Search

Evapotranspiration (ET) flux constitutes a major component of both the water and energy balances at the land surface. Among the many factors that control evapotranspiration, phenology poses a major source of uncertainty in attempts to predict ET. Contemporary approaches to ET modeling and monitoring frequently summarize the complexity of the seasonal development of vegetation cover into static phenological trajectories (or climatologies) that lack sensitivity to changing environmental conditions. The Event Driven Phenology Model (EDPM) offers an alternative, interactive approach to representing phenology. This study presents the results of an experiment designed to illustrate the differences in ET arising from various techniques used to mimic phenology in models of land surface processes. The experiment compares and contrasts two realizations of static phenologies derived from long-term satellite observations of the Normalized Difference Vegetation Index (NDVI) against canopy trajectories produced by the interactive EDPM trained on flux tower observations. The assessment was carried out through validation of predicted ET against records collected by flux tower instruments. The VegET model (Senay, 2008) was used as a framework to estimate daily actual evapotranspiration and supplied with seasonal canopy trajectories produced by the EDPM and traditional techniques. The interactive approach presented the following advantages over phenology modeled with static climatologies: (a) lower prediction bias in crops; (b) smaller root mean square error in daily ET -- 0.5mm per day on average; (c) stable level of errors throughout the season similar among different land cover types and locations; and (d) better estimation of season duration and total seasonal ET. [ABSTRACT FROM AUTHOR]

Phenologies of the vegetated land surface are being used increasingly for diagnosis and prognosis of climate change consequences. Current prospective and retrospective phenological models stand far apart in their approaches to the subject. We report on an exploratory attempt to implement a phenological model based on a new event driven concept which has both diagnostic and prognostic capabilities in the same modeling framework. This Event Driven Phenological Model (EDPM) is shown to simulate land surface phenologies and phenophase transition dates in agricultural landscapes based on assimilation of weather data and land surface observations from spaceborne sensors. The model enables growing season phenologies to develop in response to changing environmental conditions and disturbance events. It also has the ability to ingest remotely sensed data to adjust its output to improve representation of the modeled variable. We describe the model and report results of initial testing of the EDPM using Level 2 flux tower records from the Ameriflux sites at Mead, Nebraska, USA, and at Bondville, Illinois, USA. Simulating the dynamics of normalized difference vegetation index based on flux tower data, the predictions by the EDPM show good agreement (RMSE < 0.08; r² > 0.8) for maize and soybean during several growing seasons at different locations. This study presents the EDPM used in the companion paper (Kovalskyy and Henebry, 2011) in a coupling scheme to estimate daily actual evapotranspiration over multiple growing seasons. [ABSTRACT FROM AUTHOR]

We used remote sensing imagery to characterize the hydrological dynamics of 8404 temporary freshwater wetlands (playas) in Texas (Landsat 5 TM WRS-2 P30/R36) from 2008 to 2011, comparing known wet and dry periods, and related these to land use within 100 m. Hydroperiods were highly variable, and peak water availability occurred in different seasons in different years, varying by as much as two orders of magnitude with date. Land use affected the likelihood and duration of inundation, with playas in urban settings being modified in such a way as to extend hydroperiod, and playas surrounded by cropland experiencing shorter hydroperiods: 3726 playa basins never contained standing water during the four-year period, and many of these were surrounded by cotton, corn, wheat, or sorghum. In contrast, 25 playas never dried, and most of these were surrounded by urban development. Median hydroperiod was 17-109 days, being longer during the relatively wet year of 2010 compared to exceptional drought in 2011. Remote sensing was useful in monitoring playa surface water fluctuations as a function of land use, providing an alternative source of data in the absence of ground-based hydrological records, and granting a temporal perspective that may otherwise not exist for seasonal or ephemeral wetlands. [ABSTRACT FROM AUTHOR]