Regulation works in the Upper Rhine carried out since the 19th century have drastically modified the floodplain forest ecosystem in the Alsace Plain. These works have disconnected the largest part of the riparian forest from the river and interrupted the flooding. This study aims at assessing the effects of the flood interruption on woody regeneration in three hardwood stands: one still liable to flooding, the second one unflooded for 30 years and the third one unflooded for 130 years. Fieldwork was carried out to determine the species richness of seed rain, seed bank and extant vegetation of the regeneration understorey. The potential species richness was calculated using the first-order jackknife estimator. In addition, floristic composition and diversity indices of extant vegetation were calculated. Results show that species richness of the seed bank is lower than seed rain and extant vegetation. The species richness of the extant vegetation increases with the duration of interruption of the floods in the Rhine forest as a result of introduction of flood-intolerant species in the unflooded forest. Although species richness of the extant vegetation was the lowest in the flooded site, Shannon -Weaver diversity and equitability were found to be the highest, because the disturbance caused by floods allows a more equal coexistence of several species. Conversely, in the unflooded sites, a small number of species represent most of the woody individuals of the regeneration phase. Implications of flood restoration on species richness and diversity of the regeneration compartment are discussed.

No abstract

Hemispherical photography is becoming a popular technique for gap fraction measurements to characterize biophysical parameters and solar radiation in plant canopies. One of the crucial steps in the measurement of canopy gap fraction using hemispherical photography is determining the resolution of the sampling grid. In this work, the effects of varying resolutions of sampling grids by modifying the angle widths of zenithal annuli and azimuthal sectors were evaluated for leaf area and clumping indices computations. Sensitivity analysis was performed to test these effects using artificial photographs simulating ideal canopies with varying leaf area index and aggregation levels of foliage elements. Contrasting forest types, including natural tropical cloud forest and exotic plantations, were tested as real canopies. Results indicate that leaf area and clumping indices estimates are significantly affected by the variation of sampling grids. A new approach to solve the problem of null-gap segments, obscured completely by foliage, is proposed. However, the determination of optimal combinations of zenithal annuli and azimuthal sector angular widths that suit all canopy types remains a difficult practical problem that is often overlooked. Finally, theoretically sound gap fraction and size sampling regions were demonstrated for reliable estimates of canopy biophysical parameters.

No abstract