Doing NDVI when nir data is given as 3 bands

Doing NDVI when nir data is given as 3 bands

I have been provided with two data sets of the same area, one with normal RGB bands and one with NIR data. My problem is that since the NIR data is given as 3 bands, i am not sure how I should interpret this.

I am guessing that its normal in some data acquisition areas to collect NIR data with a DSL camera with some filters removed and thats why the data is captured over 3 bands. (but i really dont know much about this).

How do i transform the 3 bands of nir data into the one I need for my NDVI calculation?

Sometimes distributors drop the blue band and provide only the nIR, red and green bands so that the end user can view the data as a false color composite image. Let's assume the distributor did that. There are several ways you can deduce which bands are which, especially at the red and nIR wavelengths, using basic remote sensing principles. For example, we know that EMR in the nIR portion of the spectrum is highly reflected off healthy green vegetation and highly absorbed in water. We also know that EMR in the red portion of the spectrum is generally absorbed by healthy green vegetation. This difference effect is what makes the NDVI such a valuable vegetation index.

The following examples show the spectral properties (measured in DN here) of water in the first screenshot and healthy green vegetation in the second screenshot. Based on this analysis, we can tell that the nIR band is layer 4 and the red band is layer 1.

Temporal relationships between time series CHIRPS-rainfall estimation and eMODIS-NDVI satellite images in Amhara Region, Ethiopia

Agenagnew A. Gessesse , Assefa M. Melesse , in Extreme Hydrology and Climate Variability , 2019 NDVI reflectance and anomaly

The NDVI is computed as the difference between near-infrared (NIR) and red (RED) reflectance divided by their sum.

NDVIi represents smoothed NDVI (sNDVI) observed at time step i and their ratio yields a measure of photosynthetic activity within values between − 1 and 1. Low NDVI values indicate moisture-stressed vegetation and higher values indicate a higher density of green vegetation. It is also used for drought monitoring and famine early warning ( Wardlow et al., 2007 Javadnia et al., 2009 ).

For computing NDVI anomaly (z-score) which was to require a series of images with historical year [long-term averages (LTAs)] for each period (in our case: month) in the year. The anomaly (z-score) indicators of vegetation condition can be calculated as ZNDVI and its value is widely used for monitoring vegetation anomalies ( Klisch and Atzberger, 2016 ). It is calculated at pixel level from the sNDVI data as.

where ZNDVI is the standard difference (z-score) of NDVI at time step i, NDVIi = sNDVI is the observed at time step i, NDVI mean , m = ∑ i NDVI i / m is the monthly mean sNDVI values, and σNDVI is the standard deviation of sNDVI values at month (m) respectively. σNDVI indicates the signed number of standard deviations is above or below the mean.

NDVI Processing

NDVI, or Normalized Difference Vegetation Index, can be a great tool for gauging the health of of plants by remote means. Maps Made Easy is designed to process calibrated NDVI imagery only.

With the proper equipment, NDVI imagery can be collected with a drone and processed to create map overlays that can give a wealth of potentially cost saving information.

"Proper Equipment" is the key phrase here. There is no shortage of companies who are willing to sell users "Ag" or "NDVI" cameras that are nothing more than hacked cameras with the IR blocking filters removed and a blue filter installed. We will not link to them here because they are all very misleading in the capabilities and accuracy. Accurate NDVI data processing requires the capability to calibrate the camera and the imagery to ensure that the ratios of near infrared light and visible light throughputs are known and can be appropriately scaled. Hacked consumer cameras cannot do this and a scientific-grade camera must be used for accurate results. (We are purposefully leaving spectral content out of this for now to keep this article easy to read.)

The reason Drones Made Easy, a company founded by a team of military multispectral imaging systems experts, does not currently offer any cheap NDVI solutions is because they are all currently not scientifically accurate. We do now offer a Parrot Sequoia based system but it is not exactly cheap. With NDVI, you can pick cheap or you can pick good, as things are right now you can't have both. The Drones Made Easy Agronaut system is the cheapest system available to include a flight control app, unlimited processing and a real NDVI camera: Drones Made Easy Agronaut Agricultural Mapping System

NDVI cameras are expensive. Currently, the most inexpensive option for real NDVI is the Sequoia by Parrot at

$3500. Also available are the TetraCam's ADC Snap which costs

$4500 to MicaSense's RedEdge

$6200 camera, few people are willing to make such an investment for something that they don't fully understand when it seems like you can just buy one of these hacked cameras and get similar results.

Below is a sample of the Sequoia camera being used to make absolute measurements over time. Three visits to the same area on April 1, July 8 and August 12 show how the vegetation has dried out. These layers were all taken under pretty different light conditions and you can see it does a really good job of maintaining constant values for unchanged areas. RGB and NDVI layers for each date are shown.

The data was take using the Drones Made Easy Agronaut system which is a DJI Matrice 100 with some custom electronics to drive a H4-3D gimbal to stabilize the camera. The custom on board computer also does the data preparation for upload to Maps Made Easy.

For further examples of the data that is taken by the Agronaut and how it is processed, see our article about NDVI Image Stack Processing.

Notice the full range of red to dark green that are representative of values that range from -.3 to 1. This is a good range and you can see that dead/non-plant objects, like road, are red. Certainly not healthy plants. The plant areas of the fields and trees are green with values from .3 to .9 representing active photosynthesis.

The numbers ranging from -1 to +1 at the bottom of the map show the scale of the index values as calculated using the following formula:

NDVI = (NIR - VIS) / (NIR + VIS)

The ND in NDVI is "Normalized Difference" and means that this is a self correcting measurement of the ratios of near infrared (NIR) and visible (VIS) light. The full range of values should take up a good amount of the space of the range from -1 to 1.

In hacked cameras, it is common for the amount of NIR light being accepted by the camera to be far different than that being collected in the visible which results in the range being calculated having a range of only .2 or .3 of the full range. If the whole image is green, does that mean that everything is healthy? Nope. It means the camera (or data wasn't calibrated properly and the ratios that got calculated do not represent what is actually going on on the ground.

Here is an example of a poorly calibrated NDVI camera:

Notice that the range is between .8 and 1. This camera was allowing way too much NIR light in or was not properly calibrated to account for the large discrepancy between the NIR and VIS light levels.

The numbers between -1 and 1 actually mean something. You can't just stretch values that were calculated as ranging from .2 to .5 over the -1 to 1 range and call it accurate NDVI data. It may look nice and have some red and some green in it, but it won't MEAN anything. The information created by doing this is not actionable intelligence. If a bad NDVI mapper gives a farmer such information and he acts on it, it will likely kill his whole crop. Luckily, farmers are smart and know better. This is where Ground Truthing comes in.

The Workaround - Manual Ground Truthing

If you absolutely must use and uncalibrated camera, you can still use Maps Made Easy to stitch and calculate your index imagery.

Our tone mapping (which can be adjusted) assumes that the NDVI calculated values of 0 or less are dead or inanimate. Values of .3 or higher are generally in the healthy range. This is the science behind using NDVI as a indicator of plant health.

Custom Tone Mapping

Tone mapping is the process where we apply a color to a calculate value. A value of .8 is nice and healthy so we give it a dark green. A value of -.3 is dead so we give it a dark red. Through the following steps is is possible to

  • Make sure the camera's white balance is fixed if you can.
  • Use a fixed exposure time, if you can.
  • Take some close up ground truth images of known healthy and known dead areas (like bare dirt).
  • Manually calculate a few values for known good and known bad surfaces using a free tool like ImageJ to get the raw RGB values.
  • Determine what your captured range is.
  • Adjust the tone mapping table at the time of upload by adjusting the "NDVI Settings" to reflect your custom range.

The color map on the right is drawn from the standard values Tetracam uses that span the full range of -1 to +1. The modified color map on the right will show the compressed value range of .25 to .8 as colors ranging from red to green. This modification needs to be made at the time of upload.

After custom tone mapping, the data will still not mean anything. The values will still be uncalibrated but the imagery will at least show the colors you want to map. Maps Made Easy does not, and will not, stretch the calculated values in order to have the data "look" correct. Data is correct or it is not.

We have never been shy about sharing our opinions on the mis-service these vendors are doing the industry. If too many people start sharing this bad data commercially, the Agricultural industry will very quickly get turned off to the technology when in fact it could be a great tool as long as people do it properly.

Normalized Difference Moisture Index

This image displays a (left) Landsat 8 Surface Reflectance (SR) and (right) the SR-derived Landsat Surface Reflectance Normalized Difference Moisture Index (NDMI).

Landsat Surface Reflectance-derived Normalized Difference Moisture Index (NDMI) are acquired from Landsat 4–5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI)/Thermal Infrared Sensor (TIRS) scenes that can be successfully processed to Landsat Level-2 Surface Reflectance products.

NDMI is used to determine vegetation water content. It is calculated as a ratio between the NIR and SWIR values in traditional fashion.

In Landsat 4-7, NDMI = (Band 4 – Band 5) / (Band 4 + Band 5).

In Landsat 8, NDMI = (Band 5 – Band 6) / (Band 5 + Band 6).

NDMI is delivered as a single band product specified as shown in the table below.

Landsat Surface Reflectance-derived Normalized Difference Moisture Index (NDMI) Specifications
Attribute Value
Long Name Normalized Difference Moisture Index
File Name *_sr_ndmi.tif
Data Type Signed 16-bit Integer
Units Spectral Index (Band Ratio)
Valid Range -10,000 — 10,000
Fill Value -9999
Saturate Value 20,000
Scale Factor *0.0001

Data Access

Visit the Landsat Surface Reflectance-Derived Spectral Indices webpage for information on product contraints, citation, and reference information.

NDVI, the Foundation for Remote Sensing Phenology

Remote sensing phenology studies use data gathered by satellite sensors that measure wavelengths of light absorbed and reflected by green plants. Certain pigments in plant leaves strongly absorb wavelengths of visible (red) light. The leaves themselves strongly reflect wavelengths of near-infrared light, which is invisible to human eyes. As a plant canopy changes from early spring growth to late-season maturity and senescence, these reflectance properties also change.

A field of sunflowers near Midland, South Dakota.

(Credit: Stephen P. Shivers, USGS. Public domain.)

Many sensors carried aboard satellites measure red and near-infrared light waves reflected by land surfaces. Using mathematical formulas (algorithms), scientists transform raw satellite data about these light waves into vegetation indices. A vegetation index is an indicator that describes the greenness — the relative density and health of vegetation — for each picture element, or pixel, in a satellite image.

Although there are several vegetation indices, one of the most widely used is the Normalized Difference Vegetation Index (NDVI). NDVI values range from +1.0 to -1.0. Areas of barren rock, sand, or snow usually show very low NDVI values (for example, 0.1 or less). Sparse vegetation such as shrubs and grasslands or senescing crops may result in moderate NDVI values (approximately 0.2 to 0.5). High NDVI values (approximately 0.6 to 0.9) correspond to dense vegetation such as that found in temperate and tropical forests or crops at their peak growth stage.

By transforming raw satellite data into NDVI values, researchers can create images and other products that give a rough measure of vegetation type, amount, and condition on land surfaces around the world. NDVI is especially useful for continental- to global-scale vegetation monitoring because it can compensate for changing illumination conditions, surface slope, and viewing angle. That said, NDVI does tend to saturate over dense vegetation and is sensitive to underlying soil color.

NDVI values can be averaged over time to establish "normal" growing conditions in a region for a given time of year. Further analysis can then characterize the health of vegetation in that place relative to the norm. When analyzed through time, NDVI can reveal where vegetation is thriving and where it is under stress, as well as changes in vegetation due to human activities such as deforestation, natural disturbances such as wild fires, or changes in plants' phenological stage.

USGS EROS Archive - Vegetation Monitoring - EROS Visible Infrared Imaging Radiometer Suite (eVIIRS)

The Earth Resources Observation and Science (EROS) Center Visible Infrared Imaging Radiometer Suite (eVIIRS) collection is based on the Suomi National Polar-orbiting Partnership (S-NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) data acquired by the NPP, which is the result of a partnership between the National Aeronautics and Space Administration (NASA) and the National Oceanic and Atmospheric Administration (NOAA).

eVIIRS (EROS Visible Infrared Imaging Radiometer Suite) CONUS Sub-Sample
(Public domain)

The EROS Visible Infrared Imaging Radiometer Suite (eVIIRS) Normalized Difference Vegetation Index (NDVI) collection uses the Visible Infrared Imaging Radiometer Suite (VIIRS) collection that is available at NASA’s Land Atmosphere Near-real-time Capability for EOS (LANCE) for the expedited product and Level-1 and Atmosphere Archive & Distribution System (LAADS) for the historical products.

Moderate resolution remote sensing provides a means for operational monitoring communities to develop historical trend information and use near-real-time deviations from temporal averages to identify areas of change. High quality, consistent and well- calibrated satellite measurements are needed to detect and monitor changes and trends, especially in vegetation patterns useful for drought, crop yield, phenology and fire potential studies.

The eVIIRS collection is based on the S-NPP VIIRS data acquired by the NPP. Even though VIIRS NDVI data are available as composites in the LP DAAC as 500-m and 1-km 8-day products, the eVIIRS product addresses the need for a 375-m dataset in the 7-, 14-, as well as 10-day dekadal datasets to match EROS Moderate Resolution Imaging Spectroradiometer (eMODIS).

The historical eVIIRS suite of products includes 7- and 14-day composites for the conterminous U.S. (CONUS) and Alaska and on a monthly 10-day dekadal schedule for the NDVI Famine Early Warning Systems (FEWS) regions including Central America/Caribbean and Mexico, Africa, Central Asia, South America, Australia and New Zealand. Expedited (near-real-time) production runs daily for the 7-day (CONUS) products. Each dataset delivers acquisition, quality, and NDVI information at 375-m spatial resolution. Because each of the composites are created from a varying number of images, eVIIRS composites include acquisition files to identify which of the inputs were used to populate the final composite. The metadata accompanying the data files summarize geographic bounds, projection parameters and product contact information.

eVIIRS currently produces NDVI and surface reflectance composites over CONUS, Alaska, and Central America/Caribbean and Mexico, whereas Africa, Central Asia, South America, Australia, and New Zealand contain NDVI composites only.

Composite Building Process

The first step in the eVIIRS composite building process is to create the various geotiff bands required for building an NDVI composite for each of the days and times in the compositing period. For example, the compositing period might be 7 or 10 days.

One of the GeoTIFF bands created is the NDVI band. NDVI is created from the known equation

where red is VIIRS band I1 and NIR is band I2. These bands come from the 375-m surface reflectance bands of VNP09 granules.

The minimum NDVI values allowed are -0.1999, and all NDVI values (which normally fall between -1.0 and 1.0) are scaled by 10,000. Thus, the output NDVI values from this application fall between -2000 and 10,000 (with the exception of negative surface reflectance values) to match the LP DAAC MOD09 products.

If either of the red or NIR pixel values is background fill, then the output NDVI value is set to UNDEF (-2000). The same value applies if the red and NIR pixels are the same value. If either of the red or NIR pixels is negative, then the output NDVI value is set to NEG_SR (-3000).

Other bands created for the compositing process are Satellite Zenith Angle and Surface Reflectance Quality Bands 1, 2 and 4.

Composite Calculation

The software used to create 7-, 14- and 10-day dekadal NDVI composites was developed by the USGS EROS software development team. The software processes VIIRS 375-m swath granules to produce 375-m and/or 1-km composite products.

The software grids the required swath granules (VNP09 and NPP_IMFTS_L1) for the specified bounded region (using upper left and lower right corner coordinates). For each gridded granule, an NDVI band is generated using the red (band I1) and NIR (band I2) bands.

Once each granule has been gridded to the given coordinate extents and the NDVI has been built, then an enhanced maximum value composite (MVC) is generated. With a straight MVC algorithm, each pixel in the composite (for each band) would be filled using the pixel value of the granule with the highest NDVI for the current pixel. The MVC has been enhanced in the current eVIIRS algorithm by incorporating band quality, cloud mask, negative surface reflectance, and view angle into the algorithm for eVIIRS processing.

  1. The band quality information in the surface reflectance product is used to determine if a certain pixel is of “bad quality.” Bad quality pixels are not used in the final composite product.
  2. The NDVI algorithm flags negative surface reflectance values so that they will not be used in the output composite, if positive values are available. If either the red or NIR band surface reflectance values are negative, then the output NDVI value is flagged with a value of -3000. (NOTE: Undefined or background fill NDVI values are -2000 and that is specified as the lower limit of the “valid” NDVI values.) The composite algorithm then disregards granules with NDVI values of -3000 to omit negative surface reflectance values in the output composite.
  3. The cloud mask is used to minimize the presence of cloud pixels in the composite product. If the quality flag is probably clear and/or confident clear, then the pixel may be used to fill the current composite pixel.
  4. The view angle (distance from nadir) determines which of the two highest NDVI pixels is used in the composite. Of the non-cloudy, ideal-quality pixels, the two highest NDVI values are determined from the list of granules. Of these two granules, the pixel value which is closest to nadir is used to fill the composite.

The products are generated for S-NPP VIIRS over all areas including CONUS and Alaska and are projected to a regionally specific mapping grid and delivered in a compressed (zipped) GeoTIFF format.

Two spatial resolutions (375 m and 1,000 m) are available to download individually. These data are available in EarthExplorer as well Machine-to-Machine API. The download product is a .zip file containing 6 files, including three .tif, one jpeg, and two .txt files.

Following is a sample list of files from a CONUS 1-km NDVI composite .zip file named “”:

US_eVSH_NDVI.2020.350-356.1KM.VI_ACQI.006.2020362063105.tif Acquisition geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_ACQT.006.2020362063105.txt Acquisition text file
US_eVSH_NDVI.2020.350-356.1KM.VI_META.006.2020362063314.met Metadata text file
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.jpg Browse JPEG
US_eVSH_NDVI.2020.350-356.1KM.VI_NDVI.006.2020362063105.tif NDVI geotiff
US_eVSH_NDVI.2020.350-356.1KM.VI_QUAL.006.2020362063105.tif Quality geotiff

Acquisition Band (ACQI)

An acquisition band provides the user with a GeoTIFF image that can be overlayed onto the NDVI image so that the user can then identify each granule that provided the value for each NDVI pixel. The DOY and the acquisition number will be used to specify the acquisition. The acquisition number in this case represents not the time of the acquisition, but the order of capture. The output values are represented as an unsigned 16-bit integer using the equation

Thus, the first acquisition for DOY 117 would be 117001. The eleventh acquisition for DOY 117 would be 117011. To make it easier for the user to determine which acquisition integer value maps to which granule, an output acquisition table text file is written to match the acquisition values to the granule names.

Acquisition Text File (ACQT)

An acquisition text file provides the user a text file to accompany the Acquisition image, which gives the composite details of the data used to create the product. This list of the acquisition band values is used in the composite acquisition band product and the corresponding acquisition filename for that acquisition value.

160005 MA2RG_2020_160_2120_250m_NDVI.hdf
160004 MA2RG_2020_160_1940_250m_NDVI.hdf
160003 MA2RG_2020_160_1935_250m_NDVI.hdf
160002 MA2RG_2020_160_1800_250m_NDVI.hdf
160001 MA2RG_2020_160_1755_250m_NDVI.hdf
159007 MA2RG_2020_159_2035_250m_NDVI.hdf
159006 MA2RG_2020_159_2030_250m_NDVI.hdf
159005 MA2RG_2020_159_1900_250m_NDVI.hdf

Metadata (META)

The Metadata file provides the details about the composite image, including acquisition time period, publication dates, pixel and row counts, map projection information, datum, pixel resolution, satellite and platform, digital data type, fill values, scaling factors, and center and corner coordinates.

Browse JPEG

A full resolution browse image is provided in JPEG format and is zipped with the product and used to display the image on EarthExplorer. The color mapping is used to create a color image for the browse image and is not the same as the 16-bit single band data delivered in the NDVI GeoTIFF.

NDVI / Surface Reflectance GeoTIFF

The NDVI GeoTIFF is the product created in the compositing process detailed in the above section. All the other files including quality band, acquisition band, as well as the text files and browse are companion files for this NDVI product.

The Surface Reflectance composite is also created for CONUS, Alaska, and Central America/Caribbean and Mexico and is available as a separate download option in EarthExplorer.

Quality Band (QA)

A quality assurance (QA) band is 8 bits and produced by the composite software to identify the quality of each composite pixel. Most pixels will be filled with a pixel of good band quality. However, if none of the granules for a particular pixel are of good band quality or all of the pixels are cloudy, then the current pixel is filled with data from the best pixel possible where the preference order is valid NDVI over fill, good quality over bad, snow pixel over cloudy pixel. The following values are used in theQA band:

0 = good quality
1 = cloudy pixel
2 = bad band quality
4 = snow
10 = fill

Surface Reflectance Quality Band 1 is made up of the following bit values. Bit number 0 is the bit in the binary number which is of the lowest numerical value or Least Significant Bit (LSB).

Surface Reflectance Quality Band 1
Bit No. Parameter Name Bit Combination Definition
0-1 Cloud Mask Quality 00 Poor
01 Low
10 Medium
11 High
2-3 Cloud Detection and Confidence 00 Confident Clear
01 Probably Clear
10 Probably Cloudy
11 Confident Cloudy
4 Day/Night 0 Day
1 Night
5 Low Sun Mask 0 High
1 Low
6-7 Sun Glint 00 None
01 Geometry Based
10 Wind Speed Based
11 Geometry and Wind Speed Based

Surface Reflectance Quality Band 2 is made up of the following bit values.

Surface Reflectance Quality Band 2
Bit No. Parameter Name Bit Combination Definition
0-2 Land/Water Background 000 Land and Desert
001 Land No Desert
010 Inland Water
011 Sea Water
100 ---
101 Coastal
110 ---
3 Shadow Mask 0 No Cloud Shadow
1 Shadow

Surface Reflectance Quality Band 4 is made up of the following bit values. Bits 1 and 2 from quality band 4 are used to determine if the bits from a given gridded I1 (red) or I2 (near-infrared) band should be used in creating the NDVI band. Surface Directional Reflectance (SDR), Aerosol Optical Thickness (AOT), Ante Meridiem (AM), Precipitable Water (PW)

Surface Reflectance Quality Band 4
Bit No. Parameter Name Bit Combination Definition
0 BAD M11 SDR Data 0 No
1 Yes
1 Bad I1 SDR Data 0 No
1 Yes
2 Bad I2 SDR Data 0 No
1 Yes
3 Bad I3 SDR Data 0 No
1 Yes
4 Overall Quality of AOT 0 Good
1 Bad
5 Missing AOT Input Data 0 No
1 Yes
6 Invalid Land AM Input Data 0 Valid
1 Invalid AM Input Over Land or Over Ocean
7 Missing PW Input Data 0 No
1 Yes

Coverage Maps

Coverage Maps indicating the availability of eVIIRS NDVI products are available for download.

Additional Information

Access Data

eVIIRS NDVI products held in the USGS archive can be searched using EarthExplorer. On EarthExplorer, eVIIRS NDVI products can be found under the Vegetation Monitoring category.

Doing NDVI when nir data is given as 3 bands - Geographic Information Systems

Calculation of the Normalized Difference Vegetation Index (NDVI), which is available on-the-fly, comes first. In addition, NDVI is often used around the world to monitor drought, forecast agricultural production, assist in forecasting fire zones and desert offensive maps. Farming apps, like Crop Monitoring, integrate NDVI to facilitate crop scouting and give precision to fertilizer application and irrigation, among other field treatment activities, at specific growth stages. NDVI is preferable for global vegetation monitoring since it helps to compensate for changes in lighting conditions, surface slope, exposure, and other external factors.

NDVI is calculated in accordance with the formula:

NIR – reflection in the near-infrared spectrum
RED – reflection in the red range of the spectrum

According to this formula, the density of vegetation (NDVI) at a certain point of the image is equal to the difference in the intensities of reflected light in the red and infrared range divided by the sum of these intensities.

This index defines values ​​from -1.0 to 1.0, basically representing greens, where negative values ​​are mainly formed from clouds, water and snow, and values ​​close to zero are primarily formed from rocks and bare soil. Very small values ​​(0.1 or less) of the NDVI function correspond to empty areas of rocks, sand or snow. Moderate values ​​(from 0.2 to 0.3) represent shrubs and meadows, while large values ​​(from 0.6 to 0.8) indicate temperate and tropical forests. Crop Monitoring successfully utilizes this scale to show farmers which parts of their fields have dense, moderate, or sparse vegetation at any given moment.

Put simply, NDVI is a measure of the state of plant health based on how the plant reflects light at certain frequencies (some waves are absorbed and others are reflected).

Chlorophyll (a health indicator) strongly absorbs visible light, and the cellular structure of the leaves strongly reflect near-infrared light. When the plant becomes dehydrated, sick, afflicted with disease, etc., the spongy layer deteriorates, and the plant absorbs more of the near-infrared light, rather than reflecting it. Thus, observing how NIR changes compared to red light provides an accurate indication of the presence of chlorophyll, which correlates with plant health.

Crop Monitoring is a perfect tool for tracking the health of the crops in the field with the help of the NDVI measured on-the-fly. All you need to do is add your fields to the system, customize the NDVI settings and start receiving the notifications.

Fields analytics tool with access to high-resolution satellite images for remote problem areas identification!

Crop Monitoring tracks changes in the NDVI for individual fields throughout the season. This enables you to refer to the historical field’s productivity for up to 5 past years. You can monitor both the crop rotation patterns and the current vegetation rates. With the help of the user-friendly charts, the app visualizes different types of data, including the vegetation indices, temperature, precipitation rate, growth stages, historical weather, among others. Another important feature, based on calculating the NDVI rates, zoning allows you to identify high productivity zones, as well as discover the weak points in the field that require special treatment. Each zone, at every growth stage, needs a different amount of fertilizer and irrigation treatment (the latter is also decided based on the precipitation rates), both of which can be manually adjusted in the app to a great degree of accuracy. Precision agriculture, based on NDVI, doesn’t end there, however! Crop Monitoring updates scouting by using NDVI to find problem areas in the field and sending scouts directly to the exact location, thus saving time and resources. Users also get notified every time an abnormal change of the NDVI value has been detected, allowing farmers, traders, and insurers to make well-informed agricultural decisions in a timely manner.

1 Introduction

Water extends approximately 71% of earth's surface and it is also imperative for the existence and sustainability of living organism on the earth surface 1 . The freshwater is just 2.5% of the earth's water. About 0.3% of freshwater is found in rivers, lakes, and atmosphere 2 . In general, the understanding of the water quality plays a critical role prior to utilize for various purposes including drinking 3 . In this paper, we opted to understand the surface water quality for the Bow River, which is a major river in the Canadian province of Alberta having a total length of 587 km, and a main source of drinking water for many communities of the province 4 .

The surface water quality of the Bow River is measured every month at three fixed sampling sites (i.e. Carseland, Cluny, and Ronalane) for different water quality variables using the traditional methods. In general, these methods provide accurate measurement. However, these may not be feasible means to sample the entire river due to the huge involvement of labor and cost. Currently, the measured data of water quality variables at the sampling sites of the Bow River are grouped into five classes (i.e. excellent, good, fair, marginal and poor) using the framework of Canadian Water Quality Index (CWQI: see details in Section 2.5) 5 . These classes are obtained on the basis of fixed-point locations, which do not represent the spatial dynamics of the entire river.

In another study, we classified the surface water quality of major rivers of Alberta on the basis of clusters. We observed higher (deteriorated water quality) clusters (i.e. 4 and 5) for the rivers during the growing season (April 1–September 30) as compared to lower clusters (i.e. 1, 2, and 3) in winter months (Oct 1–March 31). During the growing season, the snowmelt wash various materials from the land surface into the rivers due to anthropogenic activities related to different types of land use/cover. Turbidity was found to be a dominant parameter associated with the deterioration in water quality during the growing season 6 . On this basis, we considered turbidity separately besides CWQI in this study. For the Bow River, the turbidity is measured at fixed sampling location, which does not represent the mean turbidity for the whole water body 7 .

In order to address the spatial variability in water quality real time data, remote sensing-based methods were found to be alternative and efficient ones 8-10 . The remote sensing methods are suitable to analyze: (i) spatial variability over a large geographic area, (ii) temporal trends over certain periods of interests, and (iii) the conditions of the water bodies in remote areas. In remote sensing, optical remote sensors are used for monitoring the water quality-related variables. The most commonly used sensors include the use of Landsat-7 ETM 11, 12 , Landsat-5 TM 13, 14 , MODIS 15 , NOAA AVHRR 16 , and SPOT HVR 17 among others. In most of the instances, the spectral bands used in these studies included blue (B), green (G), red (R), and near infrared (NIR) 11-17 . The observed planetary reflectance from these bands was used to study water quality variables including suspended sediment, turbidity, Secchi disk depth, and chlorophyll-a 12, 13, 18, 19 .

In another study, we classified and analyzed the surface water quality for twelve major rivers of Alberta. We developed a surface water quality classification system using principal component analysis, total exceedance model and clustering technique. From principal component analysis, we identified seven major principal components, which were the indicators of watershed geology, mineralization, and anthropogenic activities related to land use/cover. The principal components were used to identify the dominant parameters. The normalized data of dominant parameters were used to develop a total exceedance model. The exceedance values were used to determine the patterns for the development of five clusters. The water quality deteriorates as the cluster number increased from cluster 1 to cluster 5. The clusters showed reasonably strong agreements (i.e. 80–90%) against the classes of CWQI. The dominant clusters during the growing and winter seasons were used for the spatial and temporal patterns of the surface water quality of rivers 6 .

In the present study, we have tested remote sensing-based methods for acquiring CWQI and turbidity classes for assessing both spatial and temporal dynamics of the Bow River. The specific objectives of this paper are to: (i) develop and evaluate remote sensing based models to acquire CWQI classes using the planetary reflectance of Landsat-5 TM and ground measured data, (ii) develop and evaluate remote sensing based models to retrieve turbidity using the planetary reflectance of Landsat-5 TM and in situ data, (iii) apply the selected models to classify the source waters of the Bow River into CWQI and turbidity classes for spatial and temporal analysis, and (iv) study the impact of natural sub-regions on Bow River water quality.

Landsat Enhanced Vegetation Index

This image displays a (left) Landsat 8 Surface Reflectance (SR) and (right) the SR-derived Enhanced Vegetation Index (EVI).

Landsat Surface Reflectance-derived Enhanced Vegetation Index (EVI) are available for Landsat 4–5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), and Landsat 8 Operational Land Imager (OLI)/Thermal Infrared Sensor (TIRS) scenes that can be successfully processed to Landsat Level-2 Surface Reflectance products.

EVI is similar to Normalized Difference Vegetation Index (NDVI) and can be used to quantify vegetation greenness. However, EVI corrects for some atmospheric conditions and canopy background noise and is more sensitive in areas with dense vegetation. It incorporates an “L” value to adjust for canopy background, “C” values as coefficients for atmospheric resistance, and values from the blue band (B). These enhancements allow for index calculation as a ratio between the R and NIR values, while reducing the background noise, atmospheric noise, and saturation in most cases.

EVI = G * ((NIR - R) / (NIR + C1 * R – C2 * B + L))

In Landsat 4-7, EVI = 2.5 * ((Band 4 – Band 3) / (Band 4 + 6 * Band 3 – 7.5 * Band 1 + 1)).

In Landsat 8, EVI = 2.5 * ((Band 5 – Band 4) / (Band 5 + 6 * Band 4 – 7.5 * Band 2 + 1)).

EVI is delivered as a single band product, specified as shown in the table below.

Evaluating the performance of vegetation indices for detecting oil pollution effects on vegetation using hyperspectral (Hyperion EO-1) and multispectral (Sentinel-2A) data in the Niger Delta

Nkeiruka N. Onyia , . Juan Carlos Berrío , in Hyperspectral Remote Sensing , 2020

19.2.8 Vegetation indices

VIs to detect canopy chlorophyll content and stress pigments were derived from the three datasets, namely HS, MS, and FS. The indices calculated include:

Normalized difference vegetation index (NDVI), which is a broadband greenness index computed from the Sentinel-2A image.

Red-edge normalized difference vegetation index (RENDVI), which is a narrowband equivalent of the NDVI. This index was computed using the Hyperion and FS datasets.

Red-edge position index (REP), which was manually computed as the maximum first derivative of reflectance at Hyperion wavelengths between 671 and 782 nm (Clark et al., 2010). It is a chlorophyll related index with reduced sensitivity to variations in leaf/canopy chlorophyll content as well as environmental conditions ( Gholizadeh et al., 2016 ).

Anthocyanin reflectance index 2 (ARI2), which was used to estimate the concentration of anthocyanins in the leaf canopy. This index is often used to detect plant stress.

Structural insensitive pigment index (SIPI), which maximizes sensitivity to carotenoids while minimizing sensitivity to variation in canopy structure ( Peñuelas et al., 1993 ). The index is commonly used for plant physiological stress detection.

ARI2 and SIPI are used to detect stressed and unhealthy vegetation due to their ability to respond to changes in plant physiological status. Past studies have shown that the reflectance at the absorption maxima of these pigments decreases in stressed vegetation ( Merzylak et al., 2008 ). These indices were automatically computed in ENVI 5.4 except the REP, which was calculated in Excel for HS and FS, and in SNAP for MS. The VIs, formulae, and references used in this study are shown in Table 19–3 .

Table 19–3 . Vegetation indices analyzed in this study with the formulae and references.

Normalized difference vegetation index n − r n + r Pearson and Miller (1972)
Red-edge normalized difference vegetation index R 750 − R 705 R 705 + R 750 Gitelson and Merzlyak (1996)
Red-edge position index ( 705 + 35 ) [ ( ( R 783 − R 665 ) / 2 ) − R 705 R 740 − R 705 ] Gholizadeh et al. (2016)
REP for hyperspectral and fused images Eqs. (19.2) and (19.3) Savitzky (1964)
Anthocyanin reflectance index 2 R 800 ( 1 R 550 − 1 R 700 ) Gitelson et al. (2001)
Structure insensitive pigment index R 800 − R 445 R 800 − R 680 Peñuelas et al. (1995)

The computation of REP from the wavelengths of the HS and FS images was manually performed in Excel using the formula proposed by Savitzky (1964) as shown here:

where Ri is the reflectance at Yi Rj is the reflectance at Yj Yi is the wavelength at the start of the slope segment Yj is the wavelength at the end of the slope segment.