Resumen de Estudi de la dinàmica de la vegetació des de satèl·lit: respostes fenològiques a el canvi climàtic
Phenology is key to control physicochemical and biological processes, especially albedo, surface roughness, canopy conductance and fluxes of carbon, water and energy. High-quality retrieval of land surface phenology (LSP) is thus increasingly important for understanding the effects of climate change on ecosystem function and biosphere–atmosphere interactions. Remote sensing is a useful tool for characterizing LSP although no consensus exists on the optimal satellite dataset and the method to extract phenology metrics.
I aimed to (i) improve the retrieval of Land Surface Phenology from satellite data, (ii) validate LSP with ground observations and near surface remote sensing, and (iii) understand the relationships between climate variables and phenology in a climate change context, as well as to assess the responses of vegetation to extreme events. These three main research objectives are explored in the three chapters of the thesis.
In chapter 2, I investigated the sensitivity of phenology to (I) the input vegetation variable: normalized difference vegetation index (NDVI), leaf area index (LAI), fraction of absorbed photosynthetically active radiation (FAPAR), and fraction of vegetation cover (FCOVER); (II) the smoothing and gap filling method for deriving seasonal trajectories; and (III) the phenological extraction method: threshold, logistic-function, moving-average and first derivative based approaches. The threshold-based method applied to the smoothed and gap-filled Copernicus Global Land LAI V2 time series agreed the best with the ground phenology, with root mean square errors of ~10 d and ~25 d for the timing of the start of the season (SoS) and the end of the season (EoS), respectively.
In the third chapter, I took advantage of PhenoCam and FLUXNET capability of continuous monitoring of vegetation seasonal growth at very high temporal resolution (every 30 minutes). This allows a more robust and accurate comparison with LSP derived from satellite time series avoiding problems related to the differences in the definition of phenology metrics. I validated LSP estimated from LAI time series with near-surface PhenoCam and eddy covariance FLUXNET data over 80 sites of deciduous broadleaf forest. Results showed a strong correlation (R2 > 0.7) between the satellite LSP and ground-based observations from both PhenoCam and FLUXNET for the timing of the start (SoS) and R2 > 0.5 for the end of season (EoS). The threshold-based method performed the best with a root mean square error of ~9 d with PhenoCam and ~7 d with FLUXNET for the timing of SoS, and ~12 d and ~10 d, respectively, for the timing of EoS.
In the fourth chapter, I investigated the spatio-temporal patterns of the response of deciduous forests to climatic anomalies in the Northern Hemisphere using LSP derived in Chapter 1 and validated in Chapter 1 and Chapter 2, and multi-source climatic data sets for 2000–2018 at resolutions of 0.1°. I also assessed the impact of extreme heatwaves and droughts on deciduous forest phenology. Analyses of partial correlations of phenological metrics with the timing of the start of the season (SoS), end of the season (EoS), and climatic variables indicated that changes in preseason temperature played a stronger role than precipitation in the interannual variability of SoS anomalies: the higher the temperature, the earlier the SoS in most deciduous forests in the Northern Hemisphere (mean correlation coefficient of -0.31). Both temperature and precipitation contributed to the advance and delay of EoS. A later EoS was significantly correlated with a positive standardized precipitation-evapotranspiration index (SPEI) at the regional scale (~30% of deciduous forests). The timings of EoS and SoS shifted by >20 d in response to heat waves throughout most of Europe in 2003 and in the United States of America in 2012.
