ESUN values are described in the Landsat 8 ESUN, Radiance, and TOA Reflectance page. To GIS Ag Maps knowledge, this is the first Landsat 8 reflectance comparison made between the different reflectance models; so you can see first hand how they compare. The COST model has been shown more accurate than the DOS or TOA models with previous Landsats due to the inclusion of an additional cosine factor in the denominator; however, Landsat 8 has different bandwidths than previous Landsat satellites, particularly the NIR band. The accuracy of the models have been assessed by GIS Ag Maps and the DOS model has been shown to be more accurate based on comparisons to to known reflectance values and to LEDAPS reflectance as can be viewed through the following links: Landsat 8 COST vs. DOS NIR Reflectance Landsat 8 vs. Landsat 7 LEDAPS Soybean Reflectance Landsat 8 vs. Landsat 7 LEDAPS Reflectance
You can learn how to convert Landsat 8 digital numbers to image-based atmospherically corrected surface reflectance by following the steps in the tutorial and converting the Landsat 8 digital numbers (DNs) in the download. Landsat 7, 5, and 4 surface reflectance data can now be ordered for free (based on the LEDAPS method; pdf opens in new tab), so only Landsat 8 needs to be converted to surface reflectance (Landsat LEDAPS surface reflectance data can be accessed through the EarthExploerer link in the Data Sources folder to the left and is the recommended reflectance data for Landsat). The tutorial involves real atmospheric correction with real Landsat 8 imagery. Landsat DN data in the download can be converted to reflectance based on different models as explained in the tutorial; the necessary values are included. You can use the raster calculator in free Quantum GIS or in other GIS software. It is a good idea to verify your calculations one way or another during the process. The steps to derive reflectance are based on the example in the Landsat 8 Atmospheric Correction page for the COST, DOS, and TOA models. Reflectance has been calculated by GIS Ag Maps and statistics are provided at the end of the tutorial; your values should be the same or very similar.
Tutorial for Converting Landat DNs to Reflectance
Steps to convert digital numbers (DNs) to COST, DOS, and TOA reflectance for data download
The COST model (Chavez, 1996) for atmospheric correction can be written as:
(Eoλ is also referred to as ESUN; TAUv = 1.0 for Landsat; TAUz = cosθs for COST method)
The cosine of the solar zenith angle correction (COST) model (Chavez, 1996) is described in detail in the Atmospheric Correction Guide page. The "CO" is from cosine; the "S" is from solar; and the "T" is from "TAUz".
The dark-object subtraction (DOS) model (Chavez, 1988) is the above COST model with TAUz omitted.
The top of atmosphere (TOA) model (also known as apparent reflectance or in-band planetary albedo) is the above COST model with the deduction for scatter and TAUz omitted. It is important to realize that TOA reflectance does not represent surface reflectance. However, for Landsat 8 the USGS (2013) has developed a particular equation based on data in the .MTL file (the text for the .MTL for the data in this tutorial is included near bottom of page) that can also be applied to derive TOA reflectance; the equation is included at the bottom of the page.
Double asterisks ( * * ) below represent points in the process where a calculation needs to be made
Convert DNs to radiance (Lsatrad) for each band. Data for equations to convert to radiance are in .MTL file that is included with Landsat download from USGS. Equations used for the data download are shown at the end of Step 1.
Conversion to TOA Radiance per USGS (2013) is as follows:
OLI and TIRS band data can be converted to TOA spectral radiance using the radiance rescaling factors provided in the metadata file:
Lλ = MLQcal + AL
Lλ = TOA spectral radiance (Watts/( m2 * srad * μm))
ML = Band-specific multiplicative rescaling factor from the metadata (RADIANCE_MULT_BAND_x, where x is the band number)
AL = Band-specific additive rescaling factor from the metadata (RADIANCE_ADD_BAND_x, where x is the band number)
Qcal = Quantized and calibrated standard product pixel values (DN)
For the data download, DNs are converted to radiance based on the following from the .MTL file (need to check individual file for each scene as values may vary slightly):
RADIANCE_MULT_BAND_2 = 1.2732E-02 which equals .012732
RADIANCE_MULT_BAND_3 = 1.1658E-02 which equals .011658
RADIANCE_MULT_BAND_4 = 9.8736E-03 which equals .0098736
RADIANCE_MULT_BAND_5 = 5.9914E-03 which equals .0059914
RADIANCE_MULT_BAND_6 = 1.5095E-03 which equals .0015095
RADIANCE_ADD_BAND_2 = - 63.65864
RADIANCE_ADD_BAND_3 = - 58.28984
RADIANCE_ADD_BAND_4 = - 49.36793
RADIANCE_ADD_BAND_5 = - 29.95701
RADIANCE_ADD_BAND_6 = - 7.54767
* * Equations to covert to radiance (Lsatrad) based on the above follow (it is necessary to check the individual file for each scene as values may slightly vary). Calculation can be made with a raster calculator in GIS. Data can also be converted to a vector format in GIS and processed efficiently in a vector or spreadsheet environment if the file size is small enough. The data here have over a half million pixels. Free Quantum GIS and other GIS software, such as ArcGIS, have raster calculators.
Landsat 8 band 2 (blue): radiance = ( .012732 * DN ) - 63.65864
Landsat 8 band 3 (green): radiance = ( .011658 * DN ) - 58.28984
Landsat 8 band 4 (red): radiance = ( .0098736 * DN ) - 49.36793
Landsat 8 band 5 (NIR): radiance = ( .0059914 * DN ) - 29.95701
Landsat 8 band 6 (SWIR): radiance = ( .0015095 * DN ) - 7.54767
Step 2 (three parts)
Calculate Lhaze1%rad for bands for COST and DOS models. Part 3 shows amounts used for data download
Calculate DOS and COST haze radiance (also known as path radiance) and deduct the radiance that equals one-percent reflectance (based on specific equations listed below) for each band. This is explained in detail for this specific data below and in the Landat 8 Atmospheric Correction page.
Lhaze1%rad is the most time-consuming component of the equation. This represents the amount radiance is incorrectly increased at the sensor due to the effect of atmospheric haze (this amount is also known as path radiance); haze radiance needs to be deducted from total radiance of applicable bands. As solar radiation travels through space and enters Earth's atmosphere it can strike particles in the atmosphere and reflect back to the satellite sensor, erroneously increasing radiance values. The theory behind image-based correcting for the effect of haze is that within a Landsat scene (which contains millions of pixels) there should be a surface that reflects no radiation; the amount this surface is above zero reflectance is to due to haze. However, because of " the fact that very few targets on the Earth's surface are absolute black...an assumed one-percent minimum reflectance is better than zero percent" (Chavez, 1996). Subtracting .01 reflectance units from the haze reflectance will result in the established haze (dark object) as having one-percent reflectance. Establishing the haze amount will be detailed later in this section. There are different methods to establish a haze amount; it is not a rule to use the very lowest pixel value.
Calculate Haze Radiance (Lhaze) (also known as Path Radiance)
Different ways for calculating haze effect for bands are shown in the main Atmospheric Correction Guide page. Methods that applied to Landsat 4, 5, and 7 do not necessary apply to Landsat 8. It is not a rule that the lowest pixel value in the histogram should be selected as the haze amount (or dark object subtraction pixel value). For Landsat 8 haze selection here, values were used from a true visible band dark object area; these are represented by pixels that have very low values in all visible bands. The following graphic shows an area that has low values in the red, green, and blue bands (center-right); the dots are centered on pixels and represent red, green and blue bands.
Zoomed in to dark object area you can see there are many low values and that they correspond to a shadow in an area of vegetation.
The low value pixel area can be viewed with higher resolution imagery below (the LandsatLook imagery above is based on Landsat 30 x 30 meter pixels). The low value pixels above correspond to the center area of the image below (extent of the images are similar). It can be seen by viewing both images that the low values in the visible bands at the time of imaging correspond to green vegetation which is a high absorber of visible radiation (green vegetation reflects more green light than blue and red, but green reflectance is still low). The already low visible band reflectance amounts are lowered by the shadows from clouds, and possibly from trees and/or topography. For these reasons, it makes sense that this is an area of minimal visible light reflectance. NIR reflectance may not be the lowest here because of the associated vegetation; NIR can transmit through shadows and reflect the surface.
From the cluster of points in the dark object area above, the lowest DN in each band should be selected as the haze value (unless there is an obvious erroneous value) and converted to radiance (lowest values do not need to be from the same pixel but in the same dark object area). There should be a power relationship between band haze (path) radiance (Chavez, 1988). Longer wavelength will have relatively less haze. Clearer atmospheres will have a greater path radiance difference between bands. Hazier atmospheres will have more path radiance for all bands and the haze amounts will be more similar between bands. The haze radiance below shows that a strong power relationship exists. Haze deduction does not have to be applied to bands longer that NIR.
The following are the haze radiance values before the one-percent reduction as detailed in the Landsat 8 Atmospheric Correction page:
Band Band Center Haze DN Haze Radiance
2 (blue) .480 8289 41.876908
3 (green) .560 6993 23.234554
4 (red) .655 6140 11.255974
The band center and haze radiance values from above are plotted below with a power line.
If it is difficult to find pixels in dark object areas with your software, keep in mind that the values shown below (in the tables after the next paragraph) are very low. If you prefer, to establish haze amounts simply fit low radiance values from the blue, green, and red bands (as well as the NIR band, if applicable, as is described in the main Atmospheric Correction Guide page) to a power line with a high correlation, avoiding low values that have a large gap between them and the rest of the data (as the lowest values in the blue and green data below). This gap can commonly exist somewhere at the very low end of the histogram, it does not always exist between the lowest value and the rest of the data, but could be between a very low value and the rest of the data. Because values are low, make sure that the DNs you select for different band haze amounts correspond to radiances greater than zero; there may be some digital number near the low end of the histogram that correspond to negative radiance values. Negative radiance mainly applies to the NIR band however; is it unlikely for visible bands.
Landsat 8 low end of histogram data that show selected haze DNs; from left to right DNs are, band 2 (blue), 8289; band 3 (green), 6993; and band 4 (red), 6140. Landsat 8 data is based on a 12-bit dynamic range but are delivered as 16-bit images; previous Landsats were based on 8-bit for both. Landsat 8 DNs have a maximum value of 65,535 as opposed to previous Landsat which had a maximum value of 255. The number in the left column of each table is the ranking of the DN value (1 is the lowest DN in the entire scene); the number in the middle column is the DN value; and the number in the right column is the amount of the DN values there are in the entire scene.
If the histogram for band 5 (NIR) shows DN values that correspond to zero or negative radiance (there can be times where NIR or larger bands have some pixels that correspond to zero or negative radiance) such as it does in this example (5000 equals zero radiance), there is no need for haze deduction. Sometimes NIR should have a haze deduction and sometimes it should not. There is an example where NIR haze should be deducted in the Atmospheric Correction Guide.
Band 5 low end of histogram (5000 equals zero radiance)
As previously mentioned, bandwidths longer than NIR (such SWIR) do not need haze deducted.
As described in the Landsat 8 Atmospheric Correction page, and as is the case here, NIR may not need haze removed. (For bands longer than NIR, such as band 6 (SWIR) in this case, there is no haze deduction necessary.) For this scene, there are zero and negative radiance values for NIR in this case (there can be times where NIR or larger bands have some pixels that correspond to zero or negative radiance). If there are zero or negative radiance values, there is no haze deduction necessary. As previously mentioned, sometimes NIR will have radiance amounts over zero where a haze deduction should be made and sometimes it will not; an example of NIR haze deduction is shown in the Atmospheric Correction Guide page.
Calculate Lhaze1%rad; this is haze radiance minus the radiance that derives one-percent reflectance
The following equations should be used to calculate the radiance that derives one-percent reflectance when input into the COST or DOS model. The amount should be deducted from the haze radiance of each band, that way the pixel value that has been established as the darkest pixel (dark object subtraction pixel) will have one-percent reflectance instead of zero when the atmospheric correction process is complete.
One-percent reflectance equations:
COST model one-percent reflectance equation (ARSC, 2002) = 0.01 x ([Eoλ x cosθs²] / [d² x pi]);
DOS model one-percent reflectance equation (a cosθs value is omitted) = 0.01 x ([Eoλ x cosθs] / [d² x pi]).
The values to complete the one-percent reflectance equations follow:
Landsat 8 Eoλ, also known as ESUNλ, have not been made publicly available; however, they have been derived by GIS Ag Maps. The Landsat 8 ESUN, Radiance, & TOA Reflectance page describes in detail how the derived ESUN values were calculated; it is important to view the page to understand the basis of the values. The derived ESUN values for bands 2 - 6 (the data in the download) are as follows:
The cosθs value is the cosine of the solar zenith; the solar zenith = 90⁰ - solar elevation. The solar elevation of the scene center is included in the .MTL file (there may be an online tool to find a more local value). For the data here, cosθs = .90908487. This value needs to be squared for the COST equation but not for the DOS equation.
The d value represents the Earth-sun distance and is included in the .MTL file; it should be squared in all cases for atmospheric correction. For the data here, d² = 1.0334727
pi = 3.14159265
For the COST and DOS models, the one-percent deduct amounts for bands 2, 3, and 4 are as follows:
For the COST and DOS models, the haze radiance from Part 1 minus the one-percent deduction amounts for bands 2 (blue), 3 (green), and 4 (NIR) from above need to be calculated and are as follows:
Calculate (Lsatrad - Lhaze1%rad)
* * To derive (Lsatrad - Lhaze1%rad), the one-percent haze amounts above (Step 2: Part 3) need to be subtracted from the radiance calculated in Step 1. Each raster of band radiance can have the Lhaze1%rad values subtracted for a COST or DOS model with a raster calculator in GIS.
Calculate the numerator of COST or DOS model
* * Multiply the rasters from Step 3 by pi ( 3.14159265 ) and d² ( 1.0334727 [for this particular imagery]) to derived a raster that represents the numerator of the atmospheric correction equations (raster x pi x d²). For bands 5 and 6 there was no haze deduction so multiply the radiance rasters from Step 1.
For the TOA numerator ignore the haze deduction, and simply multiply radiance from Step 1 by pi and d².
* * Calculate surface reflectance (ρλ) by dividing the numerator in Step 4 by the denominator of COST, DOS, or TOA model shown below (all values have been included below to make calculations)
The cosine of the solar zenith angle correction (COST) model (Chavez, 1996) can be written as:
(Components of equation are described below; "CO" is from cosine; "S" is from solar; the "T" is from "TAUz".)
The dark-object subtraction (DOS) model (Chavez, 1988) is the above COST model with TAUz omittedand is the correct equation for certain situations; for example, this can be used when solar elevation is < 45° and/or the surface reflectance is not within vegetation and soil ranges.
The top of atmosphere (TOA) model (also known as apparent reflectance or in-band planetary albedo) is the above COST model with the deduction for scatter and TAUz omitted and is the correct equation for certain situations; for example, this can be used when solar elevation is < 45° and/or the surface reflectance is not within vegetation and soil ranges and if the band is mid-infrared or larger because scatter is insignificant enough.
Divide the numerator raster from Step 4 by the denominator to calculate atmospherically corrected surface reflectance. This can be done with raster calculator. The difference between the COST and DOS denominators is that the COST model has (cosine of the solar zenith angle)².
For the COST model, the denominator = [ESUN value for each band shown in Step 2: Part 2] x cosθs² x d²
For the DOS and TOA models, the denominator = [ESUN value for each band shown in Step 2: Part 2] x cosθs x d²
ESUN values for bands 2 - 6 (shown in Step 2: Part 2) are as follows:
The cosθs value is described and shown in Step 2: Part 2 and = .90908487 ; cosθs² [for COST] = .82643530
The d² value is described and shown in Step 2: Part 2 and = 1.0334727
Means have been calculated for the tutorial data and are as follows:
(Values are in reflectance [ratio] units, multiply x 100 for percent. Visible band reflectance is low because much of the area is vegetation which absorbs over 90 percent of visible light, while NIR reflectance is much higher because of vegetation.)
The below values have been updated to show reflectance values that result from entering precisely the same values from the tutorial (same amount of values after decimal point). If your results are different when calculating with raster calculator, take the mean of the DN raster (in Quantum GIS, you can right-click on raster layer, then click Properties then Metadata to find mean values) and input it into an Excel spreadsheet (or other spreadsheet) to check raster calculator results.
|Band 5||0.455||0.414 *||0.414 *|
|Band 6||0.226||0.205 *||0.205 *|
* Bands 5 (NIR) and 6 (SWIR) did not have a haze deduction as explained in the tutorial (NIR sometimes does; haze deduction is not necessary for SWIR), so the DOS and standard TOA reflectance equations are the same. TOA reflectance was calculated with the standard TOA equation instead of the USGS Landsat 8 equation (shown below). The Landsat 8 TOA equation may derive very slightly different reflectance than the standard TOA equation (values at the 10,000th place may be slightly different).
TOA Reflectance Equation for Landsat 8 (USGS, 2013)
The USGS has developed an equation (shown below) to derive TOA reflectance for Landsat 8 that can be used instead of the standard TOA equation described at top of this section.
Conversion to TOA Reflectance
OLI band data can also be converted to TOA planetary reflectance using reflectance rescaling coefficients provided in the product metadata file (MTL file). The following equation is used to convert DN values to TOA reflectance for OLI data as follows:
ρλ' = MρQcal + Aρ
ρλ' = TOA planetary reflectance, without correction for solar angle. Note that ρλ' does not contain a correction for the sun angle.
Mρ = Band-specific multiplicative rescaling factor from the metadata (REFLECTANCE_MULT_BAND_x, where x is the band number)
Aρ = Band-specific additive rescaling factor from the metadata (REFLECTANCE_ADD_BAND_x, where x is the band number)
Qcal = Quantized and calibrated standard product pixel values (DN)
TOA reflectance with a correction for the sun angle is then:
ρλ = TOA planetary reflectance
θSE = Local sun elevation angle. The scene center sun elevation angle in degrees is provided in the metadata (SUN_ELEVATION).
θSZ = Local solar zenith angle; θSZ = 90° - θSE
For more accurate reflectance calculations, per pixel solar angles could be used instead of the scene center solar angle, but per pixel solar zenith angles are not currently provided with the Landsat 8 products.
Landsat 8 .MTL file text for imagery in tutorial
GROUP = L1_METADATA_FILE
GROUP = METADATA_FILE_INFO
ORIGIN = "Image courtesy of the U.S. Geological Survey"
REQUEST_ID = "0501307110212_00017"
LANDSAT_SCENE_ID = "LC80220332013192LGN00"
FILE_DATE = 2013-07-11T19:40:19Z
STATION_ID = "LGN"
PROCESSING_SOFTWARE_VERSION = "LPGS_2.2.2"
END_GROUP = METADATA_FILE_INFO
GROUP = PRODUCT_METADATA
DATA_TYPE = "L1T"
ELEVATION_SOURCE = "GLS2000"
OUTPUT_FORMAT = "GEOTIFF"
SPACECRAFT_ID = "LANDSAT_8"
SENSOR_ID = "OLI_TIRS"
WRS_PATH = 22
WRS_ROW = 33
NADIR_OFFNADIR = "NADIR"
TARGET_WRS_PATH = 22
TARGET_WRS_ROW = 33
DATE_ACQUIRED = 2013-07-11
SCENE_CENTER_TIME = 16:31:42.2302176Z
CORNER_UL_LAT_PRODUCT = 39.95144
CORNER_UL_LON_PRODUCT = -89.06033
CORNER_UR_LAT_PRODUCT = 39.96784
CORNER_UR_LON_PRODUCT = -86.33490
CORNER_LL_LAT_PRODUCT = 37.81705
CORNER_LL_LON_PRODUCT = -88.99956
CORNER_LR_LAT_PRODUCT = 37.83226
CORNER_LR_LON_PRODUCT = -86.35453
CORNER_UL_PROJECTION_X_PRODUCT = 324000.000
CORNER_UL_PROJECTION_Y_PRODUCT = 4424400.000
CORNER_UR_PROJECTION_X_PRODUCT = 556800.000
CORNER_UR_PROJECTION_Y_PRODUCT = 4424400.000
CORNER_LL_PROJECTION_X_PRODUCT = 324000.000
CORNER_LL_PROJECTION_Y_PRODUCT = 4187400.000
CORNER_LR_PROJECTION_X_PRODUCT = 556800.000
CORNER_LR_PROJECTION_Y_PRODUCT = 4187400.000
PANCHROMATIC_LINES = 15801
PANCHROMATIC_SAMPLES = 15521
REFLECTIVE_LINES = 7901
REFLECTIVE_SAMPLES = 7761
THERMAL_LINES = 7901
THERMAL_SAMPLES = 7761
FILE_NAME_BAND_1 = "LC80220332013192LGN00_B1.TIF"
FILE_NAME_BAND_2 = "LC80220332013192LGN00_B2.TIF"
FILE_NAME_BAND_3 = "LC80220332013192LGN00_B3.TIF"
FILE_NAME_BAND_4 = "LC80220332013192LGN00_B4.TIF"
FILE_NAME_BAND_5 = "LC80220332013192LGN00_B5.TIF"
FILE_NAME_BAND_6 = "LC80220332013192LGN00_B6.TIF"
FILE_NAME_BAND_7 = "LC80220332013192LGN00_B7.TIF"
FILE_NAME_BAND_8 = "LC80220332013192LGN00_B8.TIF"
FILE_NAME_BAND_9 = "LC80220332013192LGN00_B9.TIF"
FILE_NAME_BAND_10 = "LC80220332013192LGN00_B10.TIF"
FILE_NAME_BAND_11 = "LC80220332013192LGN00_B11.TIF"
FILE_NAME_BAND_QUALITY = "LC80220332013192LGN00_BQA.TIF"
METADATA_FILE_NAME = "LC80220332013192LGN00_MTL.txt"
BPF_NAME_OLI = "LO8BPF20130711161958_20130711163900.01"
BPF_NAME_TIRS = "LT8BPF20130711161604_20130711163953.01"
CPF_NAME = "L8CPF20130701_20130930.01"
RLUT_FILE_NAME = "L8RLUT20130211_20431231v06.h5"
END_GROUP = PRODUCT_METADATA
GROUP = IMAGE_ATTRIBUTES
CLOUD_COVER = 3.56
IMAGE_QUALITY_OLI = 9
IMAGE_QUALITY_TIRS = 9
ROLL_ANGLE = -0.001
SUN_AZIMUTH = 126.58978669
SUN_ELEVATION = 65.37919226
EARTH_SUN_DISTANCE = 1.0165986
GROUND_CONTROL_POINTS_MODEL = 340
GEOMETRIC_RMSE_MODEL = 7.228
GEOMETRIC_RMSE_MODEL_Y = 4.876
GEOMETRIC_RMSE_MODEL_X = 5.336
GROUND_CONTROL_POINTS_VERIFY = 101
GEOMETRIC_RMSE_VERIFY = 3.739
END_GROUP = IMAGE_ATTRIBUTES
GROUP = MIN_MAX_RADIANCE
RADIANCE_MAXIMUM_BAND_1 = 755.79810
RADIANCE_MINIMUM_BAND_1 = -62.41405
RADIANCE_MAXIMUM_BAND_2 = 770.71515
RADIANCE_MINIMUM_BAND_2 = -63.64591
RADIANCE_MAXIMUM_BAND_3 = 705.71509
RADIANCE_MINIMUM_BAND_3 = -58.27818
RADIANCE_MAXIMUM_BAND_4 = 597.69751
RADIANCE_MINIMUM_BAND_4 = -49.35805
RADIANCE_MAXIMUM_BAND_5 = 362.68948
RADIANCE_MINIMUM_BAND_5 = -29.95102
RADIANCE_MAXIMUM_BAND_6 = 91.37965
RADIANCE_MINIMUM_BAND_6 = -7.54616
RADIANCE_MAXIMUM_BAND_7 = 29.72563
RADIANCE_MINIMUM_BAND_7 = -2.45475
RADIANCE_MAXIMUM_BAND_8 = 673.26843
RADIANCE_MINIMUM_BAND_8 = -55.59872
RADIANCE_MAXIMUM_BAND_9 = 149.04416
RADIANCE_MINIMUM_BAND_9 = -12.30812
RADIANCE_MAXIMUM_BAND_10 = 22.00180
RADIANCE_MINIMUM_BAND_10 = 0.10033
RADIANCE_MAXIMUM_BAND_11 = 22.00180
RADIANCE_MINIMUM_BAND_11 = 0.10033
END_GROUP = MIN_MAX_RADIANCE
GROUP = MIN_MAX_REFLECTANCE
REFLECTANCE_MAXIMUM_BAND_1 = 1.210700
REFLECTANCE_MINIMUM_BAND_1 = -0.099980
REFLECTANCE_MAXIMUM_BAND_2 = 1.210700
REFLECTANCE_MINIMUM_BAND_2 = -0.099980
REFLECTANCE_MAXIMUM_BAND_3 = 1.210700
REFLECTANCE_MINIMUM_BAND_3 = -0.099980
REFLECTANCE_MAXIMUM_BAND_4 = 1.210700
REFLECTANCE_MINIMUM_BAND_4 = -0.099980
REFLECTANCE_MAXIMUM_BAND_5 = 1.210700
REFLECTANCE_MINIMUM_BAND_5 = -0.099980
REFLECTANCE_MAXIMUM_BAND_6 = 1.210700
REFLECTANCE_MINIMUM_BAND_6 = -0.099980
REFLECTANCE_MAXIMUM_BAND_7 = 1.210700
REFLECTANCE_MINIMUM_BAND_7 = -0.099980
REFLECTANCE_MAXIMUM_BAND_8 = 1.210700
REFLECTANCE_MINIMUM_BAND_8 = -0.099980
REFLECTANCE_MAXIMUM_BAND_9 = 1.210700
REFLECTANCE_MINIMUM_BAND_9 = -0.099980
END_GROUP = MIN_MAX_REFLECTANCE
GROUP = MIN_MAX_PIXEL_VALUE
QUANTIZE_CAL_MAX_BAND_1 = 65535
QUANTIZE_CAL_MIN_BAND_1 = 1
QUANTIZE_CAL_MAX_BAND_2 = 65535
QUANTIZE_CAL_MIN_BAND_2 = 1
QUANTIZE_CAL_MAX_BAND_3 = 65535
QUANTIZE_CAL_MIN_BAND_3 = 1
QUANTIZE_CAL_MAX_BAND_4 = 65535
QUANTIZE_CAL_MIN_BAND_4 = 1
QUANTIZE_CAL_MAX_BAND_5 = 65535
QUANTIZE_CAL_MIN_BAND_5 = 1
QUANTIZE_CAL_MAX_BAND_6 = 65535
QUANTIZE_CAL_MIN_BAND_6 = 1
QUANTIZE_CAL_MAX_BAND_7 = 65535
QUANTIZE_CAL_MIN_BAND_7 = 1
QUANTIZE_CAL_MAX_BAND_8 = 65535
QUANTIZE_CAL_MIN_BAND_8 = 1
QUANTIZE_CAL_MAX_BAND_9 = 65535
QUANTIZE_CAL_MIN_BAND_9 = 1
QUANTIZE_CAL_MAX_BAND_10 = 65535
QUANTIZE_CAL_MIN_BAND_10 = 1
QUANTIZE_CAL_MAX_BAND_11 = 65535
QUANTIZE_CAL_MIN_BAND_11 = 1
END_GROUP = MIN_MAX_PIXEL_VALUE
GROUP = RADIOMETRIC_RESCALING
RADIANCE_MULT_BAND_1 = 1.2485E-02
RADIANCE_MULT_BAND_2 = 1.2732E-02
RADIANCE_MULT_BAND_3 = 1.1658E-02
RADIANCE_MULT_BAND_4 = 9.8736E-03
RADIANCE_MULT_BAND_5 = 5.9914E-03
RADIANCE_MULT_BAND_6 = 1.5095E-03
RADIANCE_MULT_BAND_7 = 4.9105E-04
RADIANCE_MULT_BAND_8 = 1.1122E-02
RADIANCE_MULT_BAND_9 = 2.4621E-03
RADIANCE_MULT_BAND_10 = 3.3420E-04
RADIANCE_MULT_BAND_11 = 3.3420E-04
RADIANCE_ADD_BAND_1 = -62.42654
RADIANCE_ADD_BAND_2 = -63.65864
RADIANCE_ADD_BAND_3 = -58.28984
RADIANCE_ADD_BAND_4 = -49.36793
RADIANCE_ADD_BAND_5 = -29.95701
RADIANCE_ADD_BAND_6 = -7.54767
RADIANCE_ADD_BAND_7 = -2.45524
RADIANCE_ADD_BAND_8 = -55.60985
RADIANCE_ADD_BAND_9 = -12.31058
RADIANCE_ADD_BAND_10 = 0.10000
RADIANCE_ADD_BAND_11 = 0.10000
REFLECTANCE_MULT_BAND_1 = 2.0000E-05
REFLECTANCE_MULT_BAND_2 = 2.0000E-05
REFLECTANCE_MULT_BAND_3 = 2.0000E-05
REFLECTANCE_MULT_BAND_4 = 2.0000E-05
REFLECTANCE_MULT_BAND_5 = 2.0000E-05
REFLECTANCE_MULT_BAND_6 = 2.0000E-05
REFLECTANCE_MULT_BAND_7 = 2.0000E-05
REFLECTANCE_MULT_BAND_8 = 2.0000E-05
REFLECTANCE_MULT_BAND_9 = 2.0000E-05
REFLECTANCE_ADD_BAND_1 = -0.100000
REFLECTANCE_ADD_BAND_2 = -0.100000
REFLECTANCE_ADD_BAND_3 = -0.100000
REFLECTANCE_ADD_BAND_4 = -0.100000
REFLECTANCE_ADD_BAND_5 = -0.100000
REFLECTANCE_ADD_BAND_6 = -0.100000
REFLECTANCE_ADD_BAND_7 = -0.100000
REFLECTANCE_ADD_BAND_8 = -0.100000
REFLECTANCE_ADD_BAND_9 = -0.100000
END_GROUP = RADIOMETRIC_RESCALING
GROUP = TIRS_THERMAL_CONSTANTS
K1_CONSTANT_BAND_10 = 774.89
K1_CONSTANT_BAND_11 = 480.89
K2_CONSTANT_BAND_10 = 1321.08
K2_CONSTANT_BAND_11 = 1201.14
END_GROUP = TIRS_THERMAL_CONSTANTS
GROUP = PROJECTION_PARAMETERS
MAP_PROJECTION = "UTM"
DATUM = "WGS84"
ELLIPSOID = "WGS84"
UTM_ZONE = 16
GRID_CELL_SIZE_PANCHROMATIC = 15.00
GRID_CELL_SIZE_REFLECTIVE = 30.00
GRID_CELL_SIZE_THERMAL = 30.00
ORIENTATION = "NORTH_UP"
RESAMPLING_OPTION = "CUBIC_CONVOLUTION"
END_GROUP = PROJECTION_PARAMETERS
END_GROUP = L1_METADATA_FILE
ARSC. 2002. Arizona Remote Sensing Center: Landsat 5 atmospheric and radiometric correction. Information on website adapted from Skirvin, S (2000). Cited at: http://arsc.arid.arizona.edu/resources/image_processing/landsat/ls5-atmo.html. Last accessed: July, 2011.
Chavez, P.S., Jr. 1996. Image-based atmospheric corrections–revisited and improved. Photogrammetric Engineering and Remote Sensing 62(9): pp.1025-1036.
Chavez, P.S., Jr. 1988. An improved dark-object subtraction technique for atmospheric scattering correction of multispectral data. Remote Sensing of Environment 24: pp.459-479.
USGS. 2013. Landsat Missions: Frequently Asked Questions About the Landsat Missions. USGS. Last modified: 5/30/123. Cited at: http://landsat.usgs.gov/band_designations_landsat_satellites.php
USGS. 2013b. Using the USGS Landsat 8 Product. Cited at: http://landsat7.usgs.gov/Landsat8_Using_Product.php