Ela Zangy
Jaime Kigel
Shabtai Cohen
Yossef Moshe
Mor Ashkenazi|
Ori Fragman-Sapir
Yagil Osem
Mediterranean forest ecosystems are of high conservation value and therefore increasingly managed to restore and promote their natural biodiversity and complexity. The present study aimed to assess plant species diversity of East-Mediterranean conifer forests as affected by overstory cover manipulation at multiple spatial scales. Understanding this interaction in water-limited forest ecosystems is particularly challenging as overstory-understory relations involve both water and light constrains. The experiment was conducted in a mature (40y) planted Pinus halepensis forest in the Jerusalem Mountains of Israel. Understory plant community structure was monitored for five years in sixteen experimental plots (0.5 ha) representing a range of overstory cover levels (leaf area index-LAI = 0–4 m2m−2) created by tree thinning. Light and water availability in the forest understory were monitored using fisheye camera (irradiance), predawn leaf water potential (PLWP) of dominant species and soil water content (SWC) measurements. Plant diversity was measured at increasing spatial scales (grain sizes, 0.01–400 m2), using hierarchical nested sampling. Irradiance and SWC increased with decreasing LAI while PLWP was not affected. Stand-level understory richness (α diversity) increased linearly with decreasing LAI, indicating a consistent, positive resource availability-diversity relationship. The results pointed towards strong dependency of species richness on light availability as compared to its dependency on water availability that was less definite. Understory species composition varied across LAI levels (β diversity at the forest level) indicating species turnover, mainly between annual herbs benefiting from reduced overstory cover vs. woody species benefiting from higher cover. The relationships between overstory cover and understory richness and composition were rank-invariant across grain sizes. These relationships were stronger at the larger (10, 100, 400 m2) compared to the smaller (0.01, 0.1, 1 m2) grain sizes. The 10-m2 grain-size, corresponding to the crown projection area of individual trees, showed the strongest relationship. Observed patterns of overstory cover–understory richness suggest that resource availability, rather than spatial resource heterogeneity, was the main factor shaping understory diversity. Furthermore, grain-size effects on this relationship support “environmental filtering” rather than “resource density” as the main mechanism through which reduced overstory cover promoted understory diversity.
'Synthesis and applications’: Our findings demonstrate the importance of overstory cover as a key factor determining plant diversity in water-limited East-Mediterranean forests. We highlight the important role of spatial heterogeneity of overstory cover at both the stand (intra-plot) and forest (inter-plot) scales and support dynamic variable thinning as a management strategy to enhance forest biodiversity and complexity.
Ela Zangy
Jaime Kigel
Shabtai Cohen
Yossef Moshe
Mor Ashkenazi|
Ori Fragman-Sapir
Yagil Osem
Mediterranean forest ecosystems are of high conservation value and therefore increasingly managed to restore and promote their natural biodiversity and complexity. The present study aimed to assess plant species diversity of East-Mediterranean conifer forests as affected by overstory cover manipulation at multiple spatial scales. Understanding this interaction in water-limited forest ecosystems is particularly challenging as overstory-understory relations involve both water and light constrains. The experiment was conducted in a mature (40y) planted Pinus halepensis forest in the Jerusalem Mountains of Israel. Understory plant community structure was monitored for five years in sixteen experimental plots (0.5 ha) representing a range of overstory cover levels (leaf area index-LAI = 0–4 m2m−2) created by tree thinning. Light and water availability in the forest understory were monitored using fisheye camera (irradiance), predawn leaf water potential (PLWP) of dominant species and soil water content (SWC) measurements. Plant diversity was measured at increasing spatial scales (grain sizes, 0.01–400 m2), using hierarchical nested sampling. Irradiance and SWC increased with decreasing LAI while PLWP was not affected. Stand-level understory richness (α diversity) increased linearly with decreasing LAI, indicating a consistent, positive resource availability-diversity relationship. The results pointed towards strong dependency of species richness on light availability as compared to its dependency on water availability that was less definite. Understory species composition varied across LAI levels (β diversity at the forest level) indicating species turnover, mainly between annual herbs benefiting from reduced overstory cover vs. woody species benefiting from higher cover. The relationships between overstory cover and understory richness and composition were rank-invariant across grain sizes. These relationships were stronger at the larger (10, 100, 400 m2) compared to the smaller (0.01, 0.1, 1 m2) grain sizes. The 10-m2 grain-size, corresponding to the crown projection area of individual trees, showed the strongest relationship. Observed patterns of overstory cover–understory richness suggest that resource availability, rather than spatial resource heterogeneity, was the main factor shaping understory diversity. Furthermore, grain-size effects on this relationship support “environmental filtering” rather than “resource density” as the main mechanism through which reduced overstory cover promoted understory diversity.
'Synthesis and applications’: Our findings demonstrate the importance of overstory cover as a key factor determining plant diversity in water-limited East-Mediterranean forests. We highlight the important role of spatial heterogeneity of overstory cover at both the stand (intra-plot) and forest (inter-plot) scales and support dynamic variable thinning as a management strategy to enhance forest biodiversity and complexity.