Agricultural land use efficiency and food crop production in Ghana

Despite the low productivity of the extensive agriculture system, Ghana recorded the largest reduction of undernourishment in the past two decades. We used biophysical analysis to determine the efficiency and potentials of the extensive system and its future sustainability. The results indicate that food production in Ghana has increased steadily over the past two decades and correlated highly with cropped area and population (R ² < 0.85 and 0.82), but not with fertilizer (R ² = 0.06). Sufficient food production could be sustained in the short term. In the longer term, however, the food situation in Ghana appears precarious if population growth continues while land remains the same.     

Plasticity of social organization in a forest ant species

We manipulated availability of food and nesting sites in one population of the forest ant Myrmica punctiventris. The manipulations produced significant changes in relatedness structure, reproductive allocation, and response to hierarchical selection. Food availability appeared to have a consistently stronger influence on these aspects of social organization than did availability of nesting sites. We interpret our experimental results in light of observed differences between populations, and discuss implications for kin selection dynamics. Received: 30 July 1998 / Accepted after revision: 31 October 1998     

Exposure of insects and insectivorous birds to metals and other elements from abandoned mine tailings in three Summit County drainages, Colorado

Concentrations of 31 metals, metalloids, and other elements were measured in insects and insectivorous bird tissues from three drainages with different geochemistry and mining histories in Summit Co., Colorado, in 2003, 2004, and 2005. In insect samples, all 25 elements that were analyzed in all years increased in both Snake and Deer Creeks in the mining impacted areas compared to areas above and below the mining impacted areas. This distribution of elements was predicted from known or expected sediment contamination resulting from abandoned mine tailings in those drainages. Element concentrations in avian liver tissues were in concordance with levels in insects, that is with concentrations higher in mid-drainage areas where mine tailings were present compared to both upstream and downstream locations; these differences were not always statistically different, however. The lack of statistically significant differences in liver tissues, except for a few elements, was due to relatively small sample sizes and because many of these elements are essential and therefore well regulated by the bird's homeostatic processes. Most elements were at background concentrations in avian liver tissue except for Pb which was elevated at mid-drainage sites to levels where delta-aminolevulinic acid dehydratase activity was inhibited at other mining sites in Colorado. Lead exposure, however, was not at toxic levels. Fecal samples were not a good indication of what elements birds ingested and were potentially exposed to.     

Combining functional data with hierarchical Gaussian process models

Gaussian process models have been used in applications ranging from machine learning to fisheries management. In the Bayesian framework, the Gaussian process is used as a prior for unknown functions, allowing the data to drive the relationship between inputs and outputs. In our research, we consider a scenario in which response and input data are available from several similar, but not necessarily identical, sources. When little information is known about one or more of the populations it may be advantageous to model all populations together. We present a hierarchical Gaussian process model with a structure that allows distinct features for each source as well as shared underlying characteristics. Key features and properties of the model are discussed and demonstrated in a number of simulation examples. The model is then applied to a data set consisting of three populations of Rotifer Brachionus calyciflorus Pallas. Specifically, we model the log growth rate of the populations using a combination of lagged population sizes. The various lag combinations are formally compared to obtain the best model inputs. We then formally compare the leading hierarchical Gaussian process model with the inferential results obtained under the independent Gaussian process model.