Intrinsic vegetation memory as a proxy of engineering resilience may be an oversimplification.


Repeated climate stress events may cause fundamental shifts in species compositions or ecosystem functioning. One reason for higher stability of ecosystems than previously expected may be ecological stress memory of vegetation. The study of memory effects of large-scale vegetation may therefore aid in predictions of future changes in biome distributions and resilience assessments on ecosystem or even species level. We assessed patterns of vegetation memory from 1982 to 2015 across four large dryland study regions (Iberian Peninsula, Caatinga, Australia, Contiguous US) using PCA regression analyses of Standardised Normalised Difference Vegetation Index data as well as soil moisture and air temperature data from the ERA5 reanalysis product.We have identified vegetation memory patterns using novel input variables which has led to a marked enhancement of our understanding of extrinsic and intrinsic vegetation memory components. Our findings show a strong overlay of intrinsic (NDVI[t-1]) and extrinsic (soil moisture) memory patterns across dryland regions with mixed memory lengths (most informative lags). Overall, our results show that soil moisture within the shallowest soil layers is the most important property in determining plant performance across drylands (i.e. 0-7cm). Our study highlights the usefulness of environmental variables (i.e. soil moisture) and data sets (i.e. ERA5) which have not yet found widespread application in biological studies/modelling approaches. Additionally, we have implemented a sophisticated statistical downscaling method for preparation of climate data which should prove useful for a wider spatial modelling community. We continue to work on a separation of effect components. This would be highly desirable as it would allow for a clear functional understanding of vegetation forcing via ecosystem internal or external drivers.

Oct 3, 2019 11:00 — 12:00
Salzburg, Austria
Erik Kusch
Erik Kusch
PhD Student

In my research, I focus on statistical approaches to understanding complex processes and patterns in biology using a variety of data banks.