Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #316307

Title: Framework for automated spatio-temporal enhancement of coarse resolution leaf area index (FASE-LAI) – Application to MODIS LAI

Author
item HOUBORG, RASMUS - Collaborator
item MCCABE, M. - Collaborator
item Gao, Feng

Submitted to: Remote Sensing of Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/18/2015
Publication Date: 12/12/2015
Publication URL: https://handle.nal.usda.gov/10113/61855
Citation: Houborg, R., Mccabe, M., Gao, F.N. 2015. Framework for automated spatio-temporal enhancement of coarse resolution leaf area index (FASE-LAI) – Application to MODIS LAI. Remote Sensing of Environment. 47:15-29.

Interpretive Summary: Leaf area index (LAI) is a key biophysical parameter used for land surface flux estimation, water resource management and crop growth monitoring. While coarse resolution LAI at the kilometer scale are available from the Moderate Resolution Imaging Spectroradiometer MODIS) satellite sensor and often sufficient for global, continental and regional scale applications, field scale applications require LAI with both high spatial and temporal resolution. In this paper, an Automated Spatio-temporal Enhancement of coarse-resolution leaf area index (LAI) products (FASE-LAI) is built to generate 4-day time-series of Landsat-scale LAI using MODIS LAI product and Landsat surface reflectance product. The approach integrates a reference-based LAI retrieval approach and the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM). Results are evaluated over a maize and soybean producing region in central Nebraska from 2002 to 2012. This paper demonstrates an operational approach for producing 4-day time-series of LAI from Landsat imagery for crop modeling at field scales which is required by the National Agricultural Statistics Service and Foreign Agricultural Service for crop condition monitoring and yield estimation.

Technical Abstract: A multi-scale satellite-based Framework for Automated Spatio-temporal Enhancement of coarse-resolution leaf area index (LAI) products (FASE-LAI) has ben established to generate 4-day time-series of Landsat-scale LAI, thereby meeting the critical demands of applications needing frequent and high spatially explicit information to effectively resolve rapidly evolving vegetation dynamics at sub-field scales. In this study, FASE-LAI is applied to LAI products based on Moderate Resolution Imaging Spectroradiometer (MODIS) sensor data, and integrates a reference-based regression tree approach for producing MODIS-consistent Landsat-based LAI, and the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) for intelligently interpolating the downscaled LAI between Landsat acquisitions. With input of consistent data streams, STARFM predicts Landsat-scale LAI by blending MODIS and Landsat based information from a base date with MODIS data from the prediction date. Application over an agricultural region in Nebraska, demonstrates encouraging utility of FASE-LAI for reproducing fine-scale spatial features in actual Landsat LAI with varying durations between base and prediction date, but optimal prediction results are contingent upon appropriate specification of tunable STARFM parameters and input options, and consistency of input data streams. In addition, the implementation of 250 m resolution LAI, derived from MODIS 1 km products using a NDVI-based biome-specific approach, significantly improved accuracies of spatial pattern prediction with the coefficient of efficiency (E) ranging from 0.77 – 0.94 compared to 0.01 – 0.85 when using 1 km LAI inputs. Comparisons against an 11-year record of in-situ measured LAI over maize and soybean demonstrates utility of FASE-LAI for reproducing observed LAI dynamics over different plant development stages (r2=0.86). FASE-LAI represents a powerful and intelligent interpolation mechanisms and predicts in-situ measured LAI better than estimates derived through linear interpolation between Landsat acquisitions, particularly when the measurement date is more than 10 days away from the nearest Landsat acquisition, with up to 50% reductions in prediction error. With a streamlined and completely automated processing interface, FASE-LAI represents a flexible tool for LAI disaggregation in space and time, readily adaptable to different land cover types, landscape heterogeneities and cloud cover conditions. Extension to other available global LAI products is underway and FASE-LAI may then also serve as a mechanism for efficient product validation and intercomparison over regions of interest.