Large scale ecosystem stability measures are required to understand the mechanisms determining the strength of ecosystems to withstand environmental perturbations. The three main ecosystem stability measures, i.e. resistance, resilience and variance, can be extracted from Remote Sensing (RS) time series data. RS data is available regularly and at large spatial scale and as such provides an interesting avenue to upscale our level of observations on ecosystem stability. However, noise and sensor based characteristics of the RS time series might influence the RS based ecosystem stability measures. Therefore this study quantifies the effect of (i) white noise, (ii) biased noise, (iii) missing values, (iv) the length of the time series and (v) temporal resolution on the stability measures of NDVI, EVI and LAI time series of 16 ecosystems. The study reveals a high sensitivity (decrease in R2) of the ecosystem stability measures to all types of noise. This is especially true for resilience measures. For white noise and biased noise, the decrease in R2 seems to be similar for all stability measures.