Main content area

Long term and seasonal courses of leaf area index in a semi-arid forest plantation

Sprintsin, Michael, Cohen, S., Maseyk, K., Rotenberg, E., Grünzweig, J., Karnieli, A., Berliner, P., Yakir, D.
Agricultural and forest meteorology 2011 v.151 no.5 pp. 565-574
Pinus halepensis, canopy, deserts, environment, forest management, forest plantations, forest stands, forests, leaf area index, models, monitoring, prediction, remote sensing, seasons, semiarid zones, sensors, woodlands, Israel
Effective leaf area index (LAIₑ) in the semi-arid Pinus halepensis plantation, located between arid and semi-arid climatic zones at the edge of the Negev and Judean deserts, was measured bi-annually during four years (2001–2004) and more intensively (monthly) during the following two years (2004–2006) by a number of non-contact optical devices. The measurements showed a gradual increase in LAIₑ from ∼1 (±0.25) to ∼1.8 (±0.11) during these years. All instruments, when used properly, gave similar results that were also comparable with actual leaf area index measured by litter collection and destructive sampling and allometric estimates. Because of the constraint of clear sky conditions, which limited the use of the fisheye type sensors to times of twilight, the fisheye techniques were less useful. The tracing radiation and architecture of canopies system, which includes specific treatment of two levels of clumpiness of the sparse forest stand, was used successfully for the intensive monitoring. The mean clumpiness index, 0.61, is considered representative for the specific environment. Finally, the LAIₑ measurements at the start of each season were used to constrain phenology-based estimates of annual LAIₑ development, resulting in a continuous course of LAIₑ in the forest over the five-year period. Intra-seasonal LAIₑ variation in the order of 10% of total LAIₑ predicted by the model was also observed in the intensive TRAC measurements, giving confidence in the TRAC system and indicating its sensitivity and applicability in woodlands even with low LAIₑ values. The results can be important for forest management decision support as well as for use in evaluation of remote sensing techniques for forests at the lowest range of LAIₑ values.