Skip to main content
ARS Home » Plains Area » Bushland, Texas » Conservation and Production Research Laboratory » Soil and Water Management Research » Research » Publications at this Location » Publication #214187

Title: Retrieving leaf area index from remotely sensed data using advanced statistical approaches

Author
item Gowda, Prasanna
item Oommen, Thomas - University Of Michigan
item Misra, Debu - University Of Alaska
item Schwartz, Robert
item Howell, Terry
item Wagle, Pradeep - University Of Oklahoma

Submitted to: GIScience and Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/4/2015
Publication Date: 12/7/2015
Citation: Gowda, P., Oommen, T., Misra, D., Schwartz, R.C., Howell, T.A., Wagle, P. 2015. Retrieving leaf area index from remotely sensed data using advanced statistical approaches. GIScience and Remote Sensing. 4:156, doi:10.4172/2469-4134.1000156.

Interpretive Summary: Leaf area index is a measure of crop leaf density. It is important in photosynthesis, crop water use, and nutrient uptake and yield of crops. For this reason, it is one of the sensitive input parameters in plant growth, evapotranspiration, and water quality simulation models. Manual collection of these data over a large area is costly and time intensive. Remote sensing is a useful tool for rapid collection of leaf area index over larger areas. However, remote sensing based leaf area index models developed for one region may not work in another region. In this study, a set of remote sensing based leaf area index models were developed for Texas High Plains. Results indicate that satellite data using the red and near infrared part of the electromagnetic spectrum were sensitive to leaf area index.

Technical Abstract: Mapping and monitoring leaf area index (LAI) is important for spatially distributed modeling of surface energy balance, evapotranspiration and vegetation productivity. Remote sensing can facilitate the rapid collection of LAI information on individual fields over large areas, in a time and cost-effective manner using empirical relationships between LAI and spectral vegetation indices (SVI). However, these relationships may break down when the effects of sun-surface sensor geometry, background reflectance and atmosphere-induced variations on canopy reflectance are larger than variations in the canopy itself. This requires development of region-specific LAI-SVI models. There are no remote sensing-based LAI models available for the major summer crops in the Texas High Plains. The main objective of this study was to develop empirical relationships between LAI and Landsat Thematic Mapper (TM) based SVIs for major crops in the Texas High Plains. LAI was measured in 47 randomly selected commercial fields in Moore and Ochiltree counties. Data collection was made to coincide with Landsat 5 satellite overpasses on the study area. Numerous derivations of SVIs were examined for estimating LAI using ordinary least square regression models such as linear, quadratic, power and exponential models. The R**2 values for the selected significant models varied from 0.76 to 0.84 with the power function model based on the normalized difference between TM bands 4 and 3 (NDVI) producing the best results. Analysis of the results indicated that the SVI-LAI models based on the simple ratio i.e. the ratio of TM bands 4 and 3, and NDVI are most sensitive to LAI.