Research interests

  • Satellite Remote sensing of terrestrial ecosystems
  • Ecosystem change
  • Biogeography
  • Biogeochemical cycles
  • Geographic Information system

  • Research projects


    Terrestrial remote sensing

    I worked on various remote sensing methodologies for estimating biogeophysical properties of the land surface, such as albedo and evapotranspiration (ET), and for detecting fire-affected areas. Evapotranspiration (ET) is an important flux that links water, carbon, and energy exchange in many ecosystems, but is considered the most uncertain component in water budget, due a lack of observational data and inadequate parameterization of the underlying physics in hydrology models. I developed a semi-empirical algorithm based on Priestley-Taylor (PT) approach to estimate ET at regional to continental scales using satellite observations. Monthly net radiation estimated using MODIS albedo and land surface temperature was in good agreement with field measurement. Another key parameter, PT coefficient, was successfully optimized as a function of leaf area index and soil moisture using eddy covariance measurements from 27 AmeriFlux sites. Monthly ET at a 1 km spatial resolution was produced for the conterminous United States using MODIS observations. The validation over 7 AmeriFlux sites and over continental river basins showed better performance compared with the official MODIS ET product. A spatial analysis of the newly derived continental ET estimates showed that PT coefficient was the main driver for the spatial variation of ET in arid areas, whereas net radiation controlled ET when mean annual precipitation was higher than 400 mm yr-1, indicating a shift from water-limited to energy-limited ET with increasing precipitation.


    Ecosystem response to precipitation variability

    Water availability is a major regulator of Net Primary Production (NPP) in arid and semi-arid regions. A clear distinction between the implications of short- and long-term precipitation variation, and an improved understanding of the amplitude and reversibility of the corresponding ecological impact, is needed to anticipate the ecological consequences of climate change. I separated the spatial and temporal responses of diverse ecosystems across a large productivity and climate gradient throughout California. I used 24 years of AVHRR satellite data and 11 years of MODIS data to derive gridded information on ecosystem function. Long-term mean annual NPP increased with mean precipitation spatially across the landscape and leveled off over wetter areas. The temporal sensitivity of NPP to year-to-year variation in precipitation was much lower than that derived from the spatial relationship. The temporal sensitivity decreased over the driest areas, limited by the growth potential, and decreased over wetter areas due to other biogeochemical constraints, resulting in the highest sensitivity in grasslands. Mean annual precipitation explained 34% of variations in the site-level ecological responses, and the plant functional type, precipitation variability, and stand age were also significant controls.

    Temporal slope



    The impact of fire disturbance on vegetation dynamics, biophysical properties, and the associated radiative forcing

    Fire is a dominant disturbance agent in boreal forest, affecting climate through several feedbacks, including changes in carbon balance and alteration of surface energy exchange. I combined digital fire perimeter data and MODIS satellite observations to create chronosequences of vegetation dynamics, albedo, and surface shortwave forcing in boreal Canada over the first 40 years after fire. I found that more severe fires caused more rapid vegetation recovery and considerably larger post-fire albedo increases during spring, consistent with the relationship between burn severity and post-fire species establishment documented in field studies. These results suggest that, under an intensifying fire regime, increases in negative forcing associated with the amplified post-fire albedo increase may at least partly offset carbon losses associated with increases in burn severity observed in some areas of boreal forests.

    In a separate study, I found large differences in vegetation recovery trajectories and seasonal post-fire albedo changes along a pronounced ecoclimatic transect. Changes in post-fire spring albedo and surface forcing increased with latitude, primarily because of delayed snow melt, which amplified albedo differences between unburned forests and recovering stands. This latitudinal trend was counteracted by the south-north decrease of winter incoming solar radiation and summer albedo change, resulting in a similar mean annual surface forcing (-4.1 ± 0.9 W m-2) along the transect. This study highlights the tradeoff between snow cover and vegetation structure in regulating latitudinal variations of fire-induced surface forcing in boreal forests.


    Interactions between fire, climate change, ecosystems, and human activities

    Southern California has experienced a series of extraordinarily large and destructive fires over the past decade. Currently I am leading a large interdisciplinary study to understand how interactions between climate change, ecosystems, and human activities change the fire regime and subsequent impacts on air quality in Southern California.



    Yufang Jin -- last updated in 2012