No abstract

Shade trees play an important role within agroforestry systems by influencing radiation and wind regimes as well as nutrient and hydrological cycling. However, there is a lack of quantitative assessments of their functions. One of the reasons is the rare information on structural characteristics of shade tree species. Therefore, the aim of this study is to provide basic information on the structure of frequently used shade tree species for the implementation of models simulating the ecosystem processes in agroforestry systems. The investigation of the shade trees was conducted at two cacao agroforestry sites on Sulawesi, Indonesia. The measurements of the main structural parameters: diameter at breast height, tree height, trunk height, crown length and crown radius were carried out for the shade tree species Aleurites moluccana, Cocos nucifera and Gliricidia sepium. For data collection, the National Forest Inventory Field Manual Template by FAO (2004) was applied. Based on this information allometric functions were derived for the correspondent shade tree species. The best significant relationships were obtained for the heightcrown length relationship of the dicotyledonous tree species' A. moluccana and G. sepium with a coefficient of determination r 2 = 0.925 and r 2 = 0.738, respectively, and the height-crown length relationship of the monocotyledonous palm C. nucifera with r 2 = 0.663. The transferability tests 'analysis of covariance' and 'homogeneity of slopes' have shown that the obtained allometric functions are also applicable to other cacao agroforestry systems of the region.

The increasing animosity towards wolves (Canislupus) by livestock-keeping nomads in Mongolia and the accompanying conflicts highlight the urgent need for knowledge about the feeding behavior of wolves, since information on the feeding ecology of wolves in Mongolia is rare, especially in the mountain taiga and mountain forest steppe regions of Northern Mongolia. Those regions are characterized by a relatively high wildlife diversity and are sparsely populated by humans. To face this problem, 137 wolf scats were collected in the Khentii Mountain range in Northern Mongolia between 2008 and 2012. Almost all wolf faeces contained remnants of wild ungulates, which made up 89% of the consumed biomass. Siberian roe deer (Capreoluspygargus) was the most important and positively selected prey species. It was followed by red deer (Cervuselaphus) and wild boar (Susscrofa), which was negatively selected by wolves. Wolves also fed on buffer prey species such as lagomorphs and small mammals. No evidence of domestic ungulates was found in the wolf diet. Thus, near-natural habitats with a diverse fauna of wild animals are important to limit livestock depredation.

Climate change has distinct regional and local differences in its impacts on the land surface. One of the important parameters determining the climate change signal is the emissivity (ε) of the surface. In forest-climate interactions, the leaf surface emissivity plays a decisive role. The accurate determination of leaf emissivities is crucial for the appropriate interpretation of measured energy and matter fluxes between the forest and the atmosphere. In this study, we quantified the emissivity of the five broadleaf tree species Acer pseudoplatanus, Fagus sylvatica, Fraxinus excelsior, Populus simonii and Populus candicans. Measurements of leaf surface temperatures were conducted under laboratory conditions in a controlled-climate chamber within the temperature range of +8 °C and +32 °C. Based on these measurements, broadband leaf emissivities ε (ε for the spectral range of 8-14 µm) were calculated. Average ε 8-14 µm was 0.958±0.002 for all species with very little variation among species. In a second step, the soil-vegetation-atmosphere transfer model 'MixFor-SVAT' was applied to examine the effects of ε changes on radiative, sensible and latent energy fluxes of the Hainich forest in Central Germany. Model experiments were driven by meteorological data measured at the Hainich site. The simulations were forced with the calculated ε value as well as with minimum and maximum values obtained from the literature. Significant effects of ε changes were detected. The strongest effect was identified for the sensible heat flux with a sensitivity of 20.7 % per 1 % ε change. Thus, the variability of ε should be considered in climate change studies.