Time-series of land surface phenology （LSP） data offer insights about vegetation growth patterns. They can be generated by exploiting the temporal and spectral reflectance properties of land surface components. Inte...Time-series of land surface phenology （LSP） data offer insights about vegetation growth patterns. They can be generated by exploiting the temporal and spectral reflectance properties of land surface components. Inter- annual and seasonal LSP data are important for under- standing and predicting an ecosystem＇s response to variations caused by natural and anthropogenic drivers. This research examines spatio-temporal change patterns and interactions between terrestrial phenology and 28 years of climate dynamics in Central Asia. Long-term （1981-2008） LSP records such as timing of the start, peak and length of the growing season and vegetation productivity were derived from remotely sensed vegetation greenness data. The patterns were analyzed to identify and characterize the impact of climate drivers at regional scales. We explored the relationships between phenologi- cal and precipitation and temperature variables for three generalized land use types that were exposed to decade- long regional drought events and intensified land and water resource use： rainfed agriculture, irrigated agriculture, and non-agriculture. To determine whether and how LSP dynamics are associated with climate patterns, a series of simple linear regression analyses between these two variables was executed. The three land use classes showed unique phenological responses to climate variation across Central Asia. Most of the phenological response variables were shown to be positively correlated to precipitation and negatively correlated to temperature. The most substantial climate variable affecting phenological responses of all three land use classes was a spring temperature regime. These results indicate that future higher temperatures would cause earlier and longer growing seasons.展开更多
文摘Time-series of land surface phenology （LSP） data offer insights about vegetation growth patterns. They can be generated by exploiting the temporal and spectral reflectance properties of land surface components. Inter- annual and seasonal LSP data are important for under- standing and predicting an ecosystem＇s response to variations caused by natural and anthropogenic drivers. This research examines spatio-temporal change patterns and interactions between terrestrial phenology and 28 years of climate dynamics in Central Asia. Long-term （1981-2008） LSP records such as timing of the start, peak and length of the growing season and vegetation productivity were derived from remotely sensed vegetation greenness data. The patterns were analyzed to identify and characterize the impact of climate drivers at regional scales. We explored the relationships between phenologi- cal and precipitation and temperature variables for three generalized land use types that were exposed to decade- long regional drought events and intensified land and water resource use： rainfed agriculture, irrigated agriculture, and non-agriculture. To determine whether and how LSP dynamics are associated with climate patterns, a series of simple linear regression analyses between these two variables was executed. The three land use classes showed unique phenological responses to climate variation across Central Asia. Most of the phenological response variables were shown to be positively correlated to precipitation and negatively correlated to temperature. The most substantial climate variable affecting phenological responses of all three land use classes was a spring temperature regime. These results indicate that future higher temperatures would cause earlier and longer growing seasons.
