| Title | Artificial Intelligence Analysis of Integrated Remote and In-Situ Sensing Measurements of Near Surface Air Temperature |
| |
| Author | Garabedian, A; Dagnew, M; Bitsuamlak, G; Bandari, N; Veillette, S; Fedosejevs, G; Latifovic, R; Park, W M; Fung, K B; Teillet, P M |
| Source | 25th Canadian Remote Sensing Symposium & 11th Congress of the Association québécoise de télédétection, Montréal, Québec, Canada, October 14-17; 2003. |
| Year | 2003 |
| Alt Series | Earth Sciences Sector, Contribution Series 20043253 |
| Document | book |
| Lang. | English |
| Media | paper |
| Abstract | The work described in this paper exploited data mining methodologies to integrate and increase understanding of the relationship between in-situ measurements of near surface air temperature at
meteorological stations and remotely sensed data. The remote sensing data consisted of thermal infrared data from the Advanced Very High Resolution Radiometer (AVHRR) sensors on NOAA-14 and NOAA-16. The surface data were obtained from Environment
Canada meteorological stations. A neural network approach made it possible to develop a variety of models that generated temperature estimates at reasonably fine spatial resolution guided by both surface measurements and remote sensing data. Such
estimates were obtained for all land surfaces at a resolution of one square kilometre over a region encompassing the Island of Montreal, Canada. This paper describes the models employed along with results in terms of the best combinations of input
remote sensing parameters for optimum performance in estimating temperature, as well as in terms of different spatial and temporal cases involving grouped and single meteorological stations. |
| GEOSCAN ID | 220055 |
|
|