Vegetation dynamics of the past 20 years as seen in NOAA AVHRR time series

In order to understand how the Earth functions as a system, knowledge about the global distribution of the vegetation as well as the different vegetation types and their temporal and spatial variations is necessary. Vegetation is an important part of the Earth's system at the interface between land and atmosphere because it influences processes such as the latent heat flux and surface Albedo. The Advanced Very High Resolution Radiometer (AVHRR) of the National Oceanic and Atmospheric Administration (NOAA) is qualified for the global monitoring of variable systems such as vegetation. The AVHRR has a repetition rate of 12 hours and a spatial resolution of 1.1km x 1.1 km at nadir which is sufficient for vegetation monitoring on a global scale. The AVHRR allows the derivation of vegetation indices, e.g. the Normalized Difference Vegetation Index (NDVI), from the characteristic spectral signature measured over vegetated areas.

Since mountainous areas are most likely to react more sensitively to climate change than the surrounding landscapes, it is interesting to have a closer look at the alpine vegetation dynamics of the past 20 years. This is the goal of the project presented here. The major scientific questions are:

Is there an increase of vegetation activity (length of the growing season and photosynthetic activity) in the Alps and surrounding landscapes between 1987 and 2006?

  1. Which differences are detectable in different altitude zones?
  2. Is there a clear difference between North-South and East-West Alps?
  3. Is there a correspondence between the phenological data and satellite derived vegetation index?
  4. Which are the effects of extreme events, such as dry and warm summer 2003?

 

Associated Master Thesis

Vegetation is highly sensitive to climate change. Therefore, vegetation changes are robust indicators of the dynamic responses of ecosystems to climate change. Satellite remote sensing is suited to measure phenological dynamics at a large scale. Advanced Very High Resolution Radiometer (AVHRR) sensors flown on National Oceanic and Atmospheric Administration (NOAA) satellites and since 2006 on board of European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) Meteorological Operational (MetOp) satellites provide daily daytime scenes dating back to the early 80’s, serving as an ideal medium for detecting vegetation trends. This thesis uses a 23-year (1990- 2012) AVHRR time series of the Normalized Differenced Vegetation Index (NDVI) to estimate the photosynthetic activity over the European Alps. Measuring vegetation changes using remote sensing is subject to limitations, demanding data pre-processing that entails calibration, geocoding, orthorectification, and snow/cloud masking to eliminate default values. Further noise reduction is achieved by using Maximum Value Composite (MVC) with an upper envelope, and finally the Double Logistic fitting-function (DL) is applied to smooth the time series. Based on processed data, results show a temporal shift towards a higher NDVI, induced both by earlier start of season (SOS), 3.7 days per decade, and later end of season (EOS), 3.2 days per decade. Over the 23 years, results demonstrate strong seasonal and inter-annual variability influenced by climate conditions and external factors such as the Pinatubo eruption. Four profiles across the European Alps illustrate the spatial dynamics. As expected, as elevation increases, the vegetative signal decreases and SOS occurs later, while EOS occurs at more or less the same day of the year (DOY) regardless of elevation. Over the years, there is a shift towards higher NDVI at all altitudinal levels. The results show that vegetation is dependent on temperature related to seasonality. To a certain degree, the higher the temperature, the higher the NDVI, but temperatures that are too high obstruct vegetation growth. While temperature has an immediate influence, precipitation has a delayed effect. The results show that phenological shifts can partially serve as indicators of climate change impacts. Future studies have to quantify these impacts in order to increase understanding of how vegetation feedbacks influence climate.

Master Student: Michael Burkhalter

Supervisor: Stefan Wunderle