Leaf Area Index Variations in Ecoregions of Ardabil Province, Iran

• Main Authors: Lida Andalibi, Ardavan Ghorbani, Mehdi Moameri, Zeinab Hazbavi, Arne Nothdurft, Reza Jafari, Farid Dadjou
• Format: Article
• Language: English
• Published: MDPI AG 2021-07-01
• Series: Remote Sensing
• Subjects: LaiPen; management tools; remote sensing; vegetation indices; spatiotemporal changes
• Online Access: https://www.mdpi.com/2072-4292/13/15/2879

The leaf area index (LAI) is an important vegetation biophysical index that provides broad information on the dynamic behavior of an ecosystem’s productivity and related climate, topography, and edaphic impacts. The spatiotemporal changes of LAI were assessed throughout Ardabil Province—a host of relevant plant communities within the critical ecoregion of a semi-arid climate. In a comparative study, novel data from Google Earth Engine (GEE) was tested against traditional ENVI measures to provide LAI estimations. Moreover, it is of important practical significance for institutional networks to quantitatively and accurately estimate LAI, at large areas in a short time, and using appropriate baseline vegetation indices. Therefore, LAI was characterized for ecoregions of Ardabil Province using remote sensing indices extracted from Landsat 8 Operational Land Imager (OLI), including the Enhanced Vegetation Index calculated in GEE (EVI_G) and ENVI5.3 software (EVI_E), as well as the Normalized Difference Vegetation Index estimated in ENVI5.3 software (NDVI_E). Moreover, a new field measurement method, i.e., the LaiPen LP 100 portable device (LP 100), was used to evaluate the accuracy of the derived indices. Accordingly, the LAI was measured in June and July 2020, in 822 ground points distributed in 16 different ecoregions-sub ecoregions having various plant functional types (PFTs) of the shrub, bush, and tree. The analyses revealed heterogeneous spatial and temporal variability in vegetation indices and LAIs within and between ecoregions. The mean (standard deviation) value of EVI_G, EVI_E, and NDVI_E at a province scale yielded 1.1 (0.41), 2.20 (0.78), and 3.00 (1.01), respectively in June, and 0.67 (0.37), 0.80 (0.63), and 1.88 (1.23), respectively, in July. The highest mean values of EVI_G-LAI, EVI_E-LAI, and NDVI_E-LAI in June are found in Meshginshahr (1.40), Meshginshahr (2.80), and Hir (4.33) ecoregions and in July are found in Andabil ecoregion respectively with values of 1.23, 1.5, and 3.64. The lowest mean values of EVI_G-LAI, EVI_E-LAI, and NDVI_E-LAI in June were observed for Kowsar (0.67), Meshginshahr (1.8), and Neur (2.70) ecoregions, and in July, the Bilesavar ecoregion, respectively, with values of 0.31, 0.31, and 0.81. High correlation and determination coefficients (r > 0.83 and R² > 0.68) between LP 100 and remote sensing derived LAI were observed in all three PFTs (except for NDVI_E-LAI in June with r = 0.56 and R² = 0.31). On average, all three examined LAI measures tended to underestimate compared to LP 100-LAI (r > 0.42). The findings of the present study could be promising for effective monitoring and proper management of vegetation and land use in the Ardabil Province and other similar areas.