INVESTIGADORES
MARTINEZ PASTUR Guillermo Jose
artículos
Título:
Forest phenoclusters for Argentina based on vegetation phenology and climate
Autor/es:
SILVEIRA, EDUARDA M. O.; RADELOFF, VOLKER C.; MARTÍNEZ PASTUR, GUILLERMO J.; MARTINUZZI, SEBASTIÁN; POLITI, NATALIA; LIZARRAGA, LEONIDAS; RIVERA, LUIS O.; GAVIER PIZARRO, GREGORIO I.; YIN, HE; ROSAS, YAMINA M.; CALAMARI, NOELIA C.; NAVARRO, MARÍA F.; SICA, YANINA; OLAH, ASHLEY M.; BONO, JULIETA; PIDGEON, ANNA M.
Revista:
ECOLOGICAL APPLICATIONS
Editorial:
ECOLOGICAL SOC AMER
Referencias:
Año: 2022
ISSN:
1051-0761
Resumen:
Forest biodiversity conservation and species distribution modeling greatlybenefit from broad-scale forest maps depicting tree species or forest typesrather than just presence and absence of forest, or coarse classifications. Ideally,such maps would stem from satellite image classification based on abundantfield data for both model training and accuracy assessments, but suchfield data do not exist in many parts of the globe. However, different foresttypes and tree species differ in their vegetation phenology, offering an opportunityto map and characterize forests based on the seasonal dynamic of vegetationindices and auxiliary data. Our goal was to map and characterize forestsbased on both land surface phenology and climate patterns, defined here asforest phenoclusters. We applied our methodology in Argentina (2.8 millionkm2), which has a wide variety of forests, from rainforests to cold-temperateforests. We calculated phenology measures after fitting a harmonic curve ofthe enhanced vegetation index (EVI) time series derived from 30-m Sentinel2 and Landsat 8 data from 2018?2019. For climate, we calculated land surfacetemperature (LST) from Band 10 of the thermal infrared sensor (TIRS) ofLandsat 8, and precipitation from Worldclim (BIO12). We performed stratifiedX-means cluster classifications followed by hierarchical clustering. Theresulting clusters separated well into 54 forest phenoclusters with unique combinationsof vegetation phenology and climate characteristics. The EVI 90th percentilewas more important than our climate and other phenology measures inproviding separability among different forest phenoclusters. Our results highlightthe potential of combining remotely sensed phenology measures and climatedata to improve broad-scale forest mapping for different management andconservation goals, capturing functional rather than structural or compositionalcharacteristics between and within tree species. Our approach results in classificationsthat go beyond simple forest?nonforest in areas where the lack ofdetailed ecological field data precludes tree species?level classifications