Application of Deep Learning to Community-Science-Based Mosquito Monitoring and Detection of Novel Species.

Mosquito-borne diseases account for human morbidity and mortality worldwide, caused by the parasites (e.g., malaria) or viruses (e.g., dengue, Zika) transmitted through bites of infected female mosquitoes. Globally, billions of people are at risk of infection, imposing significant economic and public health burdens. As such, efficient methods to monitor mosquito populations and prevent the spread of these diseases are at a premium. One proposed technique is to apply acoustic monitoring to the challenge of identifying wingbeats of individual mosquitoes. Although researchers have successfully used wingbeats to survey mosquito populations, implementation of these techniques in areas most affected by mosquito-borne diseases remains challenging. Here, methods utilizing easily accessible equipment and encouraging community-scientist participation are more likely to provide sufficient monitoring. We present a practical, community-science-based method of monitoring mosquito populations using smartphones. We applied deep-learning algorithms (TensorFlow Inception v3) to spectrogram images generated from smartphone recordings associated with six mosquito species to develop a multiclass mosquito identification system, and flag potential invasive vectors not present in our sound reference library. Though TensorFlow did not flag potential invasive species with high accuracy, it was able to identify species present in the reference library at an 85% correct identification rate, an identification rate markedly higher than similar studies employing expensive recording devices. Given that we used smartphone recordings with limited sample sizes, these results are promising. With further optimization, we propose this novel technique as a way to accurately and efficiently monitor mosquito populations in areas where doing so is most critical.

Effects of Climatic Change on the Potential Distribution of Lycoriella Species (Diptera: Sciaridae) of Economic Importance.

Lycoriella species (Sciaridae) are responsible for significant economic losses in greenhouse production (e.g., mushrooms, strawberries, and nurseries). The current distributions of species in the genus are restricted to cold-climate countries. Three species of Lycoriella are of particular economic concern in view of their ability to invade areas in countries across the Northern Hemisphere. We used ecological niche models to determine the potential for range expansion under future climate change scenarios (RCP 4.5 and RCP 8.5) in the distribution of these three species of Lycoriella. Stable environmental suitability under climate change was a dominant theme in these species; however, potential range increases were noted in key countries (e.g., USA, Brazil, and China). Our results illustrate the potential for range expansion in these species in the Southern Hemisphere, including some of the highest greenhouse production areas in the world.

Climate change impacts on Anopheles (K.) cruzii in urban areas of Atlantic Forest of Brazil: Challenges for malaria diseases.

Around 27% of South Americans live in central and southern Brazil. Of 19,400 human malaria cases in Brazil in 2018, some were from the southern and southeastern states. High abundance of malaria vectors is generally positively associated with malaria incidence. Expanding geographic distributions of Anopheles vector mosquito species (e.g. A. cruzii) in the face of climate change processes would increase risk of such malaria transmission; such risk is of particular concern in regions that hold human population concentrations near present limits of vector species' geographic distributions. We modeled effects of likely climate changes on the distribution of A. cruzii, evaluating two scenarios of future greenhouse gas emissions for 2050, as simulated in 21 general circulation models and two greenhouse gas scenarios (RCP 4.5 and RCP 8.5) for 2050. We tested 1305 candidate models, and chose among them based on statistical significance, predictive performance, and complexity. The models closely approximated the known geographic distribution of the species under current conditions. Under scenarios of future climate change, we noted increases in suitable area for the mosquito vector species in São Paulo and Rio de Janeiro states, including areas close to 30 densely populated cities. Under RCP 8.5, our models anticipate areal increases of >75% for this important malaria vector in the vicinity of 20 large Brazilian cities. We developed models that anticipate increased suitability for the mosquito species; around 50% of Brazilians reside in these areas, and ∼89% of foreign tourists visit coastal areas in this region. Under climate change thereefore, the risk and vulnerability of human populations to malaria transmission appears bound to increase.

Climate change implications for the distribution of the babesiosis and anaplasmosis tick vector, Rhipicephalus (Boophilus) microplus.

Climate change ranks among the most important issues globally, affecting geographic distributions of vectors and pathogens, and inducing losses in livestock production among many other damaging effects. We characterized the potential geographic distribution of the ticks Rhipicephalus (Boophilus) microplus, an important vector of babesiosis and anaplasmosis globally. We evaluated potential geographic shifts in suitability patterns for this species in two periods (2050 and 2070) and under two emissions scenarios (RCPs 4.5 and 8.5). Our results anticipate increases in suitability worldwide, particularly in the highest production areas for cattle. The Indo-Malayan region resulted in the highest cattle exposure under both climate change projections (2050), with increases in suitability of > 30%. This study illustrates how ecological niche modeling can be used to explore probable effects of climate change on disease vectors, and the possible consequences on economic dimensions.

Inventory statistics meet big data: complications for estimating numbers of species.

We point out complications inherent in biodiversity inventory metrics when applied to large-scale datasets. The number of units of inventory effort (e.g., days of inventory effort) in which a species is detected saturates, such that crucial numbers of detections of rare species approach zero. Any rare errors can then come to dominate species richness estimates, creating upward biases in estimates of species numbers. We document the problem via simulations of sampling from virtual biotas, illustrate its potential using a large empirical dataset (bird records from Cape May, NJ, USA), and outline the circumstances under which these problems may be expected to emerge.

Major challenges for correlational ecological niche model projections to future climate conditions.

Species-level forecasts of distributional potential and likely distributional shifts, in the face of changing climates, have become popular in the literature in the past 20 years. Many refinements have been made to the methodology over the years, and the result has been an approach that considers multiple sources of variation in geographic predictions, and how that variation translates into both specific predictions and uncertainty in those predictions. Although numerous previous reviews and overviews of this field have pointed out a series of assumptions and caveats associated with the methodology, three aspects of the methodology have important impacts but have not been treated previously in detail. Here, we assess those three aspects: (1) effects of niche truncation on model transfers to future climate conditions, (2) effects of model selection procedures on future-climate transfers of ecological niche models, and (3) relative contributions of several factors (replicate samples of point data, general circulation models, representative concentration pathways, and alternative model parameterizations) to overall variance in model outcomes. Overall, the view is one of caution: although resulting predictions are fascinating and attractive, this paradigm has pitfalls that may bias and limit confidence in niche model outputs as regards the implications of climate change for species' geographic distributions.

Diversidad de Vanilla spp. (Orchidaceae) y sus perfiles bioclimáticos en México / Diversity and bioclimatic profiles of Vanilla spp. (Orchidaceae) in Mexico

Resumen: El género Vanilla comprende alrededor de 110 especies, distribuidas en las partes tropicales del mundo, y es el continente americano donde se encuentra la mayor cantidad de especies reportadas. En México, el cultivo de la vainilla está relacionado con diversas culturas como la totonaca, maya, chinanteca, mazateca, entre otras. En la actualidad, este cultivo presenta factores condicionantes tales como: técnicos, económicos, sociales, ecológicos y climáticos, que limitan su producción y la conservación de especies silvestres y cultivadas, por lo que es necesario, conocer su estado actual, en relación con su diversidad, así como algunos de los principales indicadores del perfil bioclimático de cada una de las especies, que ayuden en la toma de decisiones para su conservación y mejoramiento genético. Durante 2008, se realizaron consultas a herbarios de IPN, MEXU, XAL, base de datos de la Red Mundial de Información sobre la Biodiversidad de la CONABIO (REMIB), Global Biodiversity Information Facility (GBIF) y datos de accesiones vivas del banco de germoplasma de vainilla de la BUAP, formado del 2008 al 2014. Se realizaron mapas de distribución mediante un sistema de información geográfica. Se obtuvo el perfil bioclimático de cada especie considerando 19 variables de World Clim, y altitud a una resolución espacial de aproximadamente 1 Km2. Se calcularon las medias, desviaciones estándar y varianzas de las 19 variables en cada uno de los puntos registrados. Se obtuvieron los intervalos de las condiciones ambientales extremas (mínimo, promedio y máximo) para cada una de las especies de vainilla. Para determinar las variables de mayor importancia en la distribución de las especies se realizó un análisis de componentes principales, y a las variables que resultaron significativas se les realizó pruebas de Kruskal-Wallis y Dunn. Los resultados indicaron que en México se tienen registros de V. planifolia, V. pompona, V. insignis, V. inodora V. odorata, V. cribbiana y V. sprucei distribuidas en nueve estados. V. planifolia presentó intervalos amplios de temperatura y precipitación; V. pompona, V. odorata, V. insignis y V. inodora tuvieron intervalos intermedios. La amplitud de los datos extremos de cada especie puede considerarse para ubicar los sitios donde se puedan llevar a cabo estrategias regionales de conservación ex situ y el establecimiento de cultivos. El perfil bioclimático encontrado permite inferir de manera indirecta la condición genética de cada especie que podría ser utilizada en programas de mejoramiento genético como: la alta altitud y tolerancia a bajas temperaturas (V. odorata), la tolerancia a altas temperaturas (V. inodora) y tolerancia a baja precipitación (V. odorata, V. pompona y V. planifolia).

Abstract: The genus Vanilla comprises around 110 species distributed throughout Earth's tropical regions, with the largest number of reported species growing in the American continent. Vanilla farming is associated with many Mexican cultures such as the Totonac, Mayan, Chinantec, and Mazatec, among others. Currently, this crop is threatened by technical, social, ecological, and climatic conditioning factors, limiting its production and the preservation of wild and cultivated species. It is therefore necessary to ascertain the current diversity status of each of these species, as well as some of their main bioclimatic profile indicators, in order to help decision-making, aimed at preserving and genetically improve these species. During 2008, we gathered data from IPN, MEXU, and XAL herbaria, as well as from CONABIO's World Information Network on Biodiversity (REMIB), the Global Biodiversity and Information Facility (GBIF), and we also used data from live access to BUAP's vanilla germplasm bank, obtained between 2008 and 2014. Distribution maps were generated using a geographical information system. Bioclimatic profiles for each species were obtained considering 19 WorldClim variables and altitude at a spatial resolution of approximately 1 Km2. Variance, Mean, and standard deviation for each of the 19 variables were calculated at each of the registered points. Extreme environmental condition intervals (minimum, average, and maximum) were also obtained. In order to determine the most important distribution variables of the species, we performed a principal component analysis and carried out Kruskal-Wallis and Dunn's tests on the variables identified as significant. Results indicated records for V. planifolia, V. pompona, V. insignis, V. inodora V. odorata, V. cribbiana, and V. sprucei in Mexico, distributed throughout nine states in the country. V. planifolia presented wide intervals of temperature and rain precipitation, while V. pompona, V. odorata, V. insignis and V. inodora presented intermediate intervals. The amplitudes of extreme data for each species can be considered in locating areas where ex situ regional preservation strategies could be put in place, as well as in establishing areas for cultivation. The bioclimatic profile we found, allows for an indirect inference of each species' genetic condition, which could be used in genetic improvement programs; for instance, V. odorata grows at high altitudes and tolerates low temperatures, while V. inodora tolerates high temperatures, and V. odorata, V. pompona and V. planifolia tolerate low rain precipitation. Rev. Biol. Trop. 65 (3): 975-987. Epub 2017 September 01.

Freshwater fish's spatial patterns in isolated water springs in North-eastern Mexico.

The Media Luna lake-spring was selected as representative of all thermal or no thermal springs in the zone of Valley of Rioverde, a semi-arid vegetation in the North-eastern of Mexico. This system is inhabited by 11 fish species, of which six are native. Four of the native species are endemic to the region and threatened due to touristic pressure and to the introduction of exotic species. The objectives were to determine the characteristics that influence the spatial distribution of the fish species, to analyze their spatial distribution patterns, and to describe the relationships between the different species. The general aim was to establish some basis for the conservation of these fish communities and their habitat. Several sessions were initiated in 1992 through direct observation. Later, between 1998 and 1999 five systematically seasonal sampling sessions were conducted (54 subaquatic transects/session). Finally, the data was updated by sampling in summer 2002 and winter 2006. Through the analysis was performed only for endemics of the region, like Ataeniobius toweri Meek, Cualac tessellatus Miller, Cichlasoma bartoni Bean and C. labridens Pellegrin, in at least one life stage, showed correlation with habitat variables or with other species. For these species, patterns of spatial aggregation and association with other species were observed. These results show a certain degree of specialization of endemic species to some microhabitat characteristics, as well as a significant interaction with other native species which they coexist. In addition, some significant relations between endemic and alien species suggest an antagonist relation. Management actions focused in the touristic use of the spring represent the main threat for these species, followed by an adequate management of exotic species. This study provides basis for future responsible management of these wetlands, where tourism and conservation can be combined.

Freshwater fish’s spatial patterns in isolated water springs in North-eastern Mexico

The Media Luna lake-spring was selected as representative of all thermal or no thermal springs in the zone of Valley of Rioverde, a semi-arid vegetation in the North-eastern of Mexico. This system is inhabited by 11 fish species, of which six are native. Four of the native species are endemic to the region and threatened due to touristic pressure and to the introduction of exotic species. The objectives were to determine the characteristics that influence the spatial distribution of the fish species, to analyze their spatial distribution patterns, and to describe the relationships between the different species. The general aim was to establish some basis for the conservation of these fish communities and their habitat. Several sessions were initiated in 1992 through direct observation. Later, between 1998 and 1999 five systematically seasonal sampling sessions were conducted (54 subaquatic transects/session). Finally, the data was updated by sampling in summer 2002 and winter 2006. Through the analysis was performed only for endemics of the region, like Ataeniobius toweri Meek, Cualac tessellatus Miller, Cichlasoma bartoni Bean and C. labridens Pellegrin, in at least one life stage, showed correlation with habitat variables or with other species. For these species, patterns of spatial aggregation and association with other species were observed. These results show a certain degree of specialization of endemic species to some microhabitat characteristics, as well as a significant interaction with other native species which they coexist. In addition, some significant relations between endemic and alien species suggest an antagonist relation. Management actions focused in the touristic use of the spring represent the main threat for these species, followed by an adequate management of exotic species. This study provides basis for future responsible management of these wetlands, where tourism and conservation can be combined. Rev. Biol. Trop. 58 (1): 413- 426. Epub 2010 March 01.

Este estudio se enfocó en los peces de un sistema dulceacuícola en el noreste de México, donde habitan seis especies nativas (cuatro endémicas amenazadas) y cinco exóticas. El objetivo fue establecer los patrones que determinan la distribución espacial, así como las interrelaciones de las especies. Los datos se basan en sesiones de observación directa desde 1992; entre 1998 y 1999 se llevó a cabo un muestreo sistematizado mediante transectos subacuáticos en cinco sesiones estacionales (54 transectos/ sesión), con sesiones adicionales en 2002 y 2006. Sólo las especies endémicas de la región: Ataeniobius toweri, Cualac tessellatus, Cichlasoma bartoni y C. labridens, en al menos un estadio de vida, mostraron correlación significativa con variables del hábitat o con otras especies. También mostraron patrones de agregación y asociación con otras especies nativas o introducidas. Existe especialización de los endémicos a las condiciones de su microhábitat, así como interacciones significativas con otras especies. Acciones inadecuadas para promover el turismo representaron la mayor amenaza por destrucción del hábitat, endémicos tales como A, toweri y C. bartoni enfrentan solapamiento con especies introducidas, sobre todo en sus sitios de crianza. Este estudio aporta bases para un manejo responsable de estos humedales, donde turismo y conservación pueden combinarse.