All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 132 119 3
PDF Downloads 9 8 4

A Statistical Analysis of Total Ozone Data from the Nimbus-4 BUV Satellite Experiment

View More View Less
  • 1 Department of Statistics, University of Wisconstin, Madison 53706
© Get Permissions Rent on DeepDyve
Restricted access

Abstract

Total ozone data from the Nimbus-4 BUV satellite experiment for the period April 1970–April 1977 is analyzed to: 1) provide a better appreciation and-understanding of the global distribution and behavior of total ozone, with emphasis on any possible trend in ozone, 2) compare with ground-based Dobson total-ozone data for possible instrument calibration and drift errors. and 3) examine the extent to which total ozone data at a particular network of 36 Dobson ground-station locations is representative of global ozone behavior for the detection of trend. Linear regression models containing annual, semi-annual, biennial and linear trend components are estimated for monthly time series of zonal averages of satellite data and the nature of the variation in these components with latitude is presented. An analysis of daily BUV satellite-Dobson ground ozone data is performed which shows a definite downward drift in the satellite data estimated at ∼−1.5 m atm-cm per year relative to a Dobson network of 36 stations. A comparison is also made between trend estimates obtained from satellite data over all locations with those obtained from satellite data near the 36 ground station locations. This comparison indicates that, for purposes of trend detection, the 36 ground station locations seem to provide adequate representation of the entire globe during the 1970's.

Abstract

Total ozone data from the Nimbus-4 BUV satellite experiment for the period April 1970–April 1977 is analyzed to: 1) provide a better appreciation and-understanding of the global distribution and behavior of total ozone, with emphasis on any possible trend in ozone, 2) compare with ground-based Dobson total-ozone data for possible instrument calibration and drift errors. and 3) examine the extent to which total ozone data at a particular network of 36 Dobson ground-station locations is representative of global ozone behavior for the detection of trend. Linear regression models containing annual, semi-annual, biennial and linear trend components are estimated for monthly time series of zonal averages of satellite data and the nature of the variation in these components with latitude is presented. An analysis of daily BUV satellite-Dobson ground ozone data is performed which shows a definite downward drift in the satellite data estimated at ∼−1.5 m atm-cm per year relative to a Dobson network of 36 stations. A comparison is also made between trend estimates obtained from satellite data over all locations with those obtained from satellite data near the 36 ground station locations. This comparison indicates that, for purposes of trend detection, the 36 ground station locations seem to provide adequate representation of the entire globe during the 1970's.

Save