Ostracode relative abundances and geochemistry in surface sediments from Lake Ohrid

Understanding the ecology of bioindicators such as ostracods is essential in order to reconstruct past environmental and climate change from analysis of fossil assemblages preserved in lake sediment cores. Knowledge of the ecology of ancient Lake Ohrid's ostracod fauna is very limited and open to debate. In advance of the Ohrid ICDP-Drilling project, which has potential to generate high-resolution long-term paleoenvironmental data of global importance in paleoclimate research, we sampled Lake Ohrid and a wide range of habitat types in its surroundings to assess 1) the composition of ostracod assemblages in lakes, springs, streams, and short-lived seasonal water bodies, 2) the geographical distribution of ostracods, and 3) the ecological characteristics of individual ostracod species. In total, 40 species were collected alive, and seven species were preserved as valves and empty carapaces. Of the 40 ostracod species, twelve were endemic to Lake Ohrid. The most common genus in the lake was Candona, represented by 13 living species, followed by Paralimnocythere, represented by five living species. The most frequent species was Cypria obliqua. Species with distinct distributions included Heterocypris incongruens, Candonopsis kingsleii, and Cypria lacustris. The most common species in shallow, flooded areas was H. incongruens, and the most prominent species in ditches was C. kingsleii. C. lacustris was widely distributed in channels, springs, lakes, and rivers. Statistical analyses were performed on a "Lake Ohrid" dataset, comprising the subset of samples from Lake Ohrid alone, and an "entire" dataset comprising all samples collected. The unweighted pair group mean average (UPGMA) clustering was mainly controlled by species-specific depth preferences. Canonical Correspondence Analysis (CCA) with forward selection identified water depth, water temperature, and pH as variables that best explained the ostracod distribution in Lake Ohrid. The lack of significance of conductivity and dissolved oxygen in CCA of Ohrid data highlight the uniformity across the lake of the well-mixed waters. In the entire area, CCA revealed that ostracod distribution was best explained by water depth, salinity, conductivity, pH, and dissolved oxygen. Salinity was probably selected by CCA due to the presence of Eucypris virens and Bradleystrandesia reticulata in short-lived seasonal water bodies. Water depth is an important, although indirect, influence on ostracod species distribution which is probably associated with other factors such as sediment texture and food supply. Some species appeared to be indicators for multiple environmental variables, such as lake level and water temperature.

delta Deuterium of alkenone and BIT index of sediment core 64PE304-80 samples

Reconstructing past ocean salinity is important for assessing paleoceanographic change and therefore past climatic dynamics. Commonly, sea water salinity reconstruction is based on foraminifera oxygen isotope ratio values combined with sea surface temperature reconstruction. However, the approach relies on multiple proxies, resulting in relatively large uncertainty and, consequently, relatively low accuracy of salinity estimates. An alternative tool for past ocean salinity reconstruction is the hydrogen isotope composition of long chain (C37) alkenones (dDalkenone). Here, we applied dDalkenone to a 39 ka long coastal sediment record from the Eastern South African continental shelf in the Mozambique Channel, close to the Zambezi River mouth. Despite changes in global sea water dD related to glacial - interglacial ice volume effects, no clear changes were observed in the dDalkenone record throughout the entire 39 ka. The BIT index record from the same core showed high BIT values during the glacial and low values during the Holocene. This indicates a more pronounced freshwater influence at the core location during the glacial, resulting in alkenones depleted in deuterium during that time and, thereby, explains the lack of a clear glacial-interglacial alkenone dD shift. Correlation between the BIT index and dDalkenone during the glacial period suggests that increased continental runoff potentially changed the growth conditions of the alkenone producing haptophytes, promoting coastal haptophyte species with generally more enriched dDalkenone values. We therefore suggest that the application of dDalkenone for reconstructing past salinity in coastal settings may be complicated by changes in the alkenone producing haptophyte community.

(Table 1) Iron and manganese speciations in surface layer bottom sediments of the Pacific Ocean

Distribution of iron and manganese speciations in ocean sediments of a section from the coast of Japan to the open Pacific Ocean is under consideration. Determinations of total iron, as well as of reactive iron contents and of total manganese, as well as of Mn4+ contents have been done. Significant increase of total Fe content in sediments from the coast to the pelagic zone occurs without noticeable increase in reactive Fe content. Presence of layers of volcanic and terrigenous coarse clastic material in clayey sediments results to sharp change in iron content. Manganese content increases from near coastal to pelagic sediments more than 10 times; oxidation degree of sediments also increases. There are three types of bottom sediments different by contents of iron and manganese forms: reduced, oxidized (red clay), and transitional. Content of total Fe is almost does not change with depth in sediments, content of reactive Fe increases in reduced sediments, and decreases in oxidized ones. Manganese content in red clay mass increases several times.

Leader's reflections on knowledge, knowing and not knowing: an analysis of change over time

One commonality across the leadership and knowledge related literature is the apparent neglect of the leaders own knowledge. This thesis sought to address this issue through conducting exploratory research into the content of leader’s personal knowledge and the process of knowing it. The empirical inquiry adopted a longitudinal approach, with interviews conducted at two separate time periods with an extended time-interval between each. The findings from this research contrast with images of leadership which suggest leaders are in control of what they know, that they own their own knowledge. The picture that emerges is one of individuals struggling to keep abreast of the knowledge required to deal with the dynamics and uncertainties of organisational life. Much knowledge is tacit, provisional and perishable and the related process of knowing more organic, evolutionary and informal than any structured or orchestrated approach. The collective nature of knowing is a central feature, with these leaders embedded in networks of uncontrollable relationships. In view of the indeterminate nature of knowing, the boundary between what is known and what one needs to know is both amorphous and ephemeral, and the likelihood of knowledge-absences is escalated. A significant finding in this regard is the identification of two critical points where not-knowing is most likely (entry and exit from role) and the differing implications of each. Overtime the knowledge that is legitimised or prioritised is significantly altered as these leaders replace the dogmas that were previously held in high esteem with the lessons from their own experience. This experience brings increased self-knowledge and a deeper appreciation of the values and morals instilled in their early lives. In view of the above findings, this study makes theoretical contribution to a number of core literatures: authentic leadership, role transition and knowledge-absences. In terms of leadership development, the findings point to the necessity to prepare leaders for the challenges they will encounter at the pivotal stages of the leadership role.

Learning from Geometry

Subspaces and manifolds are two powerful models for high dimensional signals. Subspaces model linear correlation and are a good fit to signals generated by physical systems, such as frontal images of human faces and multiple sources impinging at an antenna array. Manifolds model sources that are not linearly correlated, but where signals are determined by a small number of parameters. Examples are images of human faces under different poses or expressions, and handwritten digits with varying styles. However, there will always be some degree of model mismatch between the subspace or manifold model and the true statistics of the source. This dissertation exploits subspace and manifold models as prior information in various signal processing and machine learning tasks.

A near-low-rank Gaussian mixture model measures proximity to a union of linear or affine subspaces. This simple model can effectively capture the signal distribution when each class is near a subspace. This dissertation studies how the pairwise geometry between these subspaces affects classification performance. When model mismatch is vanishingly small, the probability of misclassification is determined by the product of the sines of the principal angles between subspaces. When the model mismatch is more significant, the probability of misclassification is determined by the sum of the squares of the sines of the principal angles. Reliability of classification is derived in terms of the distribution of signal energy across principal vectors. Larger principal angles lead to smaller classification error, motivating a linear transform that optimizes principal angles. This linear transformation, termed TRAIT, also preserves some specific features in each class, being complementary to a recently developed Low Rank Transform (LRT). Moreover, when the model mismatch is more significant, TRAIT shows superior performance compared to LRT.

The manifold model enforces a constraint on the freedom of data variation. Learning features that are robust to data variation is very important, especially when the size of the training set is small. A learning machine with large numbers of parameters, e.g., deep neural network, can well describe a very complicated data distribution. However, it is also more likely to be sensitive to small perturbations of the data, and to suffer from suffer from degraded performance when generalizing to unseen (test) data.

From the perspective of complexity of function classes, such a learning machine has a huge capacity (complexity), which tends to overfit. The manifold model provides us with a way of regularizing the learning machine, so as to reduce the generalization error, therefore mitigate overfiting. Two different overfiting-preventing approaches are proposed, one from the perspective of data variation, the other from capacity/complexity control. In the first approach, the learning machine is encouraged to make decisions that vary smoothly for data points in local neighborhoods on the manifold. In the second approach, a graph adjacency matrix is derived for the manifold, and the learned features are encouraged to be aligned with the principal components of this adjacency matrix. Experimental results on benchmark datasets are demonstrated, showing an obvious advantage of the proposed approaches when the training set is small.

Stochastic optimization makes it possible to track a slowly varying subspace underlying streaming data. By approximating local neighborhoods using affine subspaces, a slowly varying manifold can be efficiently tracked as well, even with corrupted and noisy data. The more the local neighborhoods, the better the approximation, but the higher the computational complexity. A multiscale approximation scheme is proposed, where the local approximating subspaces are organized in a tree structure. Splitting and merging of the tree nodes then allows efficient control of the number of neighbourhoods. Deviation (of each datum) from the learned model is estimated, yielding a series of statistics for anomaly detection. This framework extends the classical {\em changepoint detection} technique, which only works for one dimensional signals. Simulations and experiments highlight the robustness and efficacy of the proposed approach in detecting an abrupt change in an otherwise slowly varying low-dimensional manifold.

Selection of a network of large lakes and reservoirs suitable for global environmental change analysis using Earth Observation

The GloboLakes project, a global observatory of lake responses to environmental change, aims to exploit current satellite missions and long remote-sensing archives to synoptically study multiple lake ecosystems, assess their current condition, reconstruct past trends to system trajectories, and assess lake sensitivity to multiple drivers of change. Here we describe the selection protocol for including lakes in the global observatory based upon remote-sensing techniques and an initial pool of the largest 3721 lakes and reservoirs in the world, as listed in the Global Lakes and Wetlands Database. An 18-year-long archive of satellite data was used to create spatial and temporal filters for the identification of waterbodies that are appropriate for remote-sensing methods. Further criteria were applied and tested to ensure the candidate sites span a wide range of ecological settings and characteristics; a total 960 lakes, lagoons, and reservoirs were selected. The methodology proposed here is applicable to new generation satellites, such as the European Space Agency Sentinel-series.

Barite accumulation rates throughout Pliocene-Pleistocene in Eastern Equatorial Pacific at ODP Site 138-849

Export production is an important component of the carbon cycle, modulating the climate system by transferring CO2 from the atmosphere to the deep ocean via the biological pump. Here we use barite accumulation rates to reconstruct export production in the eastern equatorial Pacific over the past 4.3 Ma. We find that export production fluctuated considerably on multiple time scales. Export production was on average higher (51 g C/m**2/yr) during the Pliocene than the Pleistocene (40 g C/m**2/yr), decreasing between 3 and 1 Ma (from more than 60 to 20 g C/m**2/yr) followed by an increase over the last million years. These trends likely reflect basin-scale changes in nutrient inventory and ocean circulation. Our record reveals decoupling between export production and temperatures on these long (million years) time scale. On orbital time scales, export production was generally higher during cold periods (glacial maxima) between 4.3 and 1.1 Ma. This could be due to stronger wind stress and higher upwelling rates during glacial periods. A shift in the timing of maximum export production to deglaciations is seen in the last ~1.1 million years. Results from this study suggest that, in the eastern equatorial Pacific, mechanisms that affect nutrient supply and/or ecosystem structure and in turn carbon export on orbital time scales differ from those operating on longer time scales and that processes linking export production and climate-modulated oceanic conditions changed about 1.1 million years ago. These observations should be accounted for in climate models to ensure better predictions of future climate change.

Seawater carbonate chemistry, and Lithophyllum sp. and Feldmannia spp. coverage and dry weight during experiments, 2009

Climate-driven change represents the cumulative effect of global through local-scale conditions, and understanding their manifestation at local scales can empower local management. Change in the dominance of habitats is often the product of local nutrient pollution that occurs at relatively local scales (i.e. catchment scale), a critical scale of management at which global impacts will manifest. We tested whether forecasted global-scale change [elevated carbon dioxide (CO2) and subsequent ocean acidification] and local stressors (elevated nutrients) can combine to accelerate the expansion of filamentous turfs at the expense of calcifying algae (kelp understorey). Our results not only support this model of future change, but also highlight the synergistic effects of future CO2 and nutrient concentrations on the abundance of turfs. These results suggest that global and local stressors need to be assessed in meaningful combinations so that the anticipated effects of climate change do not create the false impression that, however complex, climate change will produce smaller effects than reality. These findings empower local managers because they show that policies of reducing local stressors (e.g. nutrient pollution) can reduce the effects of global stressors not under their governance (e.g. ocean acidification). The connection between research and government policy provides an example whereby knowledge (and decision making) across local through global scales provides solutions to some of the most vexing challenges for attaining social goals of sustainability, biological conservation and economic development.

Seawater carbonate chemistry, pigments and biological processes during experiments with coralline alga Corallina sessilis

Previous studies have shown that increasing atmospheric CO2 concentrations affect calcification in some planktonic and macroalgal calcifiers due to the changed carbonate chemistry of seawater. However, little is known regarding how calcifying algae respond to solar UV radiation (UVR, UVA+UVB, 280-400 nm). UVR may act synergistically, antagonistically or independently with ocean acidification (high CO2/low pH of seawater) to affect their calcification processes. We cultured the articulated coralline alga Corallina sessilis Yendo at 380 ppmv (low) and 1000 ppmv (high) CO2 levels while exposing the alga to solar radiation treatments with or without UVR. The presence of UVR inhibited the growth, photosynthetic O2evolution and calcification rates by13%, 6% and 3% in the low and by 47%, 20% and 8% in the high CO2 concentrations, respectively, reflecting a synergistic effect of CO2 enrichment with UVR. UVR induced significant decline of pH in the CO2-enriched cultures. The contents of key photosynthetic pigments, chlorophyll a and phycobiliproteins decreased, while UV-absorptivity increased under the highpCO2/low pH condition. Nevertheless, UV-induced inhibition of photosynthesis increased when the ratio of particulate inorganic carbon/particulate organic carbon decreased under the influence of CO2-acidified seawater, suggesting that the calcified layer played a UV-protective role. Both UVA and UVB negatively impacted photosynthesis and calcification, but the inhibition caused by UVB was about 2.5-2.6 times that caused by UVA. The results imply that coralline algae suffer from more damage caused by UVB as they calcify less and less with progressing ocean acidification.

Projections for Land Use Change in Brazil:2000-2050, links to files in ArcGIS shapefile format

This data set contains the inputs and the results of the REDD+ Policy Assessment Centre project (REDD-PAC) project (http://www.redd-pac.org), developed by a consortium of research institutes (IIASA, INPE, IPEA, UNEP-WCMC), supported by Germany's International Climate Initiative. Taking a new land use map of Brazil for 2000 as input, the research team used the global economic model GLOBIOM to project land use changes in Brazil up to 2050. Model projections show that Brazil has the potential to balance its goals of protecting the environment and becoming a major global producer of food and biofuels. The model results were taken into account by Brazilian decision-makers when developing the country's intended nationally determined contribution (INDC).