| Title | Earth observation based land cover for regional aquifer characterization |
| |
| Author | Latifovic, R; Pouliot, D; Nastev, M |
| Source | Canadian Water Resources Journal vol. 35, no. 4, 2010 p. 433-450, https://doi.org/10.4296/cwrj3504433 |
| Image |  |
| Year | 2010 |
| Alt Series | Earth Sciences Sector, Contribution Series 20100005 |
| Publisher | Informa UK Limited |
| Document | serial |
| Lang. | English |
| Media | paper; on-line; digital |
| File format | pdf |
| Subjects | hydrogeology; geophysics; remote sensing; aquifers; groundwater; groundwater resources; LANDSAT imagery; LANDSAT |
| Illustrations | Landsat images; satellite imagery; tables; histograms; diagrams |
| Program | Public Safety Geoscience Quantitative risk assessment |
| Released | 2010 01 01 |
| Abstract | Managing groundwater resources requires quantification of several complex atmosphere/surface properties affecting recharge, including precipitation, temperature, soils, land cover and land cover change.
In this review we elaborate on some of the remote sensing techniques commonly applied for generating information about changes in land cover, over large geographical areas, which are often required for hydrogeological studies. Two case studies are
presented: the Chateauguay River Basin (Quebec) and Casselman Township (Ontario). Both examples represent agriculture dominated landscapes that require a unique mapping approach to minimize confusion between agriculture and forest classes. The first
example demonstrates a procedure for mapping land cover aerial extent and change using remote sensing data acquired off the peak of the growing season where agricultural fields can be most effectively discriminated from forests. In the second example
three classification methods are compared for mapping specific crop types. Results did not show substantially greater performance for any one of the three methods. Differences in processing and theoretical advantages are considered the main criteria
for selection. The analyzes undertaken highlight the need to minimize phenology effects between image dates, by selecting images before leaf out or after leaf senesce for effective change detection and using several dates of imagery over the growing
season to effectively map crop types. |
| GEOSCAN ID | 262748 |
|
|