| Title | Accuracy assessment of annual land cover time series derived from change-based updating |
| |
| Author | Pouliot, D; Latifovic, R |
| Source | MultiTemp 2013, final program; by MultiTemp 2013; 2013 p. 148 |
| Image |  |
| Year | 2013 |
| Alt Series | Earth Sciences Sector, Contribution Series 20130349 |
| Meeting | Mulittemp 2013 7th International Workshop on the Analysis of Multi-Temporal Remote Sensing Images; Banff; CA; June 25-27, 2013 |
| Document | book |
| Lang. | English |
| Media | on-line; digital |
| File format | pdf |
| Subjects | analytical methods |
| Program | Remote Sensing Science |
| Released | 2013 01 01 |
| Abstract | Detailed information on the spatial and temporal distribution of land cover is required to evaluate the effects of land cover change on environmental processes. The development of temporally consistent
land cover time series from satellite-based earth observation has proven difficult due to variability in sensor observations. This leads to spurious land cover differences between maps when standard supervised classification approaches are applied
for land cover time series generation. To reduce this effect a common solution has been to first detect change and update a base map for only these change areas. As long as the change commission error is low this approach will ensure high consistency
between maps in the time series. Here we describe the implementation and accuracy assessment of a change based updating approach used to develop a (2000-2010) annual land cover time series over Canada from MODIS 250
spatial resolution data. The
method incorporates both abrupt and gradual changes in an effort to capture a range of change types. To predict the new land cover label, an evidential reasoning approach was used to combine spectral, spatial, and temporal sources of
evidence.
Assessing the accuracy of land cover time series is challenging because multiple maps need to be assessed for both land cover classification and change detection accuracies. Regarding a change based updating approach, for a time series
where only a small percent change occurs, accuracy is close to that of the original base map. Over longer periods where significant change has accumulated the accuracy becomes more dependent on the change and update labelling accuracy. Thus, accuracy
for a change based approach can be seen as a function of the base map, change detection, and update accuracies. A specific formulization is developed to
summarize these components. Results show that the time series accuracy was in a large degree
predetermined by the base map accuracy because there was only a small amount of change (~5%) over the period and the base map and update accuracies were similar. As the time series is extended and more change accumulates results suggest a slight
decrease due to the accumulation of omission error assuming that change and update accuracies are temporally invariant. Increasing the update accuracy by a few percent, within the precision of the accuracy estimates, would result in more accurate
time series. |
| Summary | (Plain Language Summary, not published)
This is a technical document on a method to assess accuracy of land cover time series that have been developed through a change based update
approach. |
| GEOSCAN ID | 293386 |
|
|