Artificial intelligence convolutional neural networks map giant kelp forests from satellite imagery.

Climate change is producing shifts in the distribution and abundance of marine species. Such is the case of kelp forests, important marine ecosystem-structuring species whose distributional range limits have been shifting worldwide. Synthesizing long-term time series of kelp forest observations is therefore vital for understanding the drivers shaping ecosystem dynamics and for predicting responses to ongoing and future climate changes. Traditional methods of mapping kelp from satellite imagery are time-consuming and expensive, as they require high amount of human effort for image processing and algorithm optimization. Here we propose the use of mask region-based convolutional neural networks (Mask R-CNN) to automatically assimilate data from open-source satellite imagery (Landsat Thematic Mapper) and detect kelp forest canopy cover. The analyses focused on the giant kelp Macrocystis pyrifera along the shorelines of southern California and Baja California in the northeastern Pacific. Model hyper-parameterization was tuned through cross-validation procedures testing the effect of data augmentation, and different learning rates and anchor sizes. The optimal model detected kelp forests with high performance and low levels of overprediction (Jaccard's index: 0.87 ± 0.07; Dice index: 0.93 ± 0.04; over prediction: 0.06) and allowed reconstructing a time series of 32 years in Baja California (Mexico), a region known for its high variability in kelp owing to El Niño events. The proposed framework based on Mask R-CNN now joins the list of cost-efficient tools for long-term marine ecological monitoring, facilitating well-informed biodiversity conservation, management and decision making.

Weak biodiversity connectivity in the European network of no-take marine protected areas.

The need for international cooperation in marine resource management and conservation has been reflected in the increasing number of agreements aiming for effective and well-connected networks of Marine Protected Areas (MPAs). However, the extent to which individual MPAs are connected remains mostly unknown. Here, we use a biophysical model tuned with empirical data on species dispersal ecology to predict connectivity of a vast spectrum of biodiversity in the European network of marine reserves (i.e., no-take MPAs). Our results highlight the correlation between empirical propagule duration data and connectivity potential and show weak network connectivity and strong isolation for major ecological groups, resulting from the lack of direct connectivity corridors between reserves over vast regions. The particularly high isolation predicted for ecosystem structuring species (e.g., corals, sponges, macroalgae and seagrass) might potentially undermine biodiversity conservation efforts if local retention is insufficient and unmanaged populations are at risk. Isolation might also be problematic for populations' persistence in the light of climate change and expected species range shifts. Our findings provide novel insights for management directives, highlighting the location of regions requiring additional marine reserves to function as stepping-stone connectivity corridors.

Mediterranean diet is inversely associated with steatosis and fibrosis and decreases ten-year diabetes and cardiovascular risk in NAFLD subjects: Results from the ATTICA prospective cohort study.

BACKGROUND AND AIMS: We assessed the association of Mediterranean diet with NAFLD and their interaction in predicting ten-year diabetes onset and first fatal/non-fatal cardiovascular disease (CVD) incidence. METHODS: The ATTICA prospective observational study in Athens, Greece included 1,514 men and 1,528 women (>18 years old) free-of-CVD at baseline. Liver steatosis and fibrosis indices were calculated. Mediterranean diet adherence was assessed through MedDietScore. At the ten-year follow-up visit, CVD evaluation was performed in an a priori specified subgroup of n = 2,020 participants and diabetes onset in n = 1,485 free-of-diabetes participants. RESULTS: MedDietScore was inversely associated with steatosis and fibrosis; e.g. in the case of the TyG index the Odds Ratio (OR) of the 3rd vs. 1st MedDietScore tertile was = 0·53, [95% Confidence Interval (95% CI) (0·29, 0·95)] and the associations persisted in multi-adjusted models. NAFLD predicted incident diabetes prospectively over a ten year period [HR = 1·87, 95% CI (0·75, 4·61)] and the association remained significant only in subjects with low MedDietScore (below median) whereas diabetes onset among subjects with higher MedDietScore was not influenced by NAFLD. Similarly, NAFLD predicted CVD [Hazard Ratio (HR) = 3·01, 95%CI(2·28, 3·95)]; the effect remained significant only in subjects with MedDietScore below median [HR = 1·38, 95% CI (1·00, 1·93)] whereas it was essentially null [HR = 1·00,95% CI (0·38, 2·63)] among subjects with higher score. Mediation analysis revealed that adiponectin and adiponectin-to-leptin ratio were the strongest mediators. CONCLUSIONS: We report an inverse association between Mediterranean diet and NAFLD. Mediterranean diet protected against diabetes and CVD prospectively among subjects with NAFLD.