Data Set ID: 
SNEX17_SSD

SnowEx17 Time Series Sonic Snow Depth Measurement Array, Version 1

This data set contains 15-min snow depth observations for two study sites on Grand Mesa, CO, USA, acquired as part of NASA's 2017 SnowEx campaign. The data were recorded using two arrays of Judd Communications Ultrasonic Depth Sensors, configured as a TLS K footprint on the west side of the mesa and a TLS N footprint in the east. The sensors were positioned to represent three primary vegetation conditions: open-canopy; canopy-edge; and closed-canopy. A total of 10 and 7 sensors recorded usable data at the west and east sites, respectively, from the beginning of the snow season in November 2016 through the end in June 2017.

These data can be used for a variety of purposes, including: model forcing, calibration, and validation; evaluation of airborne and satellite remote sensing data; to analyze how vegetation affects snow accumulation and ablation.

This is the most recent version of these data.

Version Summary: 

New data set

STANDARD Level of Service

Data: Data integrity and usability verified

Documentation: Key metadata and user guide available

User Support: Assistance with data access and usage; guidance on use of data in tools

See All Level of Service Details

Parameter(s):
  • SNOW/ICE > SNOW DEPTH
Data Format(s):
  • JPEG
  • Comma-Separated Values (.csv)
Spatial Coverage:
N: 39.03430447, 
S: 39.02755193, 
E: -107.9335668, 
W: -108.0545976
Platform(s):GROUND-BASED OBSERVATIONS
Spatial Resolution:
  • Varies x Varies
Sensor(s):ULTRASONIC DEPTH SENSOR
Temporal Coverage:
  • 12 October 2016 to 3 July 2017
Version(s):V1
Temporal Resolution15 minuteMetadata XML:View Metadata Record
Data Contributor(s):Keith Jennings, Theodore Barnhart, Noah Molotch

Geographic Coverage

Other Access Options

Other Access Options

Close

As a condition of using these data, you must cite the use of this data set using the following citation. For more information, see our Use and Copyright Web page.

Jennings, K. S., T. B. Barnhart, and N. P. Molotch. 2018. SnowEx17 Time Series Sonic Snow Depth Measurement Array, Version 1. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. doi: https://doi.org/10.5067/5YJEYNLS1YK4. [Date Accessed].
Created: 
9 July 2018
Last modified: 
2 October 2018

Data Description

Parameters

Snow depth.

File Information

Format

Snow depth data are provided in Comma-Separated Values (.csv) format. 

Multiple images (.jpg) of each sensor location are included. One text file (.txt) describing the contents of each image is also available.

The CSV file, JPG images, and text file are available as a single ZIP file. 

An associated Extensible Markup Language (.xml) file contains relevant metadata.

File Contents

Snow depth time series data are presented in a single CSV file. This file contains four levels of data - raw, quality controlled, infilled, and smoothed. Quality control (QC) and infill flags describe how the raw data was manipulated.

QC flags:
NA = Data passed all QC checks
1 = Observation removed because of vegetation effects
2a = Observation failed maximum / minimum threshold

2b = Observation failed rate of change threshold

Infill flags:
NA = No infilling performed
3a = Missing observation filled using linear interpolation (gap ≤ 24 h)
3b = Missing observation filled using spline interpolation (gap > 24 h)
4 = Observation failed visual inspection, was removed manually and infilled with linear interpolation
5a = Filled from beginning of snow season (2017-11-18 10:30 MST) to date of first usable observation using linear interpolation

5b = Filled from beginning of snow season (2017-11-18 10:30 MST) to date of first usable observation using average depth values from nearby working sensors, scaled to the first usable observation.

The CSV file header includes the campaign, location, names of researchers, missing data code, QC and infill flag codes, post-processed differential GPS coordinates for the 17 snow depth sensor locations, and notes. 

Starting with row 40, data are presented in columns A through H. Columns include the date & time (MST) of measurement, sensor number, raw snow depth (cm), quality controlled snow depth (cm), QC flag, infilled snow depth (cm), infill flag, and smoothed snow depth (cm). Column titles and ten rows of sample data are presented in Figure 1.

Figure 1: Sample data from file SnowEx17_snowdepth_15min_V2.csv.

Naming Convention

Image files utilize the following naming convention:

sx??_N_xxxx.JPG

where:

Table 2: File Naming Convention 
Variable Description
sx??
sensor ID number:
sxk? = sensor in the TLS-K footprint on the west mesa
sxn? = sensor in the TLS-N footprint on the east mesa
N
number of image (four images associated with each sensor site)
xxxx
description of image:
map = hand-drawn site map (not to scale) with pencil pointing to sensor location / number
site = wide-angle shot of sensor
down = sensor footprint
up = sky view taken from beneath sensor

Examples:

sxn9_1_map.JPG
sxn9_2_site.JPG
sxn9_3_down.JPG

sxn9_4_up.JPG

File Size and Volume

The CSV file is approximately 24 MB.

JPG files range between approximately 1 MB and 7.5 MB.

The text file is approximately 1 KB.

The total data set is approximately 326 MB.

Spatial Information

Coverage

Overall spatial coverage:
Northernmost Latitude: 39.03430447° N
Southernmost Latitude: 39.02755193° N
Easternmost Longitude: 107.9335668° W
Westernmost Longitude: 108.0545976° W

Resolution

1 cm

Temporal Information

Coverage

Data from the TLS-K footprint (SXK sensors) were collected between 12 October 2016 and 02 July 2017.

Data from the TLS-N footprint (SXN sensors) were collected between 12 October 2016 and 06 June 2017.

Resolution

Depth sensors recorded snow depth observations every 15 minutes

Data Acquisition and Processing

Background

Data can be used for model forcing, calibration, and validation; evaluation of airborne and satellite remote sensing products; and an analysis of vegetation's effects on snow accumulation and ablation, among other purposes.

Acquisition

Data were measured at two locations on Grand Mesa, Colorado: the TLS-K footprint (in the west mesa study site) and the TLS-N footprint (in the east mesa study site). Ten sensors recorded usable data from TLS-K; seven sensors recorded usable data from TLS-N. Sensors were positioned within each study site to capture three vegetation conditions: open-canopy, canopy-edge, and closed-canopy.

Snow depth was recorded using Judd Communications Ultrasonic Depth Sensors. Every 15 minutes, the sensors directed an ultrasonic pulse at the ground and recorded the two-way travel time. The instrument then calculated snow depth by dividing the air-termperature-corrected speed of sound by 1/2 the two-way travel time. Snow depth sensors have a 22° beamwidth and a range of 0.5 to 10 meters.

Snow depth records were stored on Campbell Scientific CR1000 data loggers.

Processing

Raw data were subjected to quality control (processing steps one through four), infilling (processing steps five through eight), and smoothing (processing step nine).

Quality Control:
  1. Data was visually inspected. Observations tainted by vegetation influences (e.g. grass growth in the sensor footprint) were removed.
  2. Observations were removed if they exceeded the maximum / minimum threshold.
  3. Data points with a change in depth greater than ±5 cm per 15 minutes were removed.
  4. Sensor offsets were corrected to correspond to 0 cm during snow-free periods. All sensors, except SXK6, SXN3, SXN6, had their offsets adjusted so that depth = 0 cm when there was no snow on the ground. Offset corrections were less than 15 cm in all cases.
Infilling:
  1. Data gaps less than 24 hours were infilled using linear interpolation between the preceding and following observations.
  2. Data gaps greater than 24 hours were infilled using spline interpolation between the preceding and following observations.
  3. Data were again visually inspected. Any questionable data points that passed the previous QC inspection were removed and infilled using linear interpolation.
  4. For those sensors whose observations were affected by vegetation during the snow-free season, depth was reconstructed by averaging values from nearby, non-affected sensors and then scaling these observations to match the first usable observation of the vegetation-affected sensor.

Smoothing

  1. Depth values were smoothed using a 6-hour moving average filter, where each observation in the 3-hour window before and after the data point was given equal weight. Other smoothing approaches are available to users and can be performed on the infilled data (column F, depth_fill_cm).

Quality, Errors, and Limitations

Sensors SXN2, SXN5, and SXN7 were installed but recorded no usable data. No data from these sensors were retained for the final product.

The percentage of infill required varied by sensor, ranging between 1.7% and 64.0%. All sensors except SXK5 and SXN3 had fewer than 18.6% infill. The mean rate of infill was 14.5%; the median rate of infill was 9.0%.

JPG images are meant to illustrate site conditions only. Cameras were not leveled or precisely located, so no attempts should be made to quantify vegetation height, leaf area index, or other parameters from the images.

Instrumentation

Description

Snow depth was measured by Judd Communications Ultrasonic Depth Sensors. See the manufacturers website for specifications.

Snow depth data were stored on Campbell Scientific CR1000 data loggers. See the manufacturers website for more details.

Related Data Sets

Other SnowEx Data Sets

Related Websites

NASA SnowEx Campaign

CU Mountain Hydrology Group

Contacts and Acknowledgments

Keith S. Jennings
University of Colorado Boulder
Boulder, CO 80309

USA

Theodore B. Barnhart
Univeristy of Colorado Boulder
Boulder, CO 80309

USA

Noah P. Molotch
University of Colorado Boulder
Boulder, CO 80309
USA

References

Molotch, N. P., P. D. Brooks, S. P. Burns, M. Litvak, R. K. Monson, J. R. McConnell, and K. Musselman. 2009. Ecohydrological controls on snowmelt partitioning in mixed conifer sub-alpine forests. Ecohydrology 2(2): 129-142.

How To

Programmatic Data Access Guide
Data from the NASA National Snow and Ice Data Center Distributed Active Archive Center (NSIDC DAAC) can be accessed directly from our HTTPS file system or through our Application Programming Interface (API). Our API offers you the ability to order data using specific temporal and spatial filters... read more