Retrieval of leaf water content from remotely sensed data using a vegetation index model constructed with shortwave infrared reflectances

Abstract

The timely and accurate estimation of leaf water content (LWC) is of great practical significance for monitoring the state of vegetation growth and forecasting crop yield. As the spectral capabilities of terrestrial spectroscopy instruments and high-spatial-resolution satellite sensors steadily increase, people pay more and more attentions to the application potential for the spectral characteristics of the plant in shortwave infrared (SWIR) domain. The major purpose of this article was to investigate the relationship between leaf spectral reflectance and LWC in SWIR. A normalized vegetation index model constructed with SWIR reflectances centred at 1725 and 2200 nm yielded a high overall performance compared to the previous normalized difference water index (NDWI). The Leaf Optical Properties Experiment 1993 data set was introduced to validate the LWC estimated with the proposed model. The results showed that the accuracy of LWC estimation from this index can reach 0.0046 g cm−2, which is better than that from traditional NDWI with root mean square error equalled to 0.0104 g cm−2. The effects of band width and random noise on the LWC estimation from these two indices were also analysed and the results showed that LWC retrieved from both indices were insensitive to band width choices and random noise variations.

Abstract from the paper’s OpenAlex record, shown for indexing and research. Copyright remains with the authors or the publisher. Read at the source ↗

The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.