Large Number Of Locations Research Articles

Three-layered location recommendation algorithm using spectral clustering

Users utilize Location-Based Social Networks (LBSNs) to check into diverse venues and share their experiences through ratings and comments. However, these platforms typically feature a considerably larger number of locations than users, resulting in a challenge known as insufficient historical data or user-location matrix sparsity. This sparsity arises because not all users can check into all available locations on a given LBSN, such as Yelp. To address this challenge, this paper proposes combining Spectral Clustering with a three-layered location recommendation model to develop a recommender system named LSC, applied to Yelp datasets. LSC leverages various information, including users’ check-in data, demographics, location demographics, and users’ friendship network data, to train the recommender system and generate recommendations. Evaluation of LSC’s performance utilizes the Yelp dataset and several comparison metrics, such as accuracy, RMSE, and F1-score. The results demonstrate that our proposed algorithm delivers reliable and significant performance improvements across various evaluation metrics compared to competing algorithms.

Task-Adaptive Meta-Learning Framework for Advancing Spatial Generalizability

Spatio-temporal machine learning is critically needed for a variety of societal applications, such as agricultural monitoring, hydrological forecast, and traffic management. These applications greatly rely on regional features that characterize spatial and temporal differences. However, spatio-temporal data often exhibit complex patterns and significant data variability across different locations. The labels in many real-world applications can also be limited, which makes it difficult to separately train independent models for different locations. Although meta learning has shown promise in model adaptation with small samples, existing meta learning methods remain limited in handling a large number of heterogeneous tasks, e.g., a large number of locations with varying data patterns. To bridge the gap, we propose task-adaptive formulations and a model-agnostic meta-learning framework that transforms regionally heterogeneous data into location-sensitive meta tasks. We conduct task adaptation following an easy-to-hard task hierarchy in which different meta models are adapted to tasks of different difficulty levels. One major advantage of our proposed method is that it improves the model adaptation to a large number of heterogeneous tasks. It also enhances the model generalization by automatically adapting the meta model of the corresponding difficulty level to any new tasks. We demonstrate the superiority of our proposed framework over a diverse set of baselines and state-of-the-art meta-learning frameworks. Our extensive experiments on real crop yield data show the effectiveness of the proposed method in handling spatial-related heterogeneous tasks in real societal applications.

Fast Bayesian inference for large occupancy datasets.

In recent years, the study of species' occurrence has benefited from the increased availability of large-scale citizen-science data. While abundance data from standardized monitoring schemes are biased toward well-studied taxa and locations, opportunistic data are available for many taxonomic groups, from a large number of locations and across long timescales. Hence, these data provide opportunities to measure species' changes in occurrence, particularly through the use of occupancy models, which account for imperfect detection. These opportunistic datasets can be substantially large, numbering hundreds of thousands of sites, and hence present a challenge from a computational perspective, especially within a Bayesian framework. In this paper, we develop a unifying framework for Bayesian inference in occupancy models that account for both spatial and temporal autocorrelation. We make use of the Pólya-Gamma scheme, which allows for fast inference, and incorporate spatio-temporal random effects using Gaussian processes (GPs), for which we consider two efficient approximations: subset of regressors and nearest neighbor GPs. We apply our model to data on two UK butterfly species, one common and widespread and one rare, using records from the Butterflies for the New Millennium database, producing occupancy indices spanning 45 years. Our framework can be applied to a wide range of taxa, providing measures of variation in species' occurrence, which are used to assess biodiversitychange.

Active Learning for Saddle Point Calculation

The saddle point (SP) calculation is a grand challenge for computationally intensive energy function in computational chemistry area, where the saddle point may represent the transition state. The traditional methods need to evaluate the gradients of the energy function at a very large number of locations. To reduce the number of expensive computations of the true gradients, we propose an active learning framework consisting of a statistical surrogate model, Gaussian process regression (GPR) for the energy function, and a single-walker dynamics method, gentle accent dynamics (GAD), for the saddle-type transition states. SP is detected by the GAD applied to the GPR surrogate for the gradient vector and the Hessian matrix. Our key ingredient for efficiency improvements is an active learning method which sequentially designs the most informative locations and takes evaluations of the original model at these locations to train GPR. We formulate this active learning task as the optimal experimental design problem and propose a very efficient sample-based sub-optimal criterion to construct the optimal locations. We show that the new method significantly decreases the required number of energy or force evaluations of the original model.

Multiplexed Analysis of Photochemical Oxidants Using a Nanoparticle-Based Optoelectronic Nose.

Photochemical pollutants pose a substantial threat to human health in both outdoor and indoor environments. Herein, we prepare a class of gold nanoparticle-based colorimetric sensor arrays on optimized hydrophobic substrates using a simple pin-printing method for accurate identification and quantification of various gas-phase oxidants, as these microdetectors are low cost, sensitive, and easy to fabricate. For an array of AuNP sensors modified with various thiol-terminated ligands, a unique and distinguishable change in color (i.e., red, green, and blue response patterns) was obtained for each specific pollutant for molecular fingerprinting. Remarkable discrimination among 15 gases at a fairly low vapor concentration (i.e., 500 ppb) was illustrated using standard chemometric methods. Using digital imaging, the AuNP colorimetric sensor array offers ultrasensitive dosimetric identification of gas-phase oxidants relevant to outdoor and indoor air pollution, with limits of detection generally at sub-ppb levels for 2 h measurement. As a practical application, the sensor array is able to predict the overall air quality in indoor office environments over 24 h. Such sensor array based on chemically induced sintering of nanoparticles has significant implications for the development of nanosensors used in continuous monitoring of potential airborne pollutants at low concentrations from a large number of locations in a cost-effective manner.

Meteorologically Stratified Gust Factors for Forecasting Peak Wind Gusts across the United States

ABSTRACTWind gusts are common to everyday life and affect a wide range of interests including wind energy, structural design, forestry, and fire danger. Strong gusts are a common environmental hazard that can damage buildings, bridges, aircraft, and trains, and interrupt electric power distribution, air traffic, waterways transport, and port operations. Despite representing the component of wind most likely to be associated with serious and costly hazards, reliable forecasts of peak wind gusts have remained elusive. A project at the University of Wisconsin–Milwaukee is addressing the need for improved peak gust forecasts with the development of the meteorologically stratified gust factor (MSGF) model. The MSGF model combines gust factors (the ratio of peak wind gust to average wind speed) with wind speed and direction forecasts to predict hourly peak wind gusts. The MSGF method thus represents a simple, viable option for the operational prediction of peak wind gusts. Here we describe the results of a project designed to provide the site-specific gust factors necessary for operational use of the MSGF model at a large number of locations across the United States. Gust web diagrams depicting the wind speed– and wind direction–stratified gust factors, as well as peak gust climatologies, are presented for all sites analyzed.

Design and Evaluation of Calibrated and Seamless Ensemble Weather Forecasts for Crop Protection Applications

AbstractAgriculture is a highly weather-dependent activity, and climatic conditions impact both directly crop growth and indirectly diseases and pest developments causing yield losses. Weather forecasts are now a major component of various decision-support systems that assist farmers to optimize the positioning of crop protection treatments. However, properly accounting for weather uncertainty in these systems still remains a challenge. In this paper, three global and regional ensemble prediction systems (EPSs), covering different spatiotemporal scales, are coupled to a temperature-driven developmental model for grapevine moths in order to provide probabilistic forecasts of treatment dates. It is first shown that a parametric postprocessing of the EPSs significantly improves the prediction of treatment dates. Anticipating the need for phytosanitary treatments also requires seamless weather forecasts from the next hour to subseasonal time scales. An approach is presented to design seamless ensemble forecasts from the combination of the three EPSs used. The proposed method is able to leverage the increased performance of high-resolution EPS at short ranges, while ensuring a smooth transition toward larger-scale EPSs for longer ranges. The added value of this seamless integration on agronomic predictions is, however, difficult to assess with the current experimental setup. Additional simulations over a larger number of locations and years may be required.

Design Basis of Movable Scaffolding Systems Following American and European Code Provisions and Recommendations

Construction of bridges span-by-span with Movable Scaffolding Systems (MSSs) is a very efficient and competitive technology. Normally used for spans between 25 and 70m, the technology has allowed reaching longer spans due to technological advances, specifically in bridge construction equipment. Thereby, the use of MSS has become widespread and well-accepted in a large number of locations across the USA and Europe. Nevertheless, despite its extended application, there is no single specific code provision that can explain, control, and give recommendations about all aspects of MSS during its design and usage. On the contrary, the information is spread over several documents. This paper aims at bridging this gap by providing an extensive review of code provisions and recommendations that can be valid for the MSS design. Applicability of these documents is discussed by analysing loads, safety factors, load combinations, limit states, as well as structural analysis and design. After this, a proposal of a design basis for MSS is presented for each aspect mentioned following provisions and recommendations of the considered codes.

Measurement of Air Pollution Parameters in Montenegro Using the Ecomar System.

Particulate matter air pollution is one of the most dangerous pollutants nowadays and an indirect cause of numerous diseases. A number of these consequences could possibly be avoided if the right information about air pollution were available at a large number of locations, especially in urban areas. Unfortunately, this is not the case today. In the whole of Europe, there are just approximately 3000 automated measuring stations for PM10, and only about 1400 stations equipped for PM2.5 measurement. In order to improve this issue and provide availability of real-time data about air pollution, different low-cost sensor-based solutions are being considered both on-field and in laboratory research. In this paper, we will present the results of PM particle monitoring using a self-developed Ecomar system. Measurements are performed in two cities in Montenegro, at seven different locations during several periods. In total, three Ecomar systems were used during 1107 days of on-field measurements. Measurements performed at two locations near official automated measuring stations during 610 days justified that the Ecomar system performance is satisfying in terms of reliability and measurement precision (NRMSE 0.33 for PM10 and 0.44 for PM2.5) and very high in terms of data validity and operating stability (Ecomar 94.13%–AMS 95.63%). Additionally, five distant urban/rural locations with different traffic, green areas, and nearby industrial objects were utilized to highlight the need for more dense spatial distributions of measuring locations. To our knowledge, this is the most extensive study of low-cost sensor-based air quality measurement systems in terms of the duration of the on-field tests in the Balkan region.

A differentially private location generalization approach to guarantee non-uniform privacy in moving objects databases

Recently there has been much interest in moving objects databases because of their applications in many domains, such as location-based services and traffic management. Moving objects databases store and manage information representing changes in the spatial properties of moving objects over time. Meanwhile, privacy protection has been one of the most important concerns in these databases. In this paper, we study this problem by presenting DPLG, a location generalization approach for moving objects databases that preserves the strong guarantee of differential privacy. Our main goal is to guarantee non-uniform privacy for locations with different privacy protection requirements while being scalable for spatial domains with a large number of locations. For this purpose, we use location generalization in such a way that locations with higher privacy protection requirements are generalized to larger ones. Location generalization also has the advantage that it enables DPLG to reduce the number of locations and, thus, to keep the running time and space requirements as reasonable as possible. We also present two post-processing techniques, namely, consistency constraints enforcement and quality improvement, to have consistent query answers and to reduce query errors caused by location generalization. The quality improvement technique divides the noisy count of each generalized location among the reference locations it contains homogeneously or heterogeneously. Extensive experiments demonstrate that, in addition to keeping reasonable the running time and space requirements, DPLG improves the utility of query answers for locations with lower privacy protection requirements in comparison to those with higher privacy protection requirements while satisfying differential privacy.

High‐dimensional multivariate geostatistics: A Bayesian matrix‐normal approach

AbstractJoint modeling of spatially oriented dependent variables is commonplace in the environmental sciences, where scientists seek to estimate the relationships among a set of environmental outcomes accounting for dependence among these outcomes and the spatial dependence for each outcome. Such modeling is now sought for massive data sets with variables measured at a very large number of locations. Bayesian inference, while attractive for accommodating uncertainties through hierarchical structures, can become computationally onerous for modeling massive spatial data sets because of its reliance on iterative estimation algorithms. This article develops a conjugate Bayesian framework for analyzing multivariate spatial data using analytically tractable posterior distributions that obviate iterative algorithms. We discuss differences between modeling the multivariate response itself as a spatial process and that of modeling a latent process in a hierarchical model. We illustrate the computational and inferential benefits of these models using simulation studies and analysis of a vegetation index data set with spatially dependent observations numbering in the millions.

Analysis of Trends in Marine Water Quality Using Environmental Impact Assessment Monitoring Data: A Case Study of Busan New Port

Lee, S.M.; Lee, E.J.; Yoo, H.S., and Lee, M.J., 2020. Analysis of trends in marine water quality using environmental impact assessment monitoring data: a case study of Busan new port. In: Jung, H.-S.; Lee, S.; Ryu, J.-H., and Cui, T. (eds.), Advances in Geospatial Research of Coastal Environments. Journal of Coastal Research, Special Issue No. 102, pp. 39-46. Coconut Creek (Florida), ISSN 0749-0208.Various development projects, including harbor construction, are carried out in coastal areas. In South Korea, the performance of environmental impact assessment (EIA) is legally required for development projects in marine areas to monitor the impacts of these projects on offshore marine environments. National Marine Environmental Measurement Network (NMEMN) data can be used to examine the overall status of the marine environment prior to a development project, but these data are very limited. In contrast, EIA data are collected from a large number of locations over various periods. The purpose of this study was to analyze trends in the coastal marine environment using NMEMN and EIA data collected in 2010–2011 for Busan New Port, where multiple development projects are underway. Use of the combined dataset enabled more effective analysis of trends in marine water quality, including those in dissolved oxygen, the chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP), than did the use of NMEMN data alone. Values increased in the bottom layer, especially in the summer, after port operation had begun. For example, average summer ranges for COD, TN, and TP in the bottom layer in 2000–2005 were 1.41–2.2 mg/L, 0–0 µg/L, and 0–17.62 µg/L, respectively, and those in 2006–2011 were 1.53–3.2 mg/L, 0–281 µg/L, and 8.09–49.35 µg/L, respectively. In the future, data from various sources should be integrated into EIAs to identify not only the impacts of a single project, but also the overall impacts of multiple changes in an area. This method also could be used to provide basic data for continuous management of the marine environment.

Low Cost Sensor With IoT LoRaWAN Connectivity and Machine Learning-Based Calibration for Air Pollution Monitoring

Air pollution poses significant risk to environment and health. Air quality monitoring stations are often confined to a small number of locations due to the high cost of the monitoring equipment. They provide a low fidelity picture of the air quality in the city; local variations are overlooked. However, recent developments in low-cost sensor technology and wireless communication systems like Internet of Things (IoT) provide an opportunity to use arrayed sensor networks to measure air pollution, in real time, at a large number of locations. This article reports the development of a novel low-cost sensor node that utilizes cost-effective electrochemical sensors to measure carbon monoxide (CO) and nitrogen dioxide (NO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ) concentrations and an infrared sensor to measure particulate matter (PM) levels. The node can be powered by either solar-recharged battery or mains supply. It is capable of long-range, low power communication over public or private long-range wide area network (LoRaWAN) IoT network and short-range high data rate communication over Wi-Fi. The developed sensor nodes were co-located with an accurate reference CO sensor for field calibration. The low-cost sensors' data, with offset and gain calibration, show good correlation with the data collected from the reference sensor. Multiple linear regression (MLR)-based temperature and humidity correction results in mean absolute percentage error (MAPE) of 48.71% and R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.607 relative to the reference sensor's data. Artificial neural network (ANN)-based calibration shows the potential for significant further improvement with MAPE of 38.89% and R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.78 for leave-one-out cross-validation.

Moving sensors for improved estimation of dynamic structures: Experimental validation

In the monitoring of structural systems, the use of multiple high-end sensors may prove to be economically prohibitive. The alternative approach would be to use fewer devices capable of moving across the span of the structural system. In the proposed approach, a velocity sensor that is able to move across the spatial domain and obtain point-wise velocity measurements is combined to a novel dynamic observer. Based on the measured velocities, a state estimator is developed, the gain of which depends on the motion of the sensor. The motion of the sensor is defined using Lyapunov redesign methods and depends only on the estimation error at the current sensor position. The guidance policy is performance based and steers the sensor to spatial regions of the structure with larger estimation errors. The proposed approach is validated with a one-dimensional flexible structure, mathematically described by an Euler–Bernoulli partial differential equation. The moving sensor is realized through the use of a laser scanning vibrometer that provides both the moving measurements and additional measurements against which the proposed approach will be validated. Once measurements over a large number of locations are acquired, the experimental results are fed to the algorithm that selects the instantaneous sensor location. Experimental results for linear and nonlinear beam cases are presented to show the feasibility and robustness of the proposed approach.

A regionalisation approach for rainfall based on extremal dependence

To mitigate the risk posed by extreme rainfall events, we require statistical models that reliably capture extremes in continuous space with dependence. However, assuming a stationary dependence structure in such models is often erroneous, particularly over large geographical domains. Furthermore, there are limitations on the ability to fit existing models, such as max-stable processes, to a large number of locations. To address these modelling challenges, we present a regionalisation method that partitions stations into regions of similar extremal dependence using clustering. To demonstrate our regionalisation approach, we consider a study region of Australia and discuss the results with respect to known climate and topographic features. To visualise and evaluate the effectiveness of the partitioning, we fit max-stable models to each of the regions. This work serves as a prelude to how one might consider undertaking a project where spatial dependence is non-stationary and is modelled on a large geographical scale.

Carbon sources supporting Australia's most widely distributed freshwater fish, Nematalosa erebi (Günther) (Clupeidae: Dorosomatinae)

Both brown (detrital-based) and green (algal-based) food pathways support freshwater food webs, although the importance of either source may vary within species, regions and different phases of the flow regime. The bony bream (Nematalosa erebi Clupeidae: Dorosomatinae) is one of Australia’s most widely distributed freshwater fish species and is a key component of freshwater food webs, especially in northern Australia. We sought to better define the feeding habits of this species, previously classified as a detritivore, algivore or zooplanktivore (or combinations thereof), by undertaking meta-analyses of published accounts based on stomach content analysis and 13C and 15N stable isotope analysis. Stomach content analysis clearly indicated that detritus was the dominant food item, although benthic algae could be an important dietary component in some habitats (inland river flood plains) and during the wet season. Zooplankton were important for small fish (i.e. juveniles &amp;lt;100mm in length). When data were pooled across a large number of locations, stable isotope analysis indicated that detritus derived from terrestrial vegetation was better aligned isotopically with values for both adult and juvenile bony bream, whereas algae were comparatively 13C enriched, indicating the latter source was not the dominant contributor to the biomass of this species. However, using site-specific data and a regression approach, a significant relationship was revealed between algal carbon and that of large fish, suggesting that carbon derived from benthic algae contributed ~20% of the carbon of adult bony bream. Zooplankton contributed a similar amount. Zooplankton provided the majority of carbon for small fish. We contend that detritus derived from terrestrial vegetation is the likely remaining carbon source for large bony bream, and this interpretation was supported by the outcomes of multiple regression analyses. Although previous studies of aquatic food webs in northern Australia have emphasised the importance of high-quality algal basal resources, this study indicates that terrestrial sources may be important for some species and demonstrates the need to better consider the circumstances that cause biota to switch between different food sources.

Krigings over space and time based on latent low-dimensional structures

We propose a new nonparametric approach to represent the linear dependence structure of a spatio-temporal process in terms of latent common factors. Though it is formally similar to the existing reduced rank approximation methods, the fundamental difference is that the low-dimensional structure is completely unknown in our setting, which is learned from the data collected irregularly over space but regularly over time. Furthermore, a graph Laplacian is incorporated in the learning in order to take the advantage of the continuity over space, and a new aggregation method via randomly partitioning space is introduced to improve the efficiency. We do not impose any stationarity conditions over space either, as the learning is facilitated by the stationarity in time. Krigings over space and time are carried out based on the learned low-dimensional structure, which is scalable to the cases when the data are taken over a large number of locations and/or over a long time period. Asymptotic properties of the proposed methods are established. An illustration with both simulated and real data sets is also reported.

A Review of Learning Effect in Perimetry

Glaucoma is the second most common cause of visual impairment in the UK, with visual impairment registrations have increased by 22% since 2010. Glaucoma refers to a group of optic neuropathies leading to visual impairment and blindness. If glaucoma remains untreated, it may produce optic nerve damage, leading to vision loss. Consequently, visual field tests can be extremely valuable for glaucoma. At the same time, visual field assessment should be performed at baseline and periodically in the glaucoma follow-up or monitor the effectiveness of adopted therapeutic schemes. Any visual field test can be masked by one or more artefacts, which can either lead to the incorrect result of visual field loss or to the possible deterioration of existing loss. One of the most important factors is the perimetric learning effect that is present in almost all types of perimetry. To minimize the learning effect, we either have to conduct a practice test procedure, as a demonstration for the patient without collecting data, or to calculate and establish a learning index of the specific patient. By the establishment of such an index, assist the clinician in detecting possible masked or overestimated visual field defects or progression of glaucoma damage.&#x0D; Conclusion: Potentially, the intense data collection at a large number of locations throughout the field in a larger cohort of subjects (visually healthy and glaucomatous) would be required for a better index establishment. The incorporation of fatigue also may be required to form a robust index enough to simulate procedures of glaucoma prognosis. The low signal to noise ratio associated with perimetric testing suggests that improvements will always be difficult to make.

Heritability and Variability of Quality Parameters of Tomatoes in Outdoor Production.

The tomato is the most important vegetable globally. In England, outdoor tomatoes are widely grown by amateur gardeners, with the number of their allotment plots exceeding 150000. For instance, in Germany, only about 16% of tomato plants are cultivated according to organic standards, although these rates are expected to increase. Breeding for yield and fruit quality can increase resource efficiency. Therefore, we need to evaluate the variability of yield and fruit quality parameters, to calculate the heritability of these traits and to identify superior genotypes for organic outdoor tomato production and breeding. With these aims, we grew 24 tomato genotypes of diverse origins in a two-year field trial. The heritability of quality traits such as glucose, fructose, organic acid, and lycopene was high. The medium heritability for yield suggests that trials with a larger number of locations are needed for the reliable selection of this character. Negative correlations of sugar concentrations with fruit weight and of organic acid concentration with fruit weight and yield suggest trade-offs in breeding for larger fruits and higher yields. Breeding for increased lycopene content is not subject to these challenges; the concentrations of the primary metabolite sugars and organic acids are positively correlated.

Smart Insect Cameras

Recent studies have shown a worrying decline in the quantity and diversity of insects at a number of locations in Europe (Hallmann et al. 2017) and elsewhere (Lister and Garcia 2018). Although the downward trend that these studies show is clear, they are limited to certain insect groups and geographical locations. Most available studies (see overview in Sánchez-Bayo and Wyckhuys 2019) were performed in nature reserves, leaving rural and urban areas largely understudied. Most studies are based on the long-term collaborative efforts of entomologists and volunteers performing labor-intensive repeat measurements, inherently limiting the number of locations that can be monitored. We propose a monitoring network for insects in the Netherlands, consisting of a large number of smart insect cameras spread across nature, rural, and urban areas. The aim of the network is to provide a labor-extensive continuous monitoring of different insect groups. In addition, we aimed to develop the cameras at a relatively cheap price point so that cameras can be installed at a large number of locations and encourage participation by citizen science enthusiasts. The cameras are made smart with image processing, consisting of image enhancement, insect detection and species identification being performed, using deep learning based algorithms. The cameras take pictures of a screen, measuring ca. 30×40 cm, every 10 seconds, capturing insects that have landed on the screen (Fig. 1). Several screen setups were evaluated. Vertical screens were used to attract flying insects. Different screen colors and lighting at night, to attract night flying insects such as moths, were used. In addition two horizontal screen orientations were used (1) to emulate pan traps to attract several pollinator species (bees and hoverflies) and (2) to capture ground-based insects and arthropods such as beetles and spiders. Time sequences of images were analyzed semi-automatically, in the following way. First, single insects are outlined and cropped using boxes at every captured image. Then the cropped single insects in every image were preliminarily identified, using a previously developed deep-learning-based automatic species identification software, Nature Identification API (https://identify.biodiversityanalysis.nl). In the next step, single insects were linked between consecutive images using a tracking algorithm that uses screen position and the preliminary identifications. This step yields for every individual insect a linked series of outlines and preliminary identifications. The preliminary identifications for individual insects can differ between multiple captured images and were therefore combined into one identification using a fusing algorithm. The result of the algorithm is a series of tracks of individual insects with species identifications, which can be subsequently translated into an estimate of the counts of insects per species or species complexes. Here we show the first set of results acquired during the spring and summer of 2019. We will discuss practical experiences with setting up cameras in the field, including the effectiveness of the different set-ups. We will also show the effectiveness of using automatic species identification in the type of images that were acquired (see attached figure) and discuss to what extent individual species can be identified reliably. Finally, we will discuss the ecological information that can be extracted from the smart insect cameras.