Towards monitoring stem growth phenology from space with high resolution satellite data

0573738 - ÚVGZ 2024 RIV NL eng J - Journal Article
Eitel, J. U.H. - Basler, D. - Braun, S. - D'Odorico, P. - Buchmann, N. - Etzold, S. - Gessler, A. - Griffin, K.L. - Krejza, Jan - Luo, Y. - Maguire, A. - Rao, M.P. - Vitasse, Y. - Walthert, L. - Zweifel, R.
Towards monitoring stem growth phenology from space with high resolution satellite data.
Agricultural and Forest Meteorology. Roč. 339, AUG (2023), č. článku 109549. ISSN 0168-1923. E-ISSN 1873-2240
Research Infrastructure: CzeCOS IV - 90248
Institutional support: RVO:86652079
Keywords : planetscope * dendrometer * intra-annual tree growth * radial growth phenology index (rgpi) * remote sensing
OECD category: Environmental sciences (social aspects to be 5.7)
Impact factor: 6.2, year: 2022
Method of publishing: Limited access
https://www.sciencedirect.com/science/article/abs/pii/S016819232300240X?via%3Dihub

Radial stem growth is a key ecosystem process resulting in long-term carbon sequestration. Despite recognition of its importance to global carbon cycling, high uncertainties remain regarding how radial growth phenology (e.g., the onset, mid, and cessation of radial growth) will be affected by climate change. In this study, we evaluated to what extent high spatially (3 × 3 m) and temporally (up to daily) resolved satellite imagery from PlanetScope can be used to monitor stem growth phenology. For this, we made use of detailed stem growth phenological observations of six common European tree species measured by automated point dendrometers at 14 distinct sites across Switzerland between 2017 and 2021. These growth phenological observations were then linked through multiple regression modeling with metrics extracted from spectral index time series. Our results show that the remote sensing-based models enable monitoring the onset (root mean squared deviation (RMSD) ranges from 5.96 to 27.04 days) and mid-stages of stem growth (RMSD ranges from 10.20 to 36.34 days) with reasonable accuracy as opposed to the cessation of stem growth that showed low accuracy (RMSD ranges from 16.02 to 153.63 days). The accuracy of the remote sensing-based prediction models and their optimal suite of predictors varied across species. The latter has important implications for the remote sensing of stem growth phenology in mixed forests, suggesting that it is important for satellite sensors to resolve individual tree crowns. Overall, our results suggest the need for novel spectral indices that capture the spectral components of mechanistic linkages between stem growth and canopy properties that go beyond the mere detection of leaf phenology. When employing such spectral indices, remote sensing could make it possible to detect not only shifts in leaf phenology caused by climate change but also those in stem growth on a broad spatial scale.
Permanent Link: https://hdl.handle.net/11104/0344121