MODIS/Terra Snow Cover L2 and L3 Daily and 8-Day 500 m


Snow cover products from the Moderate Resolution Imaging Spectroradiometer (MODIS) include level 2 swath data (MOD10_L2) at 500 m resolution, level 3 gridded daily and 8-day composites (MOD10A1 and MOD10A2, respectively) at 500 m resolution, and level 3 8-day global maps in a climate modeling grid (CMG, MOD10C2) at 0.05 degree resolution. MODIS snow cover data are based on a snow mapping algorithm that employs a Normalized Difference Snow Index (NDSI) and other criteria tests. Snow cover data and supporting attributes are available in HDF-EOS format via ftp, 8-mm tape, CD-ROM, DVD, or DLT tape. MODIS snow products are considered provisional at this time. Product quality may not be optimal, and incremental product improvements are still occurring.


Citations for use of these data should read:

Hall, D.K., G.A. Riggs, and V.V. Salomonson. 2001. MODIS/Terra Snow Cover L2 and L3 Daily and 8-Day 500 m. Boulder, CO, USA: National Snow and Ice Data Center. Digital media.

Table of Contents

1. Collection Overview

Collection Contents

This document corresponds to four separate data sets:

"V00x" at the end of each title represents the version number. V003, the most current version of MODIS processing, incorporates algorithm refinements and instrument and calibration stabilization. MOD10C2 consists only of V003 data. The MOD10A1, MOD10A2, and MOD10_L2 V003 collections will eventually incorporate all reprocessed V001 data. V001 data for a given date will be available for six months after they have been reprocessed.

MODIS snow products are archived in Hierarchical Data Format - Earth Observing System (HDF-EOS) format files. HDF was developed by the National Center for Supercomputing Applications (NCSA) and is the standard archive format for EOS Data and Information System (EOSDIS) products. HDF-EOS employs point, swath, and grid structures to geolocate parameters to geographic coordinates.

MODIS/Terra Snow Cover 5-Min L2 Swath 500m (MOD10_L2) contains the following fields:

Latitude and longitude geolocation fields are at 5 km resolution, while the snow cover field is at 500 m resolution. MOD10_L2 data are produced in five minute segments, which corresponds to approximately 203 scans. At 500 m resolution there are 20 lines per scan, yielding 4060 pixels in the along track direction and 2708 pixels in the cross track direction. At the earth's surface, the coverage of a single MOD10_L2 record is approximately 2030 km along track by 2330 km cross track.

MODIS/Terra Snow Cover Daily L3 Global 500m ISIN Grid (MOD10A1) consists of 2400 pixel by 2400 pixel tiles of data gridded in an integerized sinusoidal (ISIN) map projection. This product contains the following fields:

MODIS/Terra Snow Cover 8-Day L3 Global 500m ISIN Grid (MOD10A2) consists of 2400 pixel by 2400 pixel tiles of data gridded in an integerized sinusoidal (ISIN) map projection. This product contains the following fields:

MODIS/Terra Snow Cover 8-Day L3 Global 0.05Deg CMG (MOD10C2) consists of 3600 pixel by 7200 pixel global arrays of snow cover. MOD10A2 500 m files for each 8-day period are binned to 0.05° to create a CMG product that contains the following fields:

Related Data Collections

MOD10_L2 users requiring higher resolution geolocation data than what is actually stored in each MOD10_L2 granule can use the "MODIS/Terra Geolocation Fields Daily L1A Swath 1km" (MOD03) product. This collection contains geodetic coordinates, ground elevation, solar and satellite zenith, azimuth angle using the spacecraft attitude and orbit, instrument telemetry, and a digital elevation model, all at 1 km resolution.

Title of Investigation

MODIS/Terra Snow Cover L2 and L3 Daily and 8-Day 500 m

Investigator(s) Name and Title

Dorothy K. Hall
NASA Goddard Space Flight Center
Mailstop 400.9
Greenbelt, MD 20771

George A. Riggs
NASA Goddard Space Flight Center
Mailstop 974.0
Greenbelt, MD 20771

Vince V. Salomonson
NASA Goddard Space Flight Center
Mailstop 900.0
Greenbelt, MD 20771

Technical Contact

NSIDC User Services
National Snow and Ice Data Center
University of Colorado
Boulder, CO 80309
Phone: +1 303-492-6199
Fax: +1 303-492-2468

2. Applications and Derivation


Snow cover is an important factor in the earth's global water balance. MODIS global snow cover data are primarily used for hydrological studies such as short-term and seasonal river flow forecasts and snowmelt runoff models. Snow cover data can be combined with climatological temperature data for a given region, to determine areal snowmelt. Analysis of satellite snow cover data can reduce forecasting errors for those years which deviate from a normal runoff pattern, particularly during the snowmelt season. Snow cover data can also be used to improve discharge forecasts to ensure an efficient allocation of water in semi-arid climates, improving crop production (Hall and Martinec 1985).

Theory of Measurements

Satellites are well suited to the measurement of snow cover because the high albedo of snow presents a good contrast with most other natural surfaces except clouds. Spectral reflectivity of snow depends on grain size and shape, impurity content, liquid water content, depth, surface roughness, and solar elevation angle (Hall and Martinec 1985). Reflectance of fresh snow is very high in the visible wavelengths, but decreases in the near-infrared wavelengths, especially as grain size increases. Because of natural aging and other factors such as soot or volcanic ash deposition, reflectance of snow decreases over time. Fresh snow can have a reflectance up to about 80%, but its reflectance may decrease to below 40% after snow crystals metamorphose (Hall et al. 1998).

Snow and Cloud Discrimination
Snow and cloud discrimination techniques are based on differences between cloud and snow/ice reflectance and emittance properties (Figure 1). Clouds typically have high reflectance in visible and near-infrared wavelengths, while reflectance of snow decreases in shortwave infrared wavelengths. The Normalized Difference Snow Index (NDSI) can generally separate snow from most obscuring cumulus clouds, but it cannot always discriminate optically-thin cirrus clouds from snow (Hall et al. 1998).

Special Considerations for Dense Forests
The mapping of snow cover becomes limited in areas where snow cover is obscured by dense forest canopies. A forested landscape is never completely snow-covered because tree branches, trunks, and canopies may not be covered with snow. Often, in boreal forests, snow that falls on the coniferous tree canopy will not stay on the canopy for the entire winter because of sublimation. Thus, even in a continuously snow-covered area, much of the forested landscape will not be snow-covered. Furthermore, snow that falls onto the ground through the canopy may not be visible from above.

A canopy reflectance model (GeoSAIL) for discontinuous canopies is used to calculate the fraction of sunlit crown, sunlit background, crown reflectance, canopy transmittance, shadowed crown, and shadowed background within a forest stand. Reflectances for the sunlit snow are calculated using the Wiscombe and Warren 1980 model, while reflectances from other components were measured from direct observations. A significant difference exists between the high reflectance of snow and the low reflectance of soil, leaves, and bark. Depending on the vegetation type, snow may also cause a decrease in the mid-infrared reflectance of the forest stand. In addition, reflectance in the visible spectrum often increases with respect to the near-infrared reflectance. This lowers the Normalized Difference Vegetation Index (NDVI), a complement to the NDSI (Figure 2).

The NDSI and NDVI are used together to discriminate between snow-free and snow- covered forests (Figure 3). Forested pixels have higher NDVI values compared with non-forested pixels. Thus, by using the NDSI and NDVI in combination, it is possible to lower the NDSI threshold in forested areas without compromising the algorithm performance in other land covers (Hall et al. 1998).

Derivation Techniques and Algorithms


MODIS level 1B Calibrated Earth View data (MOD02HKM) are used as input to calculate reflectance (MCST 2000), using the following equation:

pcos(theta) = reflectance_scalesB(SIB - reflectance_offsetsB)

The reflectance_scales and reflectance_offsets are computed from calibration parameters such that the reflectance product (p) is determined directly from the integer representation (SI) of the digital number. The subscript (B) refers to band. The result of this equation is then multiplied by 1/cos(theta) to give at-satellite reflectance (p):

p = pcos(theta) * 1/cos(theta)

Processing Steps

Data products were generated by the MODIS Science Investigator-led Processing System (SIPS), and transferred to NSIDC.

The snow mapping algorithm is used to generate global snow cover products from MODIS data. It is designed to identify the presence of snow by reflectance or emittance properties in each 500 m pixel for each orbit. The primary techniques for this algorithm are threshold-based grouped criteria tests, normalized difference between infrared and visible bands, and decision rules. A global daily product was created along with an 8-day composited product. If a pixel was snow-covered on any orbit during the 8-day period, that pixel was mapped as snow-covered, even if it were snow-free on all of the other orbits during the 8-day period.

The snow mapping algorithm is based on the Normalized Difference Snow Index (NDSI), which is useful for the identification of snow and ice, and for distinguishing between snow and ice, and cumulus clouds. The NDSI is a measure of the reflectance difference between the visible and shortwave infrared reflectance of snow. The NDSI is adaptable for a number of illumination conditions, does not depend on reflectance for a specific band, and is partially normalized for atmospheric effects. The algorithm uses MODIS bands 4 (0.55 µm) and 6 (1.6 µm) to calculate the NDSI (Hall et al. 1998).

Specific processing steps vary according to product:

MODIS/Terra Snow Cover 5-Min L2 Swath 500m (MOD10_L2)
Analysis for snow in a MODIS swath is constrained to pixels that:

  • have nominal level 1B radiance data
  • are in daylight
  • are over land or inland water
  • are unobstructed by cloud

Constraints are applied in the order listed. After they are applied, only pixels having a daylight clear sky view of the land surface are analyzed for snow. Clouds are masked with the MODIS Cloud Mask data product (MOD35_L2). Initially, only the unobstructed field-of-view flag from MOD35_L2 is used to mask clouds. MOD35_L2 contains a great amount of data on the results of the processing paths and cloud tests applied within the MODIS cloud-clearing algorithm. Masking of oceans and lakes is conducted within the 1 km resolution land-water mask, contained in the MODIS geolocation product (MOD03).

Snow detection is achieved through the use of two groups of criteria tests for snow reflectance characteristics in the visible and near-infrared regions. One group is for detection of snow under different conditions. A pixel is identified as snow if all the following conditions are met:

  • (Band 4 - Band 6) / (Band 4 + Band 6) is greater than 0.4, AND
  • Band 2 reflectance is greater than 0.10, AND
  • Band 4 reflectance is greater than 0.11

Another group of criteria tests is used to better detect snow in forests. In this case, a pixel is identified as snow if the following conditions are met:

  • The pixel has NDSI and NDVI values in a defined polygon in a scatter plot of the two indices, AND
  • Band 1 reflectance is greater than 0.10, AND
  • Band 2 reflectance is greater than 0.11

This last set of criteria is applied without regard to the ecosystem, so it should not be interpreted strictly as snow-covered forest. Snow-covered ice on inland water is determined by applying the first group of criteria tests used for snow detection to pixels mapped as inland water by the land-water mask.

Intermediate checks for theoretical bounding of reflectance data and the NDSI ratio are made in the algorithm. In theory, reflectance values should lie within the 0-100% range, and the NDSI ratio should lie within the -1.0 to +1.0 range. Summary statistics are kept for pixels that exceed these theoretical limits; however, the test for snow is done regardless of violations of these limits. Violations are tracked and written as local attributes, as they suggest that error or other anomalies may have been introduced into the input data, indicating the need for further investigation. The snow algorithm also identifies missing data and reports them in the output product. Certain expected anomalous conditions may exist with the input data, such as a few missing lines or unusable data from the MODIS sensor. In these cases, the snow algorithm makes no snow decision for an affected pixel. A snow spatial QA bit is set to indicate the cause, and the algorithm moves to the next pixel. Summary statistics are calculated for these conditions, and reported as local attributes (Riggs et al. 2001).

MODIS/Terra Snow Cover Daily L3 Global 500m ISIN Grid (MOD10A1)
A daily snow cover map is constructed by examining the multiple observations acquired for a day that are mapped to each grid cell. A selection of best observations is made based on a ratio of percentage coverage of an observation in a cell to the distance of that observation from nadir. The observation that is closest to nadir with the greatest areal coverage in the cell is selected as the observation of the day. The rationale was to select the observation least affected by off nadir viewing, while maximizing coverage within a cell of the gridded projection. The relative view azimuth of the observation is also determined, and if it is outside a set limit, a bit in the QA field is set to indicate that. Results of the snow cover algorithm are used to create a daily snow map of the region, and are stored in the Day_Tile_Snow_Cover field. The QA data for that snow cover map are stored in the Snow_Spatial_QA field (Riggs et al. 2001). Gridded snow cover data are stored as coded integer values, with values being the same as assigned for MOD10_L2.

MODIS/Terra Snow Cover 8-Day L3 Global 500m ISIN Grid (MOD10A2)
The algorithm first checks that the input data from MOD10A1 files are from the same 8-day period, and then orders them chronologically. Multiple days of observations for a cell are examined. If snow cover is found for any day, then the cell in the Maximum_Snow_Extent field is labeled as "snow." If no snow is found, but there is one value that occurs more than once, that value is placed in the cell (for example, if a pixel is classified as "water" for five days, "cloud" for one day, "land" for one day, and "night" for one day, it would be ultimately labeled as "water"). If mixed observations occur, for example, land and cloud for more than one day in a given pixel, the algorithm assumes a cloud-free period and labels a pixel with the observed value. This type of logic minimizes cloud-cover extent, such that a cell needs to be cloud-obscured for all days in order to be labeled "cloud." If all observations for a cell are analyzed but a classification cannot be determined, then that cell is labeled as "no decision." A chronology of snow occurrence is recorded in the Eight_Day_Snow_Cover field. The eight bits within a byte correspond to eight days of data. If snow is found in a bit for a given day, in reading the eight bits (one byte) from right to left, that bit is set to a value of "one" in the Eight_Day_Snow_Cover field (Riggs et al. 2001).

MODIS/Terra Snow Cover 8-Day L3 Global 0.05Deg CMG (MOD10C2)
The algorithm maps MOD10A2 8-day snow cover data at 500 m resolution into the corresponding cell of a 0.05° climate modeling grid (CMG) and computes snow and cloud percentages, QA, and a confidence index based on the mapping results. The MODIS Instrument Science Team employed a University of Maryland 1 km global land cover mask to compute the percentage of land in each CMG cell and determine an appropriate classification. CMG cells with at least 12% land were classified as "land." The land extent map was also used in the calculation of the confidence index.

Because Antarctica's surface is typically less than 1% snow-free (a value less than the global error rate for MODIS snow mapping), the algorithm classifies Antarctica as completely snow-covered. This also reduces confusion with cloud signatures.

Calculated Variables

Output of the snow mapping algorithm includes snow, snow-covered lake ice (for level 3 products only), cloud, water, land, or other, based on reflectance criteria as described in the Processing Steps. Percentage of some classifications are calculated and reported in the local attributes for respective data products.

3. Data Description and Access


MODIS snow products are archived in HDF-EOS format. HDF, developed by the National Center for Supercomputing Applications (NCSA), is the standard archive format for EOSDIS products. HDF-EOS employs point, swath, and grid structures to geolocate parameters to geographic coordinates. Various software packages, including public domain, support the HDF-EOS data format. See the Software section for more information.

HDF-EOS fields by data product:

MODIS/Terra Snow Cover 5-Min L2 Swath 500m (MOD10_L2)

Results of the snow mapping algorithm are stored as coded integers in the snow cover field. The algorithm classifies pixels as snow-covered lake ice, cloud, water, or other. Latitude and longitude fields contain 5 km resolution geographic coordinates for georeferencing snow data. The latitude and longitude data correspond to the lower left of the four center pixels of a 5 km by 5 km (i.e., 10 x 10) block of 500 meter pixels from the snow cover field.

MODIS/Terra Snow Cover Daily L3 Global 500m ISIN Grid (MOD10A1)

MODIS/Terra Snow Cover 8-day L3 Global 500m ISIN Grid (MOD10A2)

A summary of MOD10A2 bit values provides an interpretation of bit values and resulting integer values for the 8-day snow cover field.

MODIS/Terra Snow Cover 8-Day L3 Global 0.05Deg CMG (MOD10C2)

The 8-day CMG Snow Cover field is a global map of snow cover extent (expressed as a percentage of land in each CMG cell) for an 8-day period. The 8-day CMG Confidence Index field indicates how much of the land surface MODIS observes. The greater the percentage of land, the higher the confidence for snow extent. Cloud obstruction reduces the confidence index. The 8-day CMG Cloud Obscured field indicates how much of the land surface in the cell is persistently obscured during the 8-day period. The Snow Spatial QA field indicates the overall quality of data in the CMG cell. QA data are stored as a 2-bit flag.


A separate ASCII text file containing metadata (with a .met file extension) accompanies the HDF-EOS file. The metadata file contains some of the same metadata as in the product file, but also includes other information regarding archiving, user support, and post production quality assurance (QA) relative to the granule ordered. The post-production QA metadata may or may not be present depending on whether or not the data granule has been investigated for quality assurance. The metadata file should be examined to determine if post-production QA has been applied to the granule (Riggs 2001).

File Naming Convention:

MODIS/Terra Snow Cover 5-Min L2 Swath 500m (MOD10_L2):

Example:  "MOD10_L2.A2000055.2135.002.2000056231924.hdf"

A = AM-1 (Terra) platform
2000 = Year of data acquisition
055 = Julian date of data acquisition (day 55)
2135 = Hour and minute of data acquisition in Greenwich Mean Time (GMT) (21:35)
002 = Collection number *
2000 = Year of production (2000)
056 = Julian date of production (day 56)
231924 = Hour/minute/second of production in GMT (23:19:24)

*(File names with "002" indicate Version 1 data)

MODIS/Terra Snow Cover Daily L3 Global 500m ISIN Grid (MOD10A1)   and
MODIS/Terra Snow Cover 8-day L3 Global 500m ISIN Grid (MOD10A2):

Example:  "MOD10A1.A2000057.h14v03.002.2000057231924.hdf"

The convention is the same as MOD10_L2, except that in place of the hour and minute of data acquisition, the following are expressed:

h14 = horizontal tile number (14)
v03 = vertical tile number (3)

MODIS/Terra Snow Cover 8-Day L3 Global 0.05Deg CMG (MOD10C2):

Example:  "MOD10C2.A2001033.003.2001268165647.hdf"

A = AM-1 (Terra) platform
2001 = Year of data acquisition
033 = Julian date of data acquisition (day 33)
003 = Version of data type (Version 3)
2001 = Year of production (2001)
268 = Julian date of production (day 268)
165647 = Hour/minute/second of production in GMT (16:56:47)

Data Organization

Data Granularity

  • MODIS/Terra Snow Cover 5-Min L2 Swath 500m (MOD10_L2): 22.9 MB per 5-minute swath
  • MODIS/Terra Snow Cover Daily L3 Global 500m ISIN Grid (MOD10A1): 11.6 MB per tile
  • MODIS/Terra Snow Cover 8-day L3 Global 500m ISIN Grid (MOD10A2): 11.6 MB per tile
  • MODIS/Terra Snow Cover 8-Day L3 Global 0.05Deg CMG (MOD10C2): 101.36 MB per global map

Data Access

Users can order data through the web-based WIST Data Gateway via ftp, 8-mm tape, CD-ROM, DVD, or DLT tape. Because of system performance considerations, users are limited to approximately five gigabytes of data for ftp requests. A brief set of ordering instructions is provided here.

Ordering MODIS snow products using the WIST Data Gateway:

  1. Enter a valid user name and password (create an account if you do not have an account), or enter as Guest.
  2. Highlight 'DATA CENTER' in the scroll window and click SELECT ->.
  3. Select 'NSIDC-ECS' as the data center. Click OK.
  4. Highlight 'DATA SET' in the scroll window and click SELECT ->.
  5. Select one or more of the following data sets from the Data Set List and click on the OK button near the bottom of the page.


  6. Choose a geographic study area and time period, then click on the Start Search button near the bottom of the page.
  7. If you queried more than one data set, the Results page displays the individual data sets from the search. Select the data set(s) and click on the List Data Granules button. A page with individual granules is then displayed. Respective data set names, geographic center points, and time periods are listed for each granule. Select the preferred data granules, then click on the Add to Cart button near the top of the page.
  8. Select the Order Options button. You may choose the same option for all granules of a data set, or you may specify individual ordering options for each data granule. Select Native Granule - FtpPull.

    "Production History" is a record of each step in the creation of a particular product. "Ancillary Data" are data required to perform an instrument's data processing, including orbit data, attitude data, time information, spacecraft engineering data, calibration data, data quality information, and data from other instruments. "Native Granule" is the data product in its raw, or native, format. "Production History" and "Ancillary Data" are not available for the MODIS snow products.

  9. Click on the OK Accept my choice and return to the shopping cart button near the bottom of the page.
  10. Click on the Go to Step 2: Order Form button.
  11. Fill out user contact information. Required fields are in red. Note that if you have an account, the fields will automatically be populated with your contact information. Click on the Go to step 3: Review Order Summary button.
  12. After reviewing the order summary for accuracy, click the Go to Step 4: Submit Order button.
  13. If you specified FtpPull in step 8 above, you will be notified by E-mail when your data are staged. The E-mail message will also contain the ftp address and directory information you will need to transfer the data to your computer.

Data Archive Center

Contact for Data Center Access Information

NSIDC User Services
National Snow and Ice Data Center
CIRES, Campus Box 449
University of Colorado
Boulder, CO 80309
Phone: +1 (303)-492-6199
Fax: +1 (303)-492-2468


Following is a list of software packages that support the HDF-EOS format:

Public Domain:


A more comprehensive list of software to read data in HDF-EOS format is available from NCSA.

4. Data Characteristics

Study Area

Coverage is global; however, only land tiles are produced for MOD10A1 and MOD10A2 products.

Spatial Resolution

Resolution at nadir is approximately 500 m for the data fields and 5 km for the latitude and longitude geolocation fields in MOD10_L2. Resolution is approximately 500 m for all fields in MOD10A1 and MOD10A2. Resolution is 0.05° in MOD10C2.


MODIS level 3 daily and 8-day snow cover data (MOD10A1 and MOD10A2, respectively) are georeferenced to an integerized sinusoidal projection. "Integerized" refers to a scheme whereby at each latitudinal row of the grid, a small adjustment is made to the cell's angular width so that there are an integral number of cells covering the entire 360 degrees in that latitude band.

The underlying map projection is sinusoidal, with meridians represented by sinusoidal curves (except for the central meridian) and parallels represented by straight lines. The central meridian and parallels are straight lines of true scale (Pearson 1990). Specific parameters follow:

  • Projection origin: 0 degrees latitude, 0 degrees longitude
  • Orientation: 0 degrees longitude, oriented vertically at top
  • Upper left corner point (m): -20015109.354(x), 10007554.677(y)
  • Lower right corner point (m): 20015109.354(x), -10007554.677(y)
  • True scale (m): 463.312717(x), 463.312717(y)

Grid Description

Level 2 data (MOD10_L2) are produced in five minute segments, which corresponds to approximately 203 scans. At 500 m resolution there are 20 lines per scan, yielding 4060 pixels in the along track direction in each data record. In the cross track direction there are 2708 pixels. At the earth's surface, the coverage of a single MOD10_L2 record is approximately 2030 km along track by 2330 km cross track.

Level 3 daily and 8-day data (MOD10A1 and MOD10A2, respectively) are gridded in equal area tiles. Each tile consists of a 1200 km by 1200 km data array, which corresponds to 2400 pixels by 2400 pixels at 500 m resolution. The following image shows tile locations for MOD10A1 and MOD10A2 data in an integerized sinusoidal projection. Click on the thumbnail image to see the full resolution image.


The CMG consists of a global array of 3600 rows by 7200 columns. Each cell is 0.05° resolution. The following is a sample image derived from MOD10C2 HDF-EOS data in the CMG.


Temporal Coverage

Temporal Resolution

MOD10_L2: Swath (scene, five minutes)
MOD10A1: Daily (day)
MOD10A2 and MOD10C2: eight days

For MOD10A2 and MOD10C2, 8-day periods begin on the first day of the year and extend into the next year. In some cases, there may not be eight days of input, so the user is encouraged to check the RangeBeginningDate and RangeEndingDate objects in the MOD10A2 and MOD10C2 global attributes to determine from which days observations were obtained.

Period Days
------ -----
1 1-8
2 9-16
3 17-24
4 25-32
5 33-40
6 41-48
7 49-56
8 57-64
9 65-72
10 73-80
11 81-88
12 89-96
13 97-104
14 105-112
15 113-120
16 121-128
17 129-136
18 137-144
19 145-152
20 153-160
21 161-168
22 169-176
23 177-184
Period Days
------ -----
24 185-192
25 193-200
26 201-208
27 209-216
28 217-224
29 225-232
30 233-240
31 241-248
32 249-256
33 257-264
34 265-272
35 273-280
36 281-288
37 289-296
38 297-304
39 305-312
40 313-320
41 321-328
42 329-336
43 337-344
44 345-352
45 353-360
46 361-368*

* Includes two or three days from the following year, depending on leap year

Parameter or Variable

Parameter Description

The snow mapping algorithm classifies pixels as snow, snow-covered lake ice, cloud, water, land, or other. Snow-cover extent is the primary variable of interest in this data set.

Parameter Range

Pixel values range from 0 to 254, with 255 used as a fill value. The following table summarizes pixel integer values and their corresponding classifications for MOD10_L2 (Snow Cover field) and MOD10A1 (Day Tile Snow Cover field).

Pixel Classifications
Integer Value Classification
255 Fill data (no data expected for pixel)
254 Saturated MODIS sensor detector
253 Dead MODIS sensor detector
200 Snow
100 Snow-covered lake ice
50 Cloud obscured
39 Ocean
37 Inland water
25 Land (no snow detected)
11 Darkness, terminator or polar
1 No decision
0 Sensor data missing

Sample Data Record

Following is sample data output from a 10 X 10 pixel subset of maximum snow extent (MOD10A2) of the Alaskan North Slope, June 6, 2000:

25 25 25 25 50 50 25 25 25 200
25 25 25 25 25 200 25 25 37 100
25 25 25 25 25 25 25 25 37 37
25 200 25 25 25 200 200 200 50 50
25 25 25 25 25 25 200 200 50 50
25 25 37 37 50 50 25 200 200 20
25 25 100 100 50 50 25 200 200 200
25 25 100 100 50 50 25 200 200 25
25 25 200 200 50 50 25 25 50 50
25 200 200 200 50 50 25 200 50 50

Error Sources

Errors may exist in the reflectance calculations due to the anisotropy of snow and ice. Snow is not a Lambertian reflector and reflects more in a forward direction, particularly with aged snow. Thus, as snow ages, its anisotropy increases. The increase in forward scattering with snow age is greater in the near infrared wavelengths, relative to the visible wavelengths. Such errors will likely be greater at larger angles (30 degrees or more) off nadir as the amount of reflected solar irradiance varies with view angle. Additionally, errors in precise reflectance value due to anisotropy related to topographic variability will be inherent in the data (Hall et al. 1998)

Quality Assessment

Quality indicators are represented by AutomaticQuality Flag and ScienceQualityFlag metadata objects, and their corresponding explanations (AutomaticQualityFlagExplanation and ScienceQualityFlagExplanation) in the CoreMetadata.0 global attribute; and also in a quality assurance field. These are generated during production, or in post-product scientific and quality checks of the data product. The AutomaticQualityFlag is automatically set according to conditions for meeting data criteria in the snow mapping algorithm. In most cases, the flag is set to either "Passed" or "Suspect", and in rare instances, it may be set to "Failed". "Suspect" means that a significant percentage of the data was found to be anomalous and that further analysis should be done to determine the source of anomalies. The AutomaticQualityFlagExplanation contains a brief message explaining the reason for the setting of the AutomaticQualityFlag. The ScienceQualityFlag and ScienceQualityFlagExplanation are set after production, either after an automated QA program is run or after the data product is inspected by a qualified snow scientist. Content and explanation of this flag are dynamic so it should always be examined if present. A sampling of products will be inspected. Random sampling or support of specific events such as field campaigns or snow events may also be conducted.

The QA field provides additional information on algorithm results for each pixel within a spatial context, and is used as a measure of usefulness for snow-cover data. QA data are stored as bit flags. QA information is extracted by reading the bits within a byte. The QA information tells if algorithm results were nominal, abnormal, or if other defined conditions were encountered for a pixel
(Riggs et al. 2001).

Validation by Source

Absolute accuracy is determined by comparing the snow mapping algorithm results with actual ground measurements or aerial photos validated by field measurements. These campaigns are summarized in (Hall et al. 1998). A Landsat TM image of the Sierra Nevada Mountains from May 10, 1992, was classified using a Snow Covered Area (SCA) algorithm, and compared with results of the snow mapping algorithm. Statistical comparison shows that the snow mapping algorithm performed well as a binary snow indicator. 89% of pixels with 50% or greater snow cover according to the SCA algorithm, were classified by the snow mapping algorithm as completely snow-covered, and 79% of those with less than 50% snow cover were classified as snow-free.

Analysis of Northern Hemisphere weekly snow-cover charts and field observations from a February 1994 MODIS Airborne Simulator (MAS) field campaign over Prince Albert National Park suggested that the landscape was 100% snow-covered for three different dates during 1993-95. The snow mapping algorithm mapped those dates with accuracies ranging from 85.7% to 99.9%. The algorithm also mapped up to 37% more snow under some tree canopies, when compared with an earlier algorithm that didn't use a NDVI correction.

A comparison of the algorithm with a Swiss Institute of Technology snow mapping algorithm was performed for a portion of two Landsat TM scenes (May 25, 1994 and July 12, 1994) covering the upper Rhein-Felsburg basin. For the May scene, the MODIS snow mapping algorithm mapped 84.7% of the snow-covered areas as snow. For the July scene, it performed less than expected by only mapping 54.1% of the snow-covered areas as snow. Researchers hypothesize that snow aging and metamorphosis may be responsible for the large errors of omission.

A vegetation density map was constructed with MAS data for April 13, 1995, as part of a validation study in Alaska. In regions with muskeg and mixed forests with a vegetation-cover density less than 50%, the snow mapping algorithm mapped 99% snow cover. In regions with coniferous and deciduous forests, with a vegetation-cover density greater than 50%, the algorithm mapped 98% snow cover. Without the NDVI component included in the algorithm, only 71% of the snow cover was mapped in the forests. The revised algorithm clearly performs better in high density vegetation cover.

Finally, a snow mapping algorithm-derived snow-cover map based on a Landsat TM scene from February 13, 1997, was compared with a snow-cover map produced by the National Operational Hydrologic Remote Sensing Center (NOHRSC). The scene covered a large area in New England. Although the comparison was purely qualitative, there was very good correspondence between the snow-covered areas derived from the two methods (Hall et al. 1998).

Post-launch validation is pending.

Measurement Error for Parameters

Errors with the snow mapping algorithm are lowest in tundra and prairie regions. The maximum expected errors are 15% for forests, 10% for mixed agriculture and forest, and 5% for other land covers. Estimating snow cover is difficult in forests because trees partially or completely conceal underlying snow. These errors were used to estimate the expected maximum monthly and annual errors in Northern Hemisphere snow-mapping methods from the algorithm. The maximum monthly errors are expected to range from 5% to 9% for North America, and from 5% to 10% for Eurasia. The maximum aggregated Northern Hemisphere snow mapping error is estimated to be 7.5%. The error is highest (around 9% to 10%) when snow covers the Boreal Forest, roughly between November and April (Hall et al. 1998). Post-launch error assessment is pending.

5. Usage Guidance

The MODIS Instrument Science Team considers the MODIS snow and ice products as "provisional products" at this time. This implies the following:

Please refer to the MODIS Land Quality Assurance and MODIS Snow/Ice Global Mapping Project Web sites for updates on known data problems, production schedule, and future plans.

Limitations of the Data

The original MOD10C2 algorithm reports only snow percentages greater than or equal to 80%. This high percentage alleviates false snow errors that occurred in the MOD10_L2 Versions 1 and 3 algorithms. An improved version of MOD10_L2 that eliminates false snow errors will be available in early 2002. This improvement will subsequently reduce errors of anomalously low snow percentages in the future CMG products.

Level 2 swaths contain a mix of visible and fill data wherever the MODIS instrument switches from day to night operation. This combination passes through to the CMG snow cover products which show a discontinuity of data through dark regions. Users will note a sharp change between daylight and polar darkness observations. Dark regions extend toward the poles but change irregularly to values of "no observation" (0), followed by another sharp change to fill the data. This happens because the top row of tiles in the integerized sinusoidal grid is not generated when the polar region is in complete darkness. Thus, MOD10C2 users should interpret all regions poleward of the first darkness data as being in darkness.

6. Acquisition Materials and Methods

Source or Platform Collection Environment

NASA/Terra satellite

Source or Platform Mission Objectives

The objective of the mission is to develop and implement algorithms that map snow and ice on a daily basis, and provide statistics of the extent and persistence of snow and ice over an 8-day period. Data at 500 m resolution enables sub-pixel snow mapping for use in regional and global climate models. A study of subgrid-scale snow-cover variability is expected to improve features of a model that simulates Earth radiation balance and land-surface hydrology (Hall et al. 1998).

Source or Platform Program Management

Coverage Information

A ±55 degree scanning pattern at 705 km achieves a 2330 km swath, with global coverage every one to two days.

Data Collection System

The MODIS sensor contains a system whereby visible light from the earth passes through a scan aperture and into a scan cavity to a scan mirror. The double-sided scan mirror reflects incoming light onto an internal telescope, which in turn focuses the light onto four different detector assemblies. Before the light reaches the detector assemblies, it passes through beam splitters and spectral filters that divide the light into four broad wavelength ranges. Each time a photon strikes a detector assembly, an electron is created. Electrons are collected in a capacitor where they are eventually transferred into the preamplifier. Electrons are converted from an analog signal to digital data, and downlinked to ground receiving stations (MODIS Web 2001).

Communication Links

See Data Processing and Acquisition

List of Sensors or Instruments

Moderate Resolution Imaging Spectroradiometer (MODIS)

Ground Segment Information

Data Acquisition and Processing

The EOS Ground System (EGS) consists of facilities, networks, and systems which archive, process, and distribute EOS and other NASA earth science data to the science and user community. The EOS Data and Operations System (EDOS) performs forward-link processing of data and return-link of science data from EOS spacecraft and instruments, processes telemetry to generate level 0 products, and maintains a backup archive of level 0 products. EOSDIS ground stations are a component of EDOS, providing space to ground communication. EOSDIS ground stations comprise the Radio Frequency (RF) ground terminal, EDOS ground station interface, and the EOSDIS Backbone Network (EBnet) telecommunication system. The RF ground terminal provides space to ground link communication channels for receipt of science data, receipt of spacecraft telemetry data and transmission of spacecraft commands for two EOS spacecraft simultaneously, including X-band and S-band capabilities. The EDOS ground station interface monitors and captures the high-rate science data and transfers data to the EDOS level 0 processing facility at the Goddard Space Flight Center (ESDIS 1996).

GSFC processes level 1A data from level 0 instrument packet data, then processes a level 1B Calibrated Radiance product (MOD02) and Geolocation Fields (MOD03). The MODIS SIPS team creates a level 2 product for snow cover (MOD10_L2), which is then used as input to create level 3 gridded products for daily and 8-day snow cover (MOD10A1 and MOD10A2, respectively). These data are archived at the NSIDC DAAC and distributed to EOS investigators and other users via external networks and interfaces (MODIS Web 2001). Data are available to the public through the WIST Data Gateway.

Latitude Crossing Times

Sensor or Instrument Description

The local equatorial crossing time of the NASA/Terra satellite is approximately 10:30 a.m. in a descending node with a sun-synchronous, near-polar, circular orbit.

Key Variables

Pixels are classified as either snow, snow-covered lake ice, cloud, water, land, or "other." Snow-cover extent is the primary variable of interest in this data set.

Principles of Operation

The MODIS instrument provides 12-bit radiometric sensitivity in 36 spectral bands, ranging in wavelength from 0.4 µm to 14.4 µm. Two bands are imaged at a nominal resolution of 250 m at nadir, five bands at 500 m, and the remaining bands at 1000 m. A ±55 degree scanning pattern at 705 km achieves a 2330 km swath, with global coverage every one to two days.

The scan mirror assembly uses a continuously rotating double-sided scan mirror to scan ±55 degrees, driven by a motor encoder built to operate 100% of the time throughout the six year instrument design life. The optical system consists of a two-mirror off-axis afocal telescope which directs energy to four refractive objective assemblies: one each for the visible, near-infrared, shortwave-infrared, and longwave-infrared spectral regions (MODIS Web 2001).

Sensor or Instrument Measurement Geometry

The MODIS scan mirror assembly uses a continuously rotating double-sided scan mirror to scan ±55 degrees, with a 20.3 rpm cross track. The viewing swath is 10 km along track at nadir, and 2330 km cross track at ±55 degrees.

Manufacturer of Sensor or Instrument

MODIS instruments were built to NASA specifications by Santa Barbara Remote Sensing.


MODIS has a series of on-board calibrators that provide radiometric, spectral, and spatial calibration of the MODIS instrument. The blackbody calibrator is the primary calibration source for thermal bands between 3.5 µm and 14.4 µm, while the Solar Diffuser (SD) provides a diffuse, solar-illuminated calibration source for visible, near-infrared, and shortwave infrared bands. The Solar Diffuser Stability Monitor (SDSM) tracks changes in the reflectance of the SD with reference to the sun so that potential instrument changes are not incorrectly attributed to changes in this calibration source. The Spectroradiometric Calibration Assembly (SRCA) provides additional spectral, radiometric, and spatial calibration.

MODIS uses the moon as an additional calibration technique and for tracking degradation of the SD, by referencing the illumination of the moon since the moon's brightness is approximately the same as that of the Earth. Finally, MODIS deep space views provide a photon input signal of zero, which is used as a point of reference for calibration (MODIS Web 2001).

7. References

Earth Science Data and Information System (ESDIS). 1996. EOS Ground System (EGS) systems and operations concept. Greenbelt, MD: Goddard Space Flight Center.

Hall, D.K., A.B. Tait, G.A. Riggs, and V.V. Salomonson. 1998. Algorithm Theoretical Basis Document (ATBD) for the MODIS Snow-, Lake Ice- and Sea Ice-Mapping Algorithms (Version 4.0). Electronic version is available online at .

Hall, D.K., G.A. Riggs, and V.V. Salomonson. 1995. Development of methods for mapping global snow cover using Moderate Resolution Imaging Spectroradiometer (MODIS). Remote Sensing of the Environment 54(2): 127-140.

Hall, D.K. and J. Martinec. 1985. Remote sensing of ice and snow. London: Chapman and Hall.

Markham, B.L. and J.L. Barker. 1986. Landsat MSS and TM post-calibration dynamic ranges, exoatmospheric reflectances and at-satellite temperatures. EOSAT Technical Notes 1:3-8.

MODIS Characterization and Support Team (MCST). 2000. MODIS level 1B product user's guide for level 1B Version 2.3.x Release 2. MCST Document #MCM-PUG-01-U-DNCN.

MODIS Web. MODIS Science and Instrument Team. 2001. .

Pearson II, F. 1990. Map projections: theory and applications. Boca Raton, FL: CRC Press, Inc.

Riggs, G.A., J.S. Barton, D.K. Hall, K. Casey, V. Salomonson. 2001. MODIS snow products users' guide. Electronic version is available online at .

Wiscombe, W.J. and S.G. Warren. 1980. A model for the spectral albedo of snow I: pure snow. Journal of the Atmospheric Sciences 37: 2712-2733.

8. Glossary and Acronyms

Please see the EOSDIS Glossary of Terms for a general list of terms.

List of Acronyms

Please see the EOSDIS Acronyms list for a general list of Acronyms. The following acronyms are used in this document:

ATBD: Advanced Theoretical Basis Document
CMG: Climate Modeling Grid
EASE-Grid: Equal Area Scalable Earth Grid
EBNet: EOSDIS Backbone Network
ECS: EOSDIS Core System
EDOS: EOSDIS Data and Operations System
EGS: EOSDIS Ground System
EOS: Earth Observing System
EOSDIS: Earth Observing System Data and Information System
ESDIS: Earth Science Data Information Project
ESDT: Earth Science Data Type
ftp: file transfer protocol
GMT: Greenwich Mean Time
GSFC: Goddard Space Flight Center
HDF-EOS: Hierarchical Data Format - Earth Observing System
ISIN: Integerized Sinusoidal
MAS: MODIS airborne simulator
MCST: MODIS Characterization Support Team
MODIS: Moderate Resolution Imaging Spectroradiometer
MSS: Multispectral Scanner
NASA: National Aeronautics and Space Administration
NCSA: National Center for Supercomputing Applications
NDSI: Normalized Difference Snow Index
NDVI: Normalized Difference Vegetation Index
NOAA: National Oceanic and Atmospheric Administration
NOHRSC: National Operational Hydrologic Remote Sensing Center
NSIDC: National Snow and Ice Data Center
PVL: Parameter Value Language
QA: Quality Assurance
RF: Radio Frequency
SCF: Science Computing Facility
SD: Solar Diffuser
SDP: Science Data Processing
SDSM: Solar Diffuser Stability Monitor
SIPS: Science Investigator-led Processing System
SRCA: Spectroradiometric Calibration Assembly
TM: Thematic Mapper

9. Document Information

Document Revision Date

February 2006

Document Review Date

January 2002

Document Curator

NSIDC Writers

Document URL