R.B. Lammers et al (2007) ART-Russia River temperature paper page

This page provides the data for the ART-Russia river temperature study:

Lammers, R.B., J.W. Pundsack, and A.I. Shiklomanov (2007), Variability in river temperature, discharge, and energy flux from the Russian pan-Arctic landmass, J. Geophys. Res. - Biogeosciences, 112, G04S59, doi:10.1029/2006JG000370.

Abstract

We introduce a new Arctic river temperature data set covering 20 gauges in 17 unique Arctic Ocean drainage basins in the Russian pan-Arctic (ART-Russia). Warm season 10-day time step data (decades) were collected from Russian archival sources covering a period from 1929 to 2003 with most data falling in the range from the mid-1930s to the early 1990s. The water temperature data was combined with river discharge data to estimate energy flux for all basins and over the Russian pan-Arctic as a whole. Tests for trend were carried out for water temperature, river discharge, and energy flux. Spatially coherent significant increases in the maximum decadal river temperature were found in the European part of the Russian pan-Arctic. Several other drainage basins showed significant changes but there was no strong pattern either in the connections between variables or spatially. The trend in area averaged energy flux for the 3 largest drainage basins (Ob, Yenisey, Lena) combined was found to be significantly decreasing. We speculate that, in the Yenisey basin, this decrease was due to large impoundments of river water. The lack of consistency between temperature and energy flux trends was due to the difference in timing between peaks in river temperature and river discharge. The mean area averaged energy flux from the Russian basins was 0.2 Wm^-2. Using this mean we estimated the total energy flux from the entire Russian pan-Arctic, both gauged and ungauged, to be 82 EJy^-1.

Data

Data is available in two files, the station attributes file which contains information about each station and the data file containing several time series of river temperature, discharge, etc.

Station attributes file: tab delimited ASCII text file with headers

Data file (time series): tab delimited ASCII text file with headers

Combined attributes and data file: R object
where the data file is the 'D4' data frame and the station attributes are in the 'AA' data.frame. The R object was created by the R program for statistical computing and was used to analyze the data and create the graphs in this paper.

Description of Fields

Station attributes

The station attribute file has 1 header line and 20 data lines with one line per station and 10 fields of data:

  Field   Description
  =====   ===========
  ID      ART-Russia unique station identification number, range 1-20.    
  Code    Station code from RosHydromet and used by R-ArcticNet and ArcticRIMS,
          use this field to link to the data file
  Name    Station Name, typically "[River] At [Location]"
  Lat     Latitude of station in decimal degrees
  Long    Longitude of station in decimal degrees
  DArea   Drainage area of drainage basin upstream from station (km²)
  Start   First year with data
  End     Last year with data
  NYears  Number of years with data
  Gaps    Free form field with notes on missing data (inherited from early form of the data)

Data file

the data file has 1 header line and 54000 data lines with one line per station-decade and 16 fields of data. The data are sorted by station first then chronologically within each station. All stations have record lines for all 75 years from 1929 to 2003 regardless of the availability of data.

54000 lines == 20 gauges x 75 years x 12 months/year x 3 decades/month.

Field Description
===== ===========
Code Station code from RosHydromet and used by R-ArcticNet and ArcticRIMS,
use this field to link to the station attribute file
Date Date in US format MM/DD/YYYY in the range "01/05/1929"-"12/25/2003"
Year Year of data point in the range 1929-2003
Month Month of data point in the range 1-12
Decade Decade (10-day period) of data point in the range 1-3
(see special note on decades below)
DaysPerDecade Number of days during each decade in the range 8-11
TemperatureCorrected River water temperature in the range 0-25.3 degrees Celsius.
This corresponds to the the "T0" data from the paper.
AvgOfQkm3perDay Mean river discharge at station in the range 0 - 15.11 km³/day
StartIceCover Start of ice cover formation in river. Value == 1 or NA. Data from Vuglinsky (2000)
StartIceDrift Start of ice drift in river. Value == 1 or NA. Data from Vuglinsky (2000)
Tflux Energy flux calculated from temperature in range 0 - 4139 PetaJoules/decade
NewTemperature River water temperature in the range 0-25.3 degrees Celsius.
This corresponds to the the "T1" data from the paper.
MonthlyQ Monthly river discharge from R-ArcticNet in the range 0 - 10.28 km³/day
Interpolated data used for stations with missing daily data.
NewQ River discharge in the range 0 - 15.11 km³/day
NewerTemperature River water temperature in the range 0-25.3 degrees Celsius.
This corresponds to the the "T2" data from the paper.
NewTflux Energy flux calculated from temperature in range 0 - 4152 PetaJoules/decade

No data values are given as "NA"

Special note on "decade"

We had several discussions on the use of the word "decade" while working on the paper and we felt decade was the least ambiguous and least cumbersome option. In English the common usage of the word decade refers to a 10 year period, however, definitions of decade include any grouping of ten not just 10 years:

"1. a. An assemblage, group, set, or series of ten" (Oxford English Dictionary)
"1: a group or set of 10" (Merriam-Webster online)
"3. a group, set, or series of ten" (dictionary.com)

However, while acceptable, we do acknowledge the word "decade" is not fully satisfactory as not all of the periods are exactly 10 days long. The periods represent each month divided into 3 periods with the mid-points of each period on the same day of the month (5th, 15th, 25th). This means the number of days per decade ranges from 8 to 11 depending on length of month and whether or not it is a leap year. We are interested to know if there is a suitable alternative. Please send us email with your suggestions.

Contact

The corresponding author for the paper and data is Richard.Lammers@unh.edu