Ecological Datasets Research Articles

A Benford’s law-based framework to determine the threshold of occurrence sites for species distribution modelling from ecological monitoring databases

The use of data-based species distribution models (SDMs) has increased significantly in recent years. However, studies of determining the minimum requirements of occurrence sites from ecological monitoring datasets used in species distribution modelling remain insufficient. Therefore, this study proposed a framework to determine the threshold of minimum occurrence sites for SDMs by assessing compliance with Benford’s law. The compliance test verified that the national-scale freshwater fish monitoring dataset was natural and reliable. Results derived from true skill statistics (TSS) determined the minimum number of occurrence sites for reliable species distribution modelling was 20 with a TSS value of 0.793 and an overall accuracy of 0.804. The Benford compliance test has shown to be a useful tool for swift and efficient evaluation of the reliability of species occurrence datasets, or the determination of the threshold of occurrence sites before species distribution modelling. Further studies regarding the evaluation of this method’s transferability to other species and validation using SDM performance are required. Overall, the framework proposed in this study demonstrates that Benford compliance test applied to species monitoring datasets can be used to derive a universal and model-independent minimum occurrence threshold for SDMs.

Extensive behavioral data contained within existing ecological datasets

Long-term ecological datasets contain vast behavioral data, enabling the quantification of among-individual behavioral variation at unprecedented spatiotemporal scales. We detail how behaviors can be extracted and describe how such data can be used to test new hypotheses, inform population and community ecology, and address pressing conservation needs.

A user-friendly guide to using distance measures to compare time series in ecology.

Time series are a critical component of ecological analysis, used to track changes in biotic and abiotic variables. Information can be extracted from the properties of time series for tasks such as classification (e.g., assigning species to individual bird calls); clustering (e.g., clustering similar responses in population dynamics to abrupt changes in the environment or management interventions); prediction (e.g., accuracy of model predictions to original time series data); and anomaly detection (e.g., detecting possible catastrophic events from population time series). These common tasks in ecological research all rely on the notion of (dis-) similarity, which can be determined using distance measures. A plethora of distance measures have been described, predominantly in the computer and information sciences, but many have not been introduced to ecologists. Furthermore, little is known about how to select appropriate distance measures for time-series-related tasks. Therefore, many potential applications remain unexplored. Here, we describe 16 properties of distance measures that are likely to be of importance to a variety of ecological questions involving time series. We then test 42 distance measures for each property and use the results to develop an objective method to select appropriate distance measures for any task and ecological dataset. We demonstrate our selection method by applying it to a set of real-world data on breeding bird populations in the UK and discuss other potential applications for distance measures, along with associated technical issues common in ecology. Our real-world population trends exhibit a common challenge for time series comparisons: a high level of stochasticity. We demonstrate two different ways of overcoming this challenge, first by selecting distance measures with properties that make them well suited to comparing noisy time series and second by applying a smoothing algorithm before selecting appropriate distance measures. In both cases, the distance measures chosen through our selection method are not only fit-for-purpose but are consistent in their rankings of the population trends. The results of our study should lead to an improved understanding of, and greater scope for, the use of distance measures for comparing ecological time series and help us answer new ecological questions.

The Australian Reference Genome Atlas (ARGA): Finding, sharing and reusing Australian genomics data in an occurrence-driven context

Fundamental to the capacity of Australia’s 15,000 biosciences researchers to answer questions in taxonomy, phylogeny, evolution, conservation, and applied fields like crop improvement and biosecurity, is access to trusted genomics (and genetics) datasets. Historically, researchers turned to single points of origin, like GenBank (part of the United States' National Center for Biotechnology Information), to find the reference or comparative data they needed, but the rapidity of data generation using next-gen methods, and the enormous size and diversity of datasets derived from next-gen sequencing methods, mean that single databases no longer contain all data of a specific class, which may be attributable to individual taxa, nor the full breadth of data types relevant for that taxon. Comprehensively searching for taxonomically relevant data, and indeed, data of types germane to the research question, is a significant challenge for researchers. Data are openly available online, but the data may be stored under synonyms or indexed via unconventional taxonomies. Data repositories are largely disconnected and researchers must visit multiple sites to have confidence that their searches have been exhaustive. Databases may focus on single data types and not store or reference other data assets, though they may be relevant for the taxon of interest. Additionally, our survey of the genomics community indicated that researchers are less likely to trust data with inadequately evidenced provenance metadata. This means that genomics data are hard to find and are often untrusted. Moreover, even once found, the data are in formats that do not interoperate with occurrence and ecological datasets, such as those housed in the Atlas of Living Australia. We built the Australian Reference Genome Atlas (ARGA) to overcome the barriers faced by researchers in finding and collating genomics data for Australia’s species, and we have built it so that researchers can search for data within taxonomically accepted contexts and defined intersections and conjunctions with verified and expert ecological datasets. Using a series of ingestion scripts, the ARGA data team has implemented new and customised data mappings that effectively integrate genomics data, ecological traits, and occurrence data within an extended Darwin Core Event framework (GBIF 2018). Here, we will demonstrate how the architecture we derived for ARGA application works, and how it can be extended as new data sources emerge. We then demonstrate how our flexible model can be used to: locate genomics data for taxa of interest; explore data within an ecological context; and calculate metrics for data availability for provincial bioregions. locate genomics data for taxa of interest; explore data within an ecological context; and calculate metrics for data availability for provincial bioregions.

The Charophytes (Charophyceae, Characeae) from Dagestan Aquatic Habitats, North Caucasus: Biogeographical and Barcoding Perspectives

The charophytes in many regions of the world are still poorly understood. This hampers the synthesis of distributional and ecological datasets at worldwide and continental scales, as well as complicates the generalization of species concepts for widely distributed and local taxa. To fill in the blanks for charophytes in the Caucasus and to improve our knowledge of species distribution areas in Eurasia, a field survey and study of available specimens from Dagestan (North Caucasus, Russia) was conducted based on morphological observation using light and scanning electron microscopy and molecular genetic analyses allowing for the precise identification and testing of the presence of cryptic and undescribed taxa. Nineteen new localities for seven Chara species and one Tolypella species, seven new species, and one new genus were identified in the studied region, and one new species in the Caspian Sea region was found. Some species records changed the outline or filled in the gaps in species distribution data. The presence of species distributed mainly in central Eurasia (C. globata, C. neglecta) with mainly Mediterranean–Middle Eastern species (C. gymnophylla) is notable for this region, as well as for other studied regions of the Caucasus characterized by a mixture combination of species with different distribution patterns. Chara gymnophylla was frequent in Dagestan, similar to the Mediterranean and Middle Eastern regions. Small brackish waterbodies on the coast of the Caspian Sea, freshwater mountain rivers, small associated waterbodies, and water reservoirs are the main habitats of charophytes in the studied region. Based on habitat preference and distribution in the Caucasus, recommendations for the protection of some species were suggested. The lack of endemic species among charophytes from Dagestan and Caucasus contrasts with the flora of terrestrial magnoliophytes that is rich in species endemism.

Automated face recognition using deep neural networks produces robust primate social networks and sociality measures

AbstractLongitudinal video archives of behaviour are crucial for examining how sociality shifts over the lifespan in wild animals. New approaches adopting computer vision technology hold serious potential to capture interactions and associations between individuals in video at large scale; however, such approaches need a priori validation, as methods of sampling and defining edges for social networks can substantially impact results.Here, we apply a deep learning face recognition model to generate association networks of wild chimpanzees using 17 years of a video archive from Bossou, Guinea. Using 7 million detections from 100 h of video footage, we examined how varying the size of fixed temporal windows (i.e. aggregation rates) for defining edges impact individual‐level gregariousness scores.The highest and lowest aggregation rates produced divergent values, indicating that different rates of aggregation capture different association patterns. To avoid any potential bias from false positives and negatives from automated detection, an intermediate aggregation rate should be used to reduce error across multiple variables. Individual‐level network‐derived traits were highly repeatable, indicating strong inter‐individual variation in association patterns across years and highlighting the reliability of the method to capture consistent individual‐level patterns of sociality over time. We found no reliable effects of age and sex on social behaviour and despite a significant drop in population size over the study period, individual estimates of gregariousness remained stable over time.We believe that our automated framework will be of broad utility to ethology and conservation, enabling the investigation of animal social behaviour from video footage at large scale, low cost and high reproducibility. We explore the implications of our findings for understanding variation in sociality patterns in wild ape populations. Furthermore, we examine the trade‐offs involved in using face recognition technology to generate social networks and sociality measures. Finally, we outline the steps for the broader deployment of this technology for analysis of large‐scale datasets in ecology and evolution.

DNA metabarcoding captures different macroinvertebrate biodiversity than morphological identification approaches across a continental scale

AbstractDNA‐based aquatic biomonitoring methods show promise to provide rapid, standardized, and efficient biodiversity assessment to supplement and in some cases replace current morphology‐based approaches that are often less efficient and can produce inconsistent results. Despite this potential, broad‐scale adoption of DNA‐based approaches by end‐users remains limited, and studies on how these two approaches differ in detecting aquatic biodiversity across large spatial scales are lacking. Here, we present a comparison of DNA metabarcoding and morphological identification, leveraging national‐scale, open‐source, ecological datasets from the National Ecological Observatory Network (NEON). Across 24 wadeable streams in North America with 179 paired sample comparisons, we found that DNA metabarcoding detected twice as many unique taxa than morphological identification overall. The two approaches showed poor congruence in detecting the same taxa, averaging 59%, 35%, and 23% of shared taxa detected at the order, family, and genus levels, respectively. Importantly, the two approaches detected different proportions of indicator taxa like %EPT and %Chironomidae. DNA metabarcoding detected far fewer Chironomid and Trichopteran taxa than morphological identification, but more Ephemeropteran and Plecopteran taxa, a result likely due to primer choice. Overall, our results showed that DNA metabarcoding and morphological identification detected different benthic macroinvertebrate communities. Despite these differences, we found that the same environmental variables were correlated with invertebrate community structure, suggesting that both approaches can accurately detect biodiversity patterns across environmental gradients. Further refinement of DNA metabarcoding protocols, primers, and reference libraries–as well as more standardized, large‐scale comparative studies–may improve our understanding of the taxonomic agreement and data linkages between DNA metabarcoding and morphological approaches.

Coevolution With Host Fishes Shapes Parasitic Life Histories in a Group of Freshwater Mussels (Unionidae: Quadrulini).

Ecological interactions among species often lead to parasitic lineages coevolving with host resources, which is often suggested as the primary driver of parasite diversification. Freshwater mussels are bivalves that possess a parasitic life cycle requiring larval encystment on freshwater vertebrates to complete metamorphosis. The North American freshwater mussel tribe Quadrulini has a suite of life history adaptations including highly specialized patterns of host use, infection strategies, and variable larval morphologies. However, the evolution of life histories has yet to be explored using phylogenetic comparative methods. In this study, we use a holistic approach incorporating biogeographical, ecological, molecular, and morphological datasets to reconstruct the evolution of Quadrulini. Comparative phylogenetic analyses suggested the diversification of Quadrulini has been driven, at least in part, by codiversification with their primary host fishes in Ictaluridae. Major diversification events in both ictalurids and quadrulines were estimated to have occurred in the Mississippi River basin throughout the Miocene. Life history characteristics associated with parasitism were supported to have coevolved with host repertories, supporting the hypothesis that ecological interactions with host fishes have shaped the evolution of highly specialized traits in this group. Our findings demonstrate the importance of ecological interactions with host resources in shaping the evolutionary history of freshwater mussels.

What's going to be on the menu with global environmental changes?

Ongoing anthropogenic change is altering the planet at an unprecedented rate, threatening biodiversity, and ecosystem functioning. Species are responding to abiotic pressures at both individual and population levels, with changes affecting trophic interactions through consumptive pathways. Collectively, these impacts alter the goods and services that natural ecosystems will provide to society, as well as the persistence of all species. Here, we describe the physiological and behavioral responses of species to global changes on individual and population levels that result in detectable changes in diet across terrestrial and marine ecosystems. We illustrate shifts in the dynamics of food webs with implications for animal communities. Additionally, we highlight the myriad of tools available for researchers to investigate the dynamics of consumption patterns and trophic interactions, arguing that diet data are a crucial component of ecological studies on global change. We suggest that a holistic approach integrating the complexities of diet choice and trophic interactions with environmental drivers may be more robust at resolving trends in biodiversity, predicting food web responses, and potentially identifying early warning signs of diversity loss. Ultimately, despite the growing body of long-term ecological datasets, there remains a dearth of diet ecology studies across temporal scales, a shortcoming that must be resolved to elucidate vulnerabilities to changing biophysical conditions.

Combining multiple data sources with different biases in state-space models for population dynamics.

The resolution at which animal populations can be modeled can be increased when multiple datasets corresponding to different life stages are available, allowing, for example, seasonal instead of annual descriptions of dynamics. However, the abundance estimates used for model fitting can have multiple sources of error, both random and systematic, namely bias. We focus here on the consequences of, and how to address, differing and unknown observation biases when fitting models.State-space models (SSMs) separate process variation and observation error, thus providing a framework to account for different and unknown estimate biases across multiple datasets. Here we study the effects on the inference of including or excluding bias parameters for a sequential life stage population dynamics SSM using a combination of theory, simulation experiments, and an empirical example.When the data, that is, abundance estimates, are unbiased, including bias parameters leads to increased imprecision compared to a model that correctly excludes bias parameters. But when observations are biased and no bias parameters are estimated, recruitment and survival processes are inaccurately estimated and estimates of process variance become biased high. These problems are substantially reduced by including bias parameters and fixing one of them at even an incorrect value. The primary inferential challenge is that models with bias parameters can show properties of being parameter redundant even when they are not in theory.Combining multiple datasets into a single analysis by using bias parameters to rescale data can offer significant improvements to inference and model diagnostics. Because their estimability in practice is dataset specific and will likely require more precise estimates than might be expected from ecological datasets, we outline some strategies for characterizing process uncertainty when it is confounded by bias parameters.

Mind wandering state detection during video-based learning via EEG.

The aim of this study is to explore the potential of technology for detecting mind wandering, particularly during video-based distance learning, with the ultimate benefit of improving learning outcomes. To overcome the challenges of previous mind wandering research in ecological validity, sample balance, and dataset size, this study utilized practical electroencephalography (EEG) recording hardware and designed a paradigm consisting of viewing short-duration video lectures under a focused learning condition and a future planning condition. Participants estimated statistics of their attentional state at the end of each video, and we combined this rating scale feedback with self-caught key press responses during video watching to obtain binary labels for classifier training. EEG was recorded using an 8-channel system, and spatial covariance features processed by Riemannian geometry were employed. The results demonstrate that a radial basis function kernel support vector machine classifier, using Riemannian-processed covariance features from delta, theta, alpha, and beta bands, can detect mind wandering with a mean area under the receiver operating characteristic curve (AUC) of 0.876 for within-participant classification and AUC of 0.703 for cross-lecture classification. Furthermore, our results suggest that a short duration of training data is sufficient to train a classifier for online decoding, as cross-lecture classification remained at an average AUC of 0.689 when using 70% of the training set (about 9 min). The findings highlight the potential for practical EEG hardware in detecting mind wandering with high accuracy, which has potential application to improving learning outcomes during video-based distance learning.

Taxonomic revision of the tribe Acraeini Boisduval, 1833 (Papilionoidea: Nymphalidae: Heliconiinae)

The systematic and taxonomic history of the tribe Acraeini is reviewed. Using a robust dated molecular phylogeny published in 2021, and combining this with morphological and ecological data-sets, we divided the clearly paraphyletic genus Acraea (sensu lato) into five genera. We raised the subgenera Rubraea Henning, 1992, Stephenia Henning, 1992 to genera and reinstate Bematistes Hemming, 1935 and Telchinia Hübner, [1819] as valid genera. The genus Tildia gen. nov. is erected for a monophyletic clade of Acraea (sensu lato), and we treat the Neotropical Actinote and the Afrotropical Telchinia as separate genera. Because of the relatively low sampling of ‘Telchinia’ species in the dated molecular phylogeny we suggest that the possibly paraphyletic Telchinia be further investigated using more complete sampling. We furthermore present an updated alpha taxonomy for the six Afrotropical genera in the tribe – Acraea, Rubraea stat. nov, Stephenia stat. nov, Tildia gen. nov., Bematistes stat. rev. and Telchinia stat. rev. The current alpha taxonomy of the New World genus Actinote and the Old World genus Cethosia is given.

Improving ecological data science with workflow management software

AbstractPressing environmental research questions demand the integration of increasingly diverse and large‐scale ecological datasets as well as complex analytical methods, which require specialized tools and resources.Computational training for ecological and evolutionary sciences has become more abundant and accessible over the past decade, but tool development has outpaced the availability of specialized training. Most training for scripted analyses focuses on individual analysis steps in one script rather than creating a scripted pipeline, where modular functions comprise an ecosystem of interdependent steps. Although current computational training creates an excellent starting place, linear styles of scripting can risk becoming labor‐ and time‐intensive and less reproducible by often requiring manual execution. Pipelines, however, can be easily automated or tracked by software to increase efficiency and reduce potential errors. Ecology and evolution would benefit from techniques that reduce these risks by managing analytical pipelines in a modular, readily parallelizable format with clear documentation of dependencies.Workflow management software (WMS) can aid in the reproducibility, intelligibility and computational efficiency of complex pipelines. To date, WMS adoption in ecology and evolutionary research has been slow. We discuss the benefits and challenges of implementing WMS and illustrate its use through a case study with thetargets rpackage to further highlight WMS benefits through workflow automation, dependency tracking and improved clarity for reviewers.Although WMS requires familiarity with function‐oriented programming and careful planning for more advanced applications and pipeline sharing, investment in training will enable access to the benefits of WMS and impart transferable computing skills that can facilitate ecological and evolutionary data science at large scales.

BIRDIE: A data pipeline to inform wetland and waterbird conservation at multiple scales

IntroductionEfforts to collect ecological data have intensified over the last decade. This is especially true for freshwater habitats, which are among the most impacted by human activity and yet lagging behind in terms of data availability. Now, to support conservation programmes and management decisions, these data need to be analyzed and interpreted; a process that can be complex and time consuming. The South African Biodiversity Data Pipeline for Wetlands and Waterbirds (BIRDIE) aims to help fast and efficient information uptake, bridging the gap between raw ecological datasets and the information final users need.MethodsBIRDIE is a full data pipeline that takes up raw data, and estimates indicators related to waterbird populations, while keeping track of their associated uncertainty. At present, we focus on the assessment of species abundance and distribution in South Africa using two citizen-science bird monitoring datasets, namely: the African Bird Atlas Project and the Coordinated Waterbird Counts. These data are analyzed with occupancy and state-space models, respectively. In addition, a suite of environmental layers help contextualize waterbird population indicators, and link these to the ecological condition of the supporting wetlands. Both data and estimated indicators are accessible to end users through an online portal and web services.Results and discussionWe have designed a modular system that includes tasks, such as: data cleaning, statistical analysis, diagnostics, and computation of indicators. Envisioned users of BIRDIE include government officials, conservation managers, researchers and the general public, all of whom have been engaged throughout the project. Acknowledging that conservation programmes run at multiple spatial and temporal scales, we have developed a granular framework in which indicators are estimated at small scales, and then these are aggregated to compute similar indicators at broader scales. Thus, the online portal is designed to provide spatial and temporal visualization of the indicators using maps, time series and pre-compiled reports for species, sites and conservation programmes. In the future, we aim to expand the geographical coverage of the pipeline to other African countries, and develop more indicators specific to the ecological structure and function of wetlands.

Synthesizing forecasts to inform decision‐making and advance ecological theory

Synthesizing forecasts to inform decision‐making and advance ecological theory

Benthic biogeographic patterns on the deep Brazilian margin.

The Brazilian continental margin (BCM) extends from the Tropical to the Subtropical Atlantic Ocean, with much of its seafloor within deep waters, supporting rich geomorphological features and under wide productivity gradients. Deep-sea biogeographic boundaries on the BCM have been limited to studies that used water mass and salinity properties of deep-water masses, partly as a result of historical under sampling and a lack of consolidation of available biological and ecological datasets. The aim of this study was to consolidate benthic assemblage datasets and test current oceanographic biogeographical deep-sea boundaries (200-5,000 m) using available faunal distributions. We retrieved over 4,000 benthic data records from open-access databases and used cluster analysis to examine assemblage distributions against the deep-sea biogeographical classification scheme from Watling et al. (2013). Starting from the assumption that vertical and horizontal distribution patterns can vary regionally, we test other schemes incorporating latitudinal and water masses stratification within the Brazilian margin. As expected, the classification scheme based on benthic biodiversity is in overall agreement with the general boundaries proposed by Watling et al. (2013). However, our analysis allowed much refinement in the former boundaries, and here we propose the use of two biogeographic realms, two provinces and seven bathyal ecoregions (200-3,500 m), and three abyssal provinces (>3,500 m) along the BCM. The main driver for these units seems to be latitudinal gradients as well as water mass characteristics such as temperature. Our study provides a significant improvement of benthic biogeographic ranges along the Brazilian continental margin allowing a more detailed recognition of its biodiversity and ecological value, and also supports the needed spatial management for industrial activities occurring in its deep waters.

An Improved Link Prediction Approach for Directed Complex Networks Using Stochastic Block Modeling

Link prediction finds the future or the missing links in a social–biological complex network such as a friendship network, citation network, or protein network. Current methods to link prediction follow the network properties, such as the node’s centrality, the number of edges, or the weights of the edges, among many others. As the properties of the networks vary, the link prediction methods also vary. These methods are inaccurate since they exploit limited information. This work presents a link prediction method based on the stochastic block model. The novelty of our approach is the three-step process to find the most-influential nodes using the m-PageRank metric, forming blocks using the global clustering coefficient and, finally, predicting the most-optimized links using maximum likelihood estimation. Through the experimental analysis of social, ecological, and biological datasets, we proved that the proposed model outperforms the existing state-of-the-art approaches to link prediction.

Shed female caribou antlers extend records of calving activity on the Arctic National Wildlife Refuge by millennia

Caribou (Rangifer tarandus) have among the longest annual migrations of any terrestrial mammal as they move from winter ranges to spring calving grounds. Biomonitoring records indicate broad consistencies in calving geography across the last several decades, but how long have herds used particular calving grounds? Furthermore, how representative are modern patterns of calving geography to periods that pre-date recent climatic perturbations and increased anthropogenic stresses? While modern ecological datasets are not long enough to address these questions, bones from past generations of caribou lying on the tundra provide unique opportunities to study historical calving geography. This is possible because female caribou shed their antlers within days of giving birth, releasing a skeletal indicator of calving. Today, the Coastal Plain of the Arctic National Wildlife Refuge (Alaska) is a key calving ground for the Porcupine Caribou Herd (PCH). To test the duration across which caribou have used this area as a calving ground, we radiocarbon dated three highly weathered female antlers collected from tundra surfaces on the Coastal Plain. Calibrated radiocarbon dates indicate that these antlers were shed between ~1,600 and more than 3,000 calendar years ago. The antiquity of these shed antlers provides the first physical evidence of calving activity on the PCH calving grounds from previous millennia, substantiating the long ecological legacy of the Coastal Plain as a caribou calving ground. Comparisons to published lake core records also reveal that dates of two of the antlers correspond to periods with average summer temperatures that were warmer than has been typical during the last several decades of biomonitoring. This finding expands the range of climatic settings in which caribou are known to use the current PCH calving grounds and suggests that the Coastal Plain of the Arctic Refuge may remain an important caribou calving ground during at least portions of predicted future warming. Discarded skeletal materials provide opportunities to assess the historical states of living populations, including aspects of reproductive biology and migration. Particularly in high-latitude settings, these insights can extend across millennia and offer rare glimpses into the past that can inform current and future management policies.

Research on land cover type classification method based on improved MaskFormer for remote sensing images.

High-resolution remote sensing images have the characteristics of wide imaging coverage, rich spectral information and unobstructed by terrain and features. All of them provide convenient conditions for people to study land cover types. However, most existing remote sensing image land cover datasets are only labeled with some remote sensing images of low elevation plain areas, which is highly different from the topography and landscape of highland mountainous areas. In this study, we construct a Qilian County grassland ecological element dataset to provide data support for highland ecological protection. To highlight the characteristics of vegetation, our dataset only includes the RGB spectrum fused with the near-infrared spectrum. We then propose a segmentation network, namely, the Shunted-MaskFormer network, by using a mask-based classification method, a multi-scale, high-efficiency feature extraction module and a data-dependent upsampling method. The extraction of grassland land types from 2 m resolution remote sensing images in Qilian County was completed, and the generalization ability of the model on a small Gaofen Image Dataset (GID) verified. Results: (1) The MIoU of the optimised network model in the Qilian grassland dataset reached 80.75%, which is 2.37% higher compared to the suboptimal results; (2) the optimized network model achieves better segmentation results even for small sample classes in data sets with unbalanced sample distribution; (3) the highest MIOU of 72.3% is achieved in the GID dataset of open remote sensing images containing five categories; (4) the size of the optimized model is only one-third of the sub-optimal model.

Personalized Digital Health Information to Substantiate Human-Delivered Exercise Support for Adults With Type 1 Diabetes.

Pilot-test personalized digital health information to substantiate human-delivered exercise support for adults with type 1 diabetes (T1D). Single-group, 2-week baseline observation, then 10-week intervention with follow-up observation. Community-based sample participating remotely with physician oversight. Volunteers aged 18 to 65 years with T1D screened for medical readiness for exercise intervention offerings. N = 20 enrolled, and N = 17 completed all outcomes with 88% to 91% biosensor adherence. Feedback on personalized data from continuous glucose monitoring (CGM), its intersection with other ecological data sets (exercise, mood, and sleep), and other informational and motivational elements (exercise videos, text-based exercise coach, and self-monitoring diary). Feasibility (use metrics and assessment completion), safety (mild and severe hypoglycemia, and diabetic ketoacidosis), acceptability (system usability scale, single items, and interview themes), and standard clinical and psychosocial assessments. Participants increased exercise from a median of 0 (Interquartile range, 0-21) to 64 (20-129) minutes per week ( P = 0.001, d = 0.71) with no severe hypoglycemia or ketoacidosis. Body mass index increased (29.5 ± 5.1 to 29.8 ± 5.4 kg/m 2 , P = 0.02, d = 0.57). Highest satisfaction ratings were for CGM use (89%) and data on exercise and its intersection with CGM and sleep (94%). Satisfaction was primarily because of improved exercise management behavioral skills, although derived motivation was transient. The intervention was feasible, safe, and acceptable. However, there is a need for more intensive, sustained support. Future interventions should perform analytics upon the digital health information and molecular biomarkers (eg, genomics) to make exercise support tools that are more personalized, automated, and intensive than our present offerings.