Variabilidade espacial da altitude elipsoidal determinada por GNSS, ASTER GDEM e SRTM

It was studied the spatial variability of elevation obtained by GNSS, SRTM and ASTER GDEM. And 3,884 altitude data were collected with receptors Global Navigation Satellite Systems (GNSS). Altitude data from SRTM and ASTER models extracted by the nearest neighbor technique were used for comparison. The geostatistical analysis was performed by adjusting semivariogram spherical, exponential and gaussian being used the Akaike information criterion, the sum of squares value and kriging standard deviation to select the best fit model. Then, the interpolation of the data by ordinary kriging. Geostatistical possible to characterize the structure and magnitude of spatial dependence of altitude. There was spatial dependence for the three data sets studied. For GNSS data, ASTER and SRTM the best model is the spherical. The largest error of kriging were obtained by SRTM data and the least error for GNSS data.