Land surface phenology dynamics and climate variations in the North East China Transect (NECT), 1982–2000

Abstract

Land surface phenology dynamics reflect the response of the Earth's biosphere to inter‐ and intra‐annual dynamics of the Earth's climate and hydrologic regimes. Investigations of land surface phenology dynamics and its relation to long‐term climate variation could help us to detect the response of regional vegetation to climate variation. The present study developed a new algorithm for detecting regional land surface phenology dynamics (ARLSPD) and demonstrated it in detecting the vegetation response to inter‐annual climate variability in the North East China Transect (NECT), a mid‐latitude semi‐arid terrestrial transect with strong gradients in environmental conditions and vegetation formations. The spatial–temporal patterns of greenup‐onset date, maturity date, and senescence date during the period of 1982–2000 are presented. The resultant spatial–temporal patterns of land surface phenology were quite consistent with the land‐cover characteristics, moisture, and temperature gradients. The relations between inter‐annual variations in phenology and seasonal climate were investigated. It was found that besides human disturbance, land surface phenology depended primarily on the combined effects of preseason temperature and precipitation. The relative influence of preseason temperature and precipitation on land surface phenology was changing, which led to the different responses of land surface dynamics to climate variation along the moisture gradient in the NECT. In the arid and semi‐arid region of NECT, the dates of onset for phonological events in temperate typical grassland were most significantly related to the precipitation during the preceding 2–4 months. Temperature‐induced drought stress during the preceding months could delay greenup onset in cropland/grassland mosaic, and advance senescence in temporal typical grassland, and in cropland/grassland mosaic. The regional phenology algorithm, theoretically also applicable for complex ecosystems characterized by annual multiple growth cycles, is expected to couple with large‐scale biogeochemical models to regulate dynamically land surface phenology.