Productivity, carbon assimilation and intra-annual change in tropical reef platform seagrass communities of the Torres Strait, north-eastern Australia

Detailed data on seagrass distribution, abundance, growth rates and community structure information were collected at Orman Reefs in March 2004 to estimate the above-ground productivity and carbon assimilated by seagrass meadows. Seagrass meadows were re-examined in November 2004 for comparison at the seasonal extremes of seagrass abundance. Ten seagrass species were identified in the meadows on Orman Reefs. Extensive seagrass coverage was found in March (18,700 ha) and November (21,600 ha), with seagrass covering the majority of the intertidal reef-top areas and a large proportion of the subtidal areas examined. There were marked differences in seagrass above-ground biomass, distribution and species composition between the two surveys. Major changes between March and November included a substantial decline in biomass for intertidal meadows and an expansion in area of subtidal meadows. Changes were most likely a result of greater tidal exposure of intertidal meadows prior to November leading to desiccation and temperature-related stress. The Orman Reef seagrass meadows had a total above-ground productivity of 259.8 t DW day-1 and estimated carbon assimilation of 89.4 t C day-1 in March. The majority of this production came from the intertidal meadows which accounted for 81% of the total production. Intra-annual changes in seagrass species composition, shoot density and size of meadows measured in this study were likely to have a strong influence on the total above-ground production during the year. The net estimated above-ground productivity of Orman Reefs meadows in March 2004 (1.19 g C m-2 day-1) was high compared with other tropical seagrass areas that have been studied and also higher than many other marine, estuarine and terrestrial plant communities.

Genetic erosion and changes in distribution of sorghum (Sorghum bicolor L. (Moench)) landraces in north-eastern Ethiopia

Ethiopia is believed to be the centre of origin and domestication for sorghum, where sorghum remains one of the main staple crops. Loss of biodiversity is occurring at an alarming rate in Ethiopia and crops, including sorghum, have long been recognized as vulnerable to genetic erosion. A major collection of sorghum germplasm was made in 1973 by Gebrekidan and Ejeta from north-eastern Ethiopia. A new collection of landraces was made in 2003, and these were field evaluated at Sirinka in 2004 along with representative samples from the 1973 collection. Farmer surveys and soil and climate surveys were also performed. Preliminary analysis demonstrated that some important landraces have disappeared either locally or regionally in the past 30 years and many other landraces have become marginalized. Landraces which are less preferred in terms of agronomic value and end use, and introductions, have become increasingly important. Late maturing landraces were found to be particularly vulnerable, with a number disappearing altogether. Farmers have become more risk averse, and factors such as declining soil fertility, more frequent drought and unreliable rainfall, and increased pest infestation have contributed to a change in farmer landrace selection. Data are presented on the variability and unique characters of some of the Ethiopian landraces, and implications for conservation are discussed.

Projecting future expansion of invasive species: Comparing and improving methodologies for species distribution modeling

Modeling the distributions of species, especially of invasive species in non-native ranges, involves multiple challenges. Here, we developed some novel approaches to species distribution modeling aimed at reducing the influences of such challenges and improving the realism of projections. We estimated species-environment relationships with four modeling methods run with multiple scenarios of (1) sources of occurrences and geographically isolated background ranges for absences, (2) approaches to drawing background (absence) points, and (3) alternate sets of predictor variables. We further tested various quantitative metrics of model evaluation against biological insight. Model projections were very sensitive to the choice of training dataset. Model accuracy was much improved by using a global dataset for model training, rather than restricting data input to the species’ native range. AUC score was a poor metric for model evaluation and, if used alone, was not a useful criterion for assessing model performance. Projections away from the sampled space (i.e. into areas of potential future invasion) were very different depending on the modeling methods used, raising questions about the reliability of ensemble projections. Generalized linear models gave very unrealistic projections far away from the training region. Models that efficiently fit the dominant pattern, but exclude highly local patterns in the dataset and capture interactions as they appear in data (e.g. boosted regression trees), improved generalization of the models. Biological knowledge of the species and its distribution was important in refining choices about the best set of projections. A post-hoc test conducted on a new Partenium dataset from Nepal validated excellent predictive performance of our “best” model. We showed that vast stretches of currently uninvaded geographic areas on multiple continents harbor highly suitable habitats for Parthenium hysterophorus L. (Asteraceae; parthenium). However, discrepancies between model predictions and parthenium invasion in Australia indicate successful management for this globally significant weed. This article is protected by copyright. All rights reserved.

Projecting future expansion of invasive species: Comparing and improving methodologies for species distribution modeling

Modeling the distributions of species, especially of invasive species in non-native ranges, involves multiple challenges. Here, we developed some novel approaches to species distribution modeling aimed at reducing the influences of such challenges and improving the realism of projections. We estimated species-environment relationships with four modeling methods run with multiple scenarios of (1) sources of occurrences and geographically isolated background ranges for absences, (2) approaches to drawing background (absence) points, and (3) alternate sets of predictor variables. We further tested various quantitative metrics of model evaluation against biological insight. Model projections were very sensitive to the choice of training dataset. Model accuracy was much improved by using a global dataset for model training, rather than restricting data input to the species’ native range. AUC score was a poor metric for model evaluation and, if used alone, was not a useful criterion for assessing model performance. Projections away from the sampled space (i.e. into areas of potential future invasion) were very different depending on the modeling methods used, raising questions about the reliability of ensemble projections. Generalized linear models gave very unrealistic projections far away from the training region. Models that efficiently fit the dominant pattern, but exclude highly local patterns in the dataset and capture interactions as they appear in data (e.g. boosted regression trees), improved generalization of the models. Biological knowledge of the species and its distribution was important in refining choices about the best set of projections. A post-hoc test conducted on a new Partenium dataset from Nepal validated excellent predictive performance of our “best” model. We showed that vast stretches of currently uninvaded geographic areas on multiple continents harbor highly suitable habitats for Parthenium hysterophorus L. (Asteraceae; parthenium). However, discrepancies between model predictions and parthenium invasion in Australia indicate successful management for this globally significant weed. This article is protected by copyright. All rights reserved.