Virtual Climate Station Network (VCSN) data technical description

Virtual Climate Station Network (VCSN) data technical description

  • VCSN data (except Soil Moisture, see notes below) is derived from climate station observations using a thin-plate smoothing spline model for the VCSN spatial interpolations. This model incorporates two location variables (latitude and longitude) and a third "pattern" variable. For example, for rainfall the 1951–80 mean annual rainfall digitised from an expert-guided contour map is used to aid the interpolation. The software used for the interpolations is ANUSPLIN1.
  • VCSN data is generally provided as CSV files, containing daily data for all variables as per the following field description. 

Field definitions 

Note: Note the time range of the data in the following fields e.g. "at 9am", "from 9am", "to 9am", etc, especially when comparing with datatypes from other sources. 

Field 
Units 
Example Time of Observation (Local) 
Description 
Station 
  
  
This is the CLIDB agent number of the station i.e. the "station identifier" 
The file header also contains the VCSN “Network Number” and Latitude / Longitude Information  
Date 
local 
  
Refers to the day starting at the date indicated. The date includes the hour/minute "00:00" and refers to the day starting at the date indicated. Time should be ignored 
  
The time range of the data in the following fields e.g. "at 9am", "from 9am", "to 9am", etc, especially when comparing with datatypes from other sources. (See note on time range change below.) 
WindSpeed 
mps 
2015-06-15 00:00 to 2015-06-16 00:00 
Mean wind speed in m/s at 10m above ground level over 24 hours FROM midnight local day. 
VapPress 
hPa 
2015-06-15 09:00 
Vapour pressure in hPa AT 9am local day. 
Tmin 
degC 
2015-06-14 09:00 to 2015-06-15 09:00 
Minimum 24 hour temperature in DegC TO 9am local day. Note take care when comparing this with the daily combined "Form 301" minimum temperature - this will be one day earlier. 
Tmax 
degC 
2015-06-15 09:00 to 2015-06-16 09:00 
Maximum 24 hour temperature in degC FROM 9am local day. 
TempEarth10cm 
degC 
2015-06-15 09:00 
The 10cm earth temperature in degC AT 9am local day. 
SoilM 
mm 
2015-06-15 09:00 to 2015-06-16 09:00 
24 hour soil moisture index in mm FROM 9am local day calculated from rainfall and evapotranspiration. The base value is -150mm ("permanent wilting point") based on "soil store capacity". A value of "0" indicates the soil is at "field capacity" (amount of water held in the soil after the excess has drained away). A value greater than "0" indicates runoff. See the note below "Note on Soil Moisture Calculation" below for more information. 
RH 
% 
2015-06-15 09:00 
Relative humdity in percent AT 9am local day. 
Rain 
mm 
2015-06-15 09:00 to 2015-06-16 09:00 
Total 24 hour amount of rain in mm FROM 9am local day. 
Radiation 
MJ/m2 
2015-06-15 00:00 to 2015-06-16 00:00 
24 hour amount of accumulated global solar radiation in MJ/m2 (Mega Joules per square metre) FROM midnight local day. 
PET 
mm 
2015-06-15 09:00 to 2015-06-16 09:00 
24 hour Penman Potential Evapotranspiration total in mm FROM 9am local day. 
MSLPress 
hPa 
2015-06-15 09:00 
Pressure reduced to Mean Sea Level in hPa AT 9am local day. 

Soil moisture data

Soil moisture is derived (in mm) using a 1D water balance model, as described in Kandel et al. (2005)1. The model has one parameter, the soil store capacity (Smax) and utilises the VCSN data of daily rainfall (Rain) and daily Penman Potential Evapotranspiration (PET). Rainfall is assumed to infiltrate into a soil store from which moisture is depleted by evapotranspiration loss. The model keeps track of the soil store for each timestep. When the soil store is full, it is assumed to be saturated and the excess rainfall becomes surface runoff. Evapotranspiration is assumed to continue at its potential rate until the soil water store drops to half capacity. Evapotranspiration then decreases linearly to zero when the store is empty. For the simulations here, a fixed Smax value of 150 mm is used for all the modelling grids (so no variability of soil type is considered). 

Data currency (mainly relevant for automated access) 

VCSN data requires many climate stations observations to calculate reliable VCSN surfaces. Due to the timing of incoming observational data for interpolation, there is some latency when VCSN data are generated. This may be up to 3 days prior to the current day. 

Examples for latest data: 

  • Before 1500 (current time in NZST), today's VCSN run has not been completed, and the latest data available is 3 days old. 
  • After 1500 (current time in NZST), today's VCSN run has been completed, and the latest data available is 2 days old. 
  • Example: 
  1. Access date is 18-Apr-2013 0900, latest data available is 15-Apr-2013. 
  2. Access date is 18-Apr-2013 1700, latest data available is 16-Apr-2013. 

Each month the VCSN data for the previous month are regenerated to account for any issues with data quality and to use additional manually collected rainfall data. Therefore, there may be changes to VCSN data previously accessed.

Reference

1. Kandel DD, Chiew FHS, Grayson RB, 2005. A tool for mapping and forecasting soil moisture deficit over Australia. Cooperative Research Centre for Catchment Hydrology, Technical Report 05/02, 18 pp. (ISBN 1-920813-21-7).