On investigation of density correction and improved covariance effect on local modeling of gravity field by least squares collocation method in Iran

Due to the lack of sufficient information about the density of subsurface masses, in the modeling of the earth gravity field, usually a constant global average density value is used for the entire studied area. However, any accuracy improvement of the mass density used in the modeling of topographic gravitation will increase the accuracy of gravity field modeling. To approve this quantitatively, a topographic density model with a resolution of 30×30 arc second is prepared from the processing of seismographic maps and satellite data of the layers of the lithosphere and used instead of a constant mass density in four study regions inside Iran with diverse topographic and data distribution. These four regions have the dimension of 2.5×3 degrees. The point intervals in the first and third regions (R1 and R3) are approximately 5 minutes, while in the second and fourth regions (R2 and R4) the intervals are approximately 13 minutes. In areas with the same point distribution, R1 has a relatively smoother topography than R3. The topography in R2 is relatively rougher than R4. In addition to the global average density value, the average value for Iran and the region is also included in modeling the gravity topographic masses. Owe to topography diversity, these areas seem to be suitable for a fear investigation of the RTM technique performance. In modeling the gravity field, the least squares collocation method and, consequently, the RTM technique are used in modeling the effect of topographic mass gravitation. In order to evaluate the effect of lateral density variations when using the RTM technique for gravity field modeling of the earth, the least squares collocation method is used in this research. The RCR technique is used for gravity field modeling by least squares collocation method and the effect of the global topography is removed. To remove the global gravitational effect from the observations, the EIGEN6C4 model degree and order 360 is used. To remove the effect of topography by RTM method, a digital elevation model with a point density of 1 second of arc is used. In addition to density correction, the use of improved covariance algorithm in gravity field modeling is also evaluated in this research. The results show that in the areas with more topography, and as such more density variations, the effect of density modification in removing the effect of topography from the gravity anomalies signal of the region is more significant. Furthermore, comparison to the control points of this study shows that application of density correction and the improved covariance algorithm in areas with rough topography and lack of sufficient gravimetric data and proper distribution, the accuracy of gravity field modeling can be improved by 15.6%. Using the IC approach in four regions leads to an increase in modeling accuracy. Among these regions, the highest increase in accuracy is related to region 2 (with a decrease of 1 mGal of standard deviation) and relatively related to region 3 (with a decrease of 11.5% in standard deviation). These two areas have rougher topography than areas 1 and 4.