Upscaling of leaf area index in Terai forest plantations using fine- and moderate-resolution satellite data

Abstract

Ecophysiological variables, such as leaf area index (LAI), play a key role in the functioning of ecosystem processes and are thus a useful determinant of primary production, evapotranspiration, and biogeochemical cycling. In the present study, upscaling of LAI was carried out by applying a transfer function from field LAI measurements to fine-resolution LISS III images and subsequently to coarse resolution Moderate Resolution Imaging Spectroradiometer (MODIS) data using a photosynthetically active radiation (PAR)/LAI ceptometer (AccuPAR model LP-80). Field data were collected from differently aged forest plantation types in the Terai Central Forest Division of Nainital district in Uttarakhand, India. The upscaling was done by establishing an empirical exponential relationship between normalized difference vegetation index (NDVI) and LAI. Results reveal a significant relationship (p < 0.01) between NDVI and LAI for each of the studied plantation types, i.e. teak, poplar, eucalyptus, and mixed plantation. The LAI was mapped at 23.5 m resolution by applying a plantation-specific LAI versus NDVI relationship derived from IRS Linear Imaging Self-Scanning Sensor images. The LAI maps were upscaled by using a simple linear averaging within nonoverlapping windows to match with Terra-MODIS 250 m resolution NDVI images. The upscaled time series of LAI was compared with representative field-measured LAI measurements and also with MODIS LAI at a resolution of 1000 m. Upscaled LAI was found to be significantly related to the field-measured LAI with a value of 0.69 for coefficient of determination and with a root mean square error of 0.77. On the other hand, upscaled LAI has less agreement with the MODIS LAI product (R 2 = 0.5 and root mean square error = 0.70).

Retrieving leaf area index of rubber plantation in Hainan Island using empirical and neural network models with Landsat images

“…However, due to the single parameter input and simple calculation, the model has become a practical method for estimating the LAI by remote sensing. [11][12][13]15,16 In previous studies, the research on LAI estimation was based on remote sensing data and field measurement data, mainly focuses on natural forest [70][71][72] and grassland, 51,73 rice, 74,75 wheat, 17,76 corn, 77,78 and other large crops. However, little attention has been paid to the remote sensing research of the tropical crop LAI, such as rubber forest, oil palm forest, cassava, tropical fruit tree, etc.…”

“…For the crop, tree, and fruit variety breeding, growth monitoring, yield prediction, and production management, the LAI is usually used to quantitatively characterize the density and the canopy structure of crop leaves, dynamically monitor crop growth change, diagnose and evaluate crop disease, and estimate crop yield 8 – 16 Furthermore, the LAI is also one of the important indicators for tree and stand structures 17 . Therefore, the rapid and accurate acquisition of the vegetation or agroforestry LAI has great scientific significance in agroforestry ecological ecosystems research and a very important practical value for guiding agricultural and forestry production.…”

“…However, due to the single parameter input and simple calculation, the model has become a practical method for estimating the LAI by remote sensing 11 – 13 , 15 , 16 …”

Predicting maize leaf area index by partial least square combined with wavelet transform

“…Although this method has high measurement accuracy, it is a destructive and time‐consuming method and is limited in practical application (Bréda, 2003). Researchers have developed many approaches for monitoring LAI (Ma et al., 2017; Tripathi et al., 2014; Vyas et al., 2010). For example, the transmittance of canopy radiation is measured, and LAI is calculated with the radiation transfer equation (Ross, 1981).…”