Evaluating canopy water content in grasslands using in situ hyperspectral data: A case from the Eastern Mediterranean, Turkey

• Ahmet Karakoç ^[1] ; Murat Karabulut ^[2]
  1. [1] Osmaniye Korkut Ata University

     Osmaniye Korkut Ata University

     Turquía

     
  2. [2] Department of Geography, Faculty of Humanities and Social Sciences, Kahramanmaraş Sütçü İmam University, Kahramanmaraş, Turkey
• Localización: Mediterranean Botany, ISSN-e 2603-9109, Vol. 45, Vol. 2, 2024
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • The water content of vegetation is considered a key parameter for ecological analysis and agricultural and forestry applications. Remote sensing techniques provide substantial benefits over conventional field methods in determining vegetation water content at the leaf, canopy, and landscape scales. This study evaluated the potential of hyperspectral vegetation indices in predicting canopy water content in grasslands. Data was gathered from three different grasslands situated at approximately 500 m asl, 1200 m asl, and 1400 m asl elevations. Each study area provided 71 samples, and a total of 213 samples were analyzed. In this context, 59 ratio-based hyperspectral vegetation indices were tested. The correlation between hyperspectral vegetation indices and canopy water content was evaluated using linear, exponential, logarithmic, and power regression models. The results showed that the NW-3 (920,970) index significantly represents the canopy water content variable. It was determined that the exponential regression model created with this index could explain the variations in canopy water content up to 85%.On the other hand, it has been detected that the high level of water content in the vegetation creates a significant saturation problem. Another finding of this study is that the predictive power reaches higher levels in low canopy water content characteristics. The results of thi