2008 Alaska Park Science Symposium
October 14, 2008
On the Difficulty to Gain Regionally Representative Measurements in National Parks
Debasish Pai Mazumder1, Nicole Mölders2
1Department of Atmospheric Sciences, University of Alaska Fairbanks, 903 Koyukuk Drive, Fairbanks, AK, 99775, USA, Phone 907-474-7618, debasish@gi.alaska.edu
2Department of Atmospheric Sciences, Geophysical Institute & College of Natural Science and Mathematics, 903 Koyukuk Drive, Fairbanks, AK, 99775-7320, USA, molders@gi.alaska.edu
Due to power logistic reasons and to maintain the attractiveness to the tourist, observational sites in national parks can often not be placed in those locations that would be optimal for obtaining representative regional values. The location of sites and coarseness of observational networks in national parks may introduce uncertainty in the reliability of regional observational values. To estimate this uncertainty related to network density and/or design, a case study has been performed over Siberia by using Weather Research and Forecasting (WRF) Model simulations for July and December, 2005. Regional averages for various meteorological and soil quantities determined for areas of 2.8° x 2.8° based on all WRF data are considered as “reference”. Regional averages determined from four artificial networks with 500, 400, 200, and 100 randomly taken WRF grid-points as “sites” and 411 “sites” that correspond to the locations of a real network are compared with the reference regional averages. The real network has appreciable difficulties in reproducing the reference regional averages for all quantities because its non-uniform site distribution misrepresents the landscape. Its biases in regional averages of sea-level pressure, wind-speed, 2m-temperature, relative humidity, precipitation, short- and long-wave radiation reach up to 140 hPa (160 hPa), 2.8 m/s (8 m/s), 12 K (20 K), 40% (24%), 4 mm/d (9 mm/d), 140 Wm-2 (100 Wm-2) and 60 Wm-2 (100 Wm-2), in July (December). Such errors can strongly affect the ability to assess fire danger and the possibility of flash floods in national parks. The greatest errors occur in coastal and mountainous terrain. The 100-sites-network has some difficulties in reproducing the reference sea-level pressure due to its low density. The artificial networks with 200 sites or more reproduce the reference regional averages, trends and variability for all quantities.
Previous Abstract |
Next Abstract