Loading...

Monitoring of Precipitation By Merging Surface Gauge Measurements and PERSIANN Satellite Precipitation Product (Case Study: Lake Urmia Basin)

Mirshahi, Amir | 2013

665 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45675 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Abrishamchi, Ahmad
  7. Abstract:
  8. Worldwide coverage, online accessibility and unique accuracy of spatial and temporal satellite data such as rainfall, has encouraged researchers to apply this information in studies such as water resources engineering, hydrologic modeling, drought studies and flood forecast. This research aims to introduce a method that reduces the estimation error of rainfall data derived from a satellite product. “Satellites rainfall products” such as PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks), which has been used in this research, has significant error in comparison with geo-prediction precipitation, due to stochastic and systematic errors. In order to reduce the error many efforts have been made by combining the geo-rainfall data with the satellite products. In this method in order to reduce the estimation error, instead of using the rain gauge data, the Bayesian Theory has been applied to establish a statistical relationship between the geo-rainfall data and the satellite-rainfall data in the past years. The developed statistical model can then be used to reduce the error of the satellite-rainfall product around the real time when no geo-rainfall data is available. This approach had been assessed using the data from the Uroomia Lake basin. 8 years of data at monthly scale (November 2000 to May 2008) was used for model development, and 2 years (2009, 2010) for model evaluation. To derive a monthly geo-rainfall data at the studied basin, different regression methods including Kriging, Regression-Kriging, General Kriging, Co-Kriging and GIDS using the Kernel smoothing method were assessed and Kriging was found to be the best approach. Also to downscale the satellite data from 0.25 to 0.05 degree, the stochastic-spatial downscaling algorithm using the NDVI and elevation variables were used. The final precipitation output derived from the combined Bayesian approach indicated that in our basin in 79% of months of evaluation period it had less error than the satellite-rainfall. On average the two statistical indices of MAE and RMSE of the final output for the region has decreased for 3.92 and 2.99 mm respectively
  9. Keywords:
  10. Bayesian Method ; Kriging Metamodel ; Lake Urmia Watershed ; Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN) ; Downscaling

 Digital Object List

 Bookmark

No TOC