Avian species identity and predation success in tropical cacao agroforestry of the Napu Valley in Central Sulawesi, Indonesia

Avian ecosystem services such as the suppression of pests are considered being of high ecological and economic importance in a range of ecosystems, especially in tropical agroforestry. But how bird predation success is related to the diversity and composition of the bird community, as well as local and landscape factors, is poorly understood. The author quantified arthropod predation in relation to the identity and diversity of insectivorous birds, using experimental exposure of artificial, caterpillar-like prey on smallholder cacao agroforestry systems, differing in local shade management and distance to primary forest. The bird community was assessed using both mist netting (targeting on active understory insectivores) and point count (higher completeness of species inventories) sampling. The study was conducted in a land use dominated area in Central Sulawesi, Indonesia, adjacent to the Lore Lindu National Park. We selected 15 smallholder cacao plantations as sites for bird and bat exclosure experiments in March 2010. Until July 2011, we recorded several data in this study area, including the bird community data, cacao tree data and bird predation experiments that are presented here. We found that avian predation success can be driven by single and abundant insectivorous species, rather than by overall bird species richness. Forest proximity was important for enhancing the density of this key species, but did also promote bird species richness. The availability of local shade trees had no effects on the local bird community or avian predation success. Our findings are both of economical as well as ecological interest because the conservation of nearby forest remnants will likely benefit human needs and biodiversity conservation alike.

Bird observation and sampling in tropical agroforestry landscapes of the Napu Valley in Central Sulawesi, Indonesia

We investigated the local bird community in Central Sulawesi (Indonesia), with focus on insectivorous species in the agroforestry landscapes adjacent to the Lore Lindu National Park. All study sites were situated at the northern tip of Napu Valley in Central Sulawesi, Indonesia. After an initial mapping of the study area, we selected 15 smallholder cacao plantations as sites for our study in March 2010. These sides were mainly used for bird and bat exclosure experiments. All sited were situated along a local gradient (shade availability on each plantation) and a landscape gradient (distance to primary forest), which were independent from each other. In September 2010 and from February until June 2011, we assessed the bird community on our 15 study sites using monthly point count and mist netting sampling. Point count (20 minutes between 07 am and 10 am and in between the net checking hours) and mist netting surveys (12 hours, between 05:30 am and 17:30 pm) were conducted simultaneously but only once per month on each study site, to avoid habituation of the local bird community to our surveys. Further, point counts were conducted at least 100 m apart from the mist netting sites, to avoid potential disturbance between the two methods. We discarded all observations beyond 50 m (including those individuals that flew over the canopy) from the statistical analysis, as well as recaptures of individuals within identical mist netting rounds.

Mesopelagic fish collected during Maria S. Merian cruise MSM26, spring 2013

Mesopelagic fish were collected using a 1 m**2 Double-MOCNESS (Multiple Opening and Closing Net and Environmental Sensing System) and 4.5 m**2 IKMT (Isaacs-Kidd midwater trawl). The main portion of the IKMT was 20 mm knotted nylon, and the tail bag was 3 mm knotless nylon. Oblique IKMT tows were made to a maximum depth of 500 m at a tow speed of 3.5 knots. The original cruise plan intended for nighttime IKMT tows, but tow times varied due to operational constraints. The MOCNESS was equipped with 20 nets of 333 µm mesh size; 10 nets per side. The towing speed was 2 knots. Samples were collected to a maximum depth of 1250 m. The first oblique nets sampled from the surface to the max depth, and the other nets sampled depth stratified bins of the water column. MOCNESS hauls were performed during day and night to investigate diel vertical migrations. Mesoplelagic fish were processed on board. All fish were picked from all IKMT nets, most oblique MOCNESS nets, and the left side nets of the depth stratified MOCNESS samples. The Depth stratified nets from the right side of the MOCNESS frame were preserved in 5 % formalin for future quantitative analyses of the nekton. Fish were identified to the lowest possible taxa using Whitehead et al. (1984) and Fahay (2007). Standard length of each fish was measured to the nearest 0.1 mm using a digital caliper. Measured and identified fish were frozen in an -80 °C freezer, and shipped to the University of Hamburg at the end of the cruise.

Abundance mesopelagic fish collected during Maria S. Merian cruise MSM26, spring 2013

Mesopelagic fish were collected using a 1 m**2 Double-MOCNESS (Multiple Opening and Closing Net and Environmental Sensing System) and 4.5 m**2 IKMT (Isaacs-Kidd midwater trawl). The main portion of the IKMT was 20 mm knotted nylon, and the tail bag was 3 mm knotless nylon. Oblique IKMT tows were made to a maximum depth of 500 m at a tow speed of 3.5 knots. The original cruise plan intended for nighttime IKMT tows, but tow times varied due to operational constraints. The MOCNESS was equipped with 20 nets of 333 µm mesh size; 10 nets per side. The towing speed was 2 knots. Samples were collected to a maximum depth of 1250 m. The first oblique nets sampled from the surface to the max depth, and the other nets sampled depth stratified bins of the water column. MOCNESS hauls were performed during day and night to investigate diel vertical migrations. Mesoplelagic fish were processed on board. All fish were picked from all IKMT nets, most oblique MOCNESS nets, and the left side nets of the depth stratified MOCNESS samples. The Depth stratified nets from the right side of the MOCNESS frame were preserved in 5 % formalin for future quantitative analyses of the nekton. Fish were identified to the lowest possible taxa using Whitehead et al. (1984) and Fahay (2007). Standard length of each fish was measured to the nearest 0.1 mm using a digital caliper. Measured and identified fish were frozen in an -80 °C freezer, and shipped to the University of Hamburg at the end of the cruise.