Remote sensors provide us unprecedent abilities to observe phenomena at large scales. Ecological studies shed light on mechanisms linked to the impacts of global change on ecosystems, but obtaining accurate field data is a hard work and, as a result, many of these studies have been developed at small scales and extrapolating to broader territories is not straightforward. SEÑALES project is aimed at establishing links between ground at remote sensors data to scale up from plot to regional views. In doing so, SEÑALES relates remote sensor data to ground data in three study cases to solve three different applied problems: 1) we aim to explain fungal productivity time-series with climate and remote sensing data, developing models for spatio-temporal patterns of mushroom harvest; 2) we use satellite data and Artificial Intelligence to reconstruct late frost defoliation along Spanish beech forest distribution between 2003 and 2018, and 3) we try to relate tree growth in beech forests to NDVI data as a tool to predict shifts in carbon gain.
Project VA026P17 funded by Junta de Castilla y León.
- García-Pedrero A, García-Cervigón AI, Olano JM, García-Hidalgo M, Lillo-Saavedra M, Gonzalo-Martín C, Caetano C, Calderón-Ramírez S (in press) Convolutional Neural Networks for segmenting xylem vessels in stained cross-sectional images. Neural Computing and Applications -link-
- Sangüesa-Barreda G, Villalba R, Rozas V, Christie DA, Olano JM (2019) Detecting Nothofagus pumilio growth reductions induced by past spring frosts at the northern Patagonian Andes. Frontiers in Plant Science 10: 1413