The latest addition to our online maps are concerned with land cover trends at national scale. A super-temporal 16-day MOD13Q1 250m NDVI (normalised difference vegetation index) time series was acquired, covering the period between 2000 and 2014. As introduced by Hird et al. (2015), the second highest annual greenness (shNDVI) values per pixel were extracted and transformed into new variables:
- CHANGE – absolute interannual maximum difference of shNDVI
- AFTEREFFECT – mean shNDVI after i) occurred
- TREND – slope of shNDVI regression covering the whole interannual period
- TIME – Timestamp of i)
The results can be visualised per variable and as a RGB composite raster (MODIS_NDVI_change_2000_2014), where R: CHANGE, G: AFTEREFFECT and B: TREND. When looking at the RGB composite, land form changes such as urbanisation or deforestation, are displayed in yellowish to reddish colours (e.g. negative NDVI trend). Light blueish areas can be related to e.g. transitions from no or young forest to a more mature forest (e.g. positive NDVI trend). For further details please feel free to look at the presentation Christoph Raab (IEOS_2015_CRaab) gave at the IEOS 2015.