Modeling grassland spring onset across the Western United States using climate variables and MODIS-derived phenology metrics Qinchuan Xin a, ⁎, Mark Broich b , Peng Zhu a , Peng Gong a,c,d a Ministry of Education Key Laboratory for Earth System Modeling, Tsinghua University, Beijing, China b School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, Australia c Environmental Science, Policy and Management and Geography, University of California, Berkeley, CA, USA d Joint Center for Global Change Studies, Beijing, China abstract article info Article history: Received 23 July 2014 Received in revised form 3 February 2015 Accepted 3 February 2015 Available online xxxx Keywords: Remote sensing Phenology model Flux tower Climate variability Vegetation phenology strongly controls photosynthetic activity and ecosystem function and is essential for mon- itoring the response of vegetation to climate change and variability. Terrestrial ecosystem models require robust phenology models to understand and simulate the relationship between ecosystems and a changing climate. While current phenology models are able to capture inter-annual variation in the timing of vegetation spring onset, their spatiotemporal performances are not well understood. Using green-up dates derived from MODIS, we test 9 phenological models that predict the timing of grassland spring onset via commonly available climato- logical variables. Model evaluation using satellite observations suggests that Modified Growing-Degree Day (MGDD) models and Accumulated Growing Season Index (AGSI) models achieve reasonable accuracy (RMSE b 20 days) after model calibration. Inclusion of a photoperiod trigger and varied critical forcing thresholds in the temperature-based phenology model improves model applicability at a regional scale. In addition, we ob- serve that AGSI models outperform MGDD models by capturing inter-annual phenology variation in large semi- arid areas, likely due to the explicit consideration of water availability. Further validation based on flux tower sites shows good agreement between the modeled timing of spring onset and references derived from satellite observations and in-situ measurements. Our results confirm recent studies and indicate that there is a need to calibrate current phenology models to predict grassland spring onsets accurately across space and time. We dem- onstrate the feasibility of combining satellite observations and climatic datasets to develop and refine phenology models for characterizing the spatiotemporal patterns of grassland green-up variations. © 2015 Elsevier Inc. All rights reserved. 1. Introduction Vegetation phenology, characterizing the recurring and periodic cy- cles of vegetation green-up and senescence, is highly sensitive to cli- mate change and variability (Cleland, Chuine, Menzel, Mooney, & Schwartz, 2007; Koerner & Basler, 2010; Piao, Fang, Zhou, Ciais, & Zhu, 2006; Richardson et al., 2013). Environmental drivers, such as tempera- ture, photoperiod, water and nutrient availability, regulate the timing of the spring onset of natural vegetation (Friedl et al., 2014; Piao et al., 2011; Yu, Price, Ellis, & Shi, 2003). Numerous studies using in-situ mea- surements and satellite observations have documented decadal shifts in vegetation phenology under a changing climate at both regional and global scales (Broich et al., 2014; Julien & Sobrino, 2009; Wu & Liu, 2013; Yang, Mustard, Tang, & Xu, 2012). The shifts of key phenophases, such as spring onset and autumn senescence, control vegetation photo- synthetic activities (Churkina, Schimel, Braswell, & Xiao, 2005; Richardson et al., 2010) and have profound impacts on global carbon and water cycles in both field measurements and model simulations (Dragoni et al., 2011; Jeong, Medvigy, Shevliakova, & Malyshev, 2012; Piao, Friedlingstein, Ciais, Viovy, & Demarty, 2007). Robust climate- driven models of vegetation phenology are therefore critical for projecting climate change scenarios (Cramer et al., 2001; Levis & Bonan, 2004). Modeling springtime vegetation phenology via climate variables has received extensive attention in recent publications, and a variety of climate-driven phenological models have been proposed and tested using in-situ measurements (Cesaraccio, Spano, Snyder, & Duce, 2004; Melaas, Richardson, et al., 2013; Richardson, Bailey, Denny, Martin, & O'Keefe, 2006; Yang et al., 2012). Based on species-level observations of tree budburst, it is generally considered that temperature is the main driver for spring onsets of temperate forests (Bale et al., 2002; Chuine, Cour, & Rousseau, 1999; Hanninen & Kramer, 2007; Kaduk & Los, 2011; Wu, Gonsamo, Gough, Chen, & Xu, 2014). The temperature- based phenology models have been widely employed as sub-models in terrestrial biosphere models (Cramer et al., 2001; Kucharik et al., 2006). Most of these phenology models are empirical, with prescribed Remote Sensing of Environment xxx (2015) xxx–xxx ⁎ Corresponding author at: Tsinghua University, Mengminwei South Building Room 920, Beijing 100084, China. E-mail address: xqcchina@gmail.com (Q. Xin). RSE-09312; No of Pages 15 http://dx.doi.org/10.1016/j.rse.2015.02.003 0034-4257/© 2015 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Remote Sensing of Environment journal homepage: www.elsevier.com/locate/rse Please cite this article as: Xin, Q., et al., Modeling grassland spring onset across the Western United States using climate variables and MODIS-derived phenology metrics, Remote Sensing of Environment (2015), http://dx.doi.org/10.1016/j.rse.2015.02.003