Web search activity data accurately predict population chronic disease risk in the USA

BACKGROUND: The WHO framework for non-communicable disease (NCD) describes risks and outcomes comprising the majority of the global burden of disease. These factors are complex and interact at biological, behavioural, environmental and policy levels presenting challenges for population monitoring and intervention evaluation. This paper explores the utility of machine learning methods applied to population-level web search activity behaviour as a proxy for chronic disease risk factors. METHODS: Web activity output for each element of the WHO's Causes of NCD framework was used as a basis for identifying relevant web search activity from 2004 to 2013 for the USA. Multiple linear regression models with regularisation were used to generate predictive algorithms, mapping web search activity to Centers for Disease Control and Prevention (CDC) measured risk factor/disease prevalence. Predictions for subsequent target years not included in the model derivation were tested against CDC data from population surveys using Pearson correlation and Spearman's r. RESULTS: For 2011 and 2012, predicted prevalence was very strongly correlated with measured risk data ranging from fruits and vegetables consumed (r=0.81; 95% CI 0.68 to 0.89) to alcohol consumption (r=0.96; 95% CI 0.93 to 0.98). Mean difference between predicted and measured differences by State ranged from 0.03 to 2.16. Spearman's r for state-wise predicted versus measured prevalence varied from 0.82 to 0.93. CONCLUSIONS: The high predictive validity of web search activity for NCD risk has potential to provide real-time information on population risk during policy implementation and other population-level NCD prevention efforts.

Rhodolith Beds Are Major CaCO3 Bio-Factories in the Tropical South West Atlantic

Rhodoliths are nodules of non-geniculate coralline algae that occur in shallow waters (<150 m depth) subjected to episodic disturbance. Rhodolith beds stand with kelp beds, seagrass meadows, and coralline algal reefs as one of the world's four largest macrophyte-dominated benthic communities. Geographic distribution of rhodolith beds is discontinuous, with large concentrations off Japan, Australia and the Gulf of California, as well as in the Mediterranean, North Atlantic, eastern Caribbean and Brazil. Although there are major gaps in terms of seabed habitat mapping, the largest rhodolith beds are purported to occur off Brazil, where these communities are recorded across a wide latitudinal range (2 degrees N - 27 degrees S). To quantify their extent, we carried out an inter-reefal seabed habitat survey on the Abrolhos Shelf (16 degrees 50' - 19 degrees 45'S) off eastern Brazil, and confirmed the most expansive and contiguous rhodolith bed in the world, covering about 20,900 km(2). Distribution, extent, composition and structure of this bed were assessed with side scan sonar, remotely operated vehicles, and SCUBA. The mean rate of CaCO3 production was estimated from in situ growth assays at 1.07 kg m(-2) yr(-1), with a total production rate of 0.025 Gt yr(-1), comparable to those of the world's largest biogenic CaCO3 deposits. These gigantic rhodolith beds, of areal extent equivalent to the Great Barrier Reef, Australia, are a critical, yet poorly understood component of the tropical South Atlantic Ocean. Based on the relatively high vulnerability of coralline algae to ocean acidification, these beds are likely to experience a profound restructuring in the coming decades.

Macrofauna abundance and sediment characteristics during R/V Senckenberg cruise west of Sylt in september 2010

The continuously influence of human impacts on the seafloor and benthic habitats demands the knowledge of clearly defined habitats to assess recent conditions and to monitor future changes. In this study, a benthic habitat dominated by sorted bedforms was mapped in 2010 using biological, sedimentological and acoustic data. This approach reveals the first interdisciplinary analysis of macrofauna communities in sorted bedforms in the German Bight. The study area covered 4 km², and was located ca. 3.5 km west of island of Sylt. Sorted bedforms formed as sinuous depressions with an east west orientation. Inside these depressions coarse sand covers the seafloor, while outside predominantly fine to medium sand was found. Based on the hydroacoustic data, two seafloor classes were identified. Acoustic class 1 was linked to coarse sand (type A) found inside these sorted bedforms, whereas acoustic class 2 was related to mainly fine to medium sands (type B). The two acoustic classes and sediment types corresponded with the macrofauna communities 1 and 2. The Aoinides paucibranchiata-Goniadella bobretzkii community on coarse sand and the Spiophanes bombyx - Magelona johnstonii community on fine sand. A transitional community 3 (Scoloplos armiger - Ophelia community), with species found in communities 1 and 2, could not be detected by hydroacoustic methods. This study showed the limits of the used acoustic methods, which were unable to detect insignificant differences in the fauna composition of sandy areas.

Occurrence (presence-only) of living Lophelia pertusa reefs in the Irish continental margin

The present data set was used as a training set for a Habitat Suitability Model. It contains occurrence (presence-only) of living Lophelia pertusa reefs in the Irish continental margin, which were assembled from databases, cruise reports and publications. A total of 4423 records were inspected and quality assessed to ensure that they (1) represented confirmed living L. pertusa reefs (so excluding 2900 records of dead and isolated coral colony records); (2) were derived from sampling equipment that allows for accurate (<200 m) geo-referencing (so excluding 620 records derived mainly from trawling and dredging activities); and (3) were not duplicated. A total of 245 occurrences were retained for the analysis. Coral observations are highly clustered in regions targeted by research expeditions, which might lead to falsely inflated model evaluation measures (Veloz, 2009). Therefore, we coarsened the distribution data by deleting all but one record within grid cells of 0.02° resolution (Davies & Guinotte 2011). The remaining 53 points were subject to a spatial cross-validation process: a random presence point was chosen, grouped with its 12 closest neighbour presence points based on Euclidean distance and withheld from model training. This process was repeated for all records, resulting in 53 replicates of spatially non-overlapping sets of test (n=13) and training (n=40) data. The final 53 occurrence records were used for model training.

Lithofacies distribution in the western Bay of Plenty, New Zealand

Sediment dynamics on a storm-dominated shelf (western Bay of Plenty, New Zealand) were mapped and analyzed using the newly developed multi-sensor benthic profiler MARUM NERIDIS III. An area of 60 km × 7 km between 2 and 35 m water depth was surveyed with this bottom-towed sled equipped with a high-resolution camera for continuous close-up seafloor photography and a CTD with connected turbidity sensor. Here we introduce our approach of using this multi-parameter dataset combined with sidescan sonography and sedimentological analyses to create detailed lithofacies and bedform distribution maps and to derive regional sediment transport patterns. For the assessment of sediment distribution, photographs were classified and their spatial distribution mapped out according to associated acoustic backscatter from a sidescan sonar. This provisional map was used to choose target locations for surficial sediment sampling and subsequent laboratory analysis of grain size distribution and mineralogical composition. Finally, photographic, granulometric and mineralogical facies were combined into a unified lithofacies map and corresponding stratigraphic model. Eight distinct types of lithofacies with seawards increasing grain size were discriminated and interpreted as reworked relict deposits overlain by post-transgressional fluvial sediments. The dominant transport processes in different water depths were identified based on type and orientation of bedforms, as well as bottom water turbidity and lithofacies distribution. Observed bedforms include subaquatic dunes, coarse sand ribbons and sorted bedforms of varying dimensions, which were interpreted as being initially formed by erosion. Under fair weather conditions, sediment is transported from the northwest towards the southeast by littoral drift. During storm events, a current from the southeast to the northweast is induced which is transporting sediment along the shore in up to 35 m water depth. Shorewards oriented cross-shore transport is taking place in up to 60 m water depth and is likewise initiated by storm events. Our study demonstrates how benthic photographic profiling delivers comprehensive compositional, structural and environmental information, which compares well with results obtained by traditional probing methods, but offers much higher spatial resolution while covering larger areas. Multi-sensor benthic profiling enhances the interpretability of acoustic seafloor mapping techniques and is a rapid and economic approach to seabed and habitat mapping especially in muddy to sandy facies.

Georeferenced standardized apparent wave power along the coastline of Cabrera National Park for analysis of essential habitats of littoral species

Wind- induced exposure is one of the major forces shaping the geomorphology and biota in coastal areas. The effect of wave exposure on littoral biota is well known in marine environments (Ekebon et al., 2003; Burrows et al., 2008). In the Cabrera Archipelago National Park wave exposure has demostrated to have an effect on the spatial distribution of different stages of E.marginatus (Alvarez et al., 2010). Standarized average wave exposures during 2008 along the Cabrera Archipelago National park coast line were calculated to be applied in studies of littoral species distribution within the archipelago. Average wave exposure (or apparent wave power) was calculated for points located 50 m equidistant on the coastline following the EXA methodology (EXposure estimates for fragmented Archipelagos) (Ekebon et al., 2003). The average wave exposures were standardized from 1 to 100 (minimum and maximum in the area), showing coastal areas with different levels of mea wave exposure during the year. Input wind data (direction and intensity) from 2008 was registered at the Cabrera mooring located north of Cabrera Archipelago. Data were provided by IMEDEA (CSIC-UIB, TMMOS http://www.imedea.uib-csic.es/tmoos/boyas/). This cartography has been developed under the framework of the project EPIMHAR, funded by the National Park's Network (Spanish Ministry of Environment, Maritime and Rural Affairs, reference: 012/2007 ). Part of this work has been developed under the research programs funded by "Fons de Garantia Agrària i Pesquera de les Illes Balears (FOGAIBA)".

Mapping multi-species habitat use for marine conservation planning

Tese de Doutoramento, Ciências do Mar (Biologia Marinha)

Predictive mapping of powerful owl (Ninox strenua) breeding sites using Geographical Information Systems (GIS) in urban Melbourne, Australia

Urban expansion is a principal process threatening biodiversity globally. It is predicted that over half of the world's population will reside in urban centres by 2010. If we are to conserve biodiversity, a shift in perspective from traditional ecological studies based in natural environments, to studies based in less natural environments is paramount. To effectively conserve species which occur in urban environments, comprehensive analysis is necessary to determine the processes that are driving this urban usage. Geographical Information Systems (GIS) technology provides a valuable tool for efficient spatial analysis and predictive mapping of species distributions.

This study used GIS to analyze current breeding sites for the powerful owl, a vulnerable top order predator in urban Melbourne, Australia. GIS analysis suggests that a number of ecological attributes were influencing powerful owl usage of urban environments. Using these ecological attributes, predictive mapping was undertaken, which identified a number of potential breeding sites for powerful owls within urbanized Melbourne.

Urban environments are traditionally perceived as “the wastelands” of natural environments, however, this study demonstrates that they have the potential to support apex predators, an important finding for the management of rare and threatened species.