Machine and deep learning approaches to understand and predict habitat suitability for seabird breeding.

The way animals select their breeding habitat may have great impacts on individual fitness. This complex process depends on the integration of information on various environmental factors, over a wide range of spatiotemporal scales. For seabirds, breeding habitat selection integrates both land and sea features over several spatial scales. Seabirds explore these features prior to breeding, assessing habitats' quality. However, the information-gathering and decision-making process by seabirds when choosing a breeding habitat remains poorly understood. We compiled 49 historical records of larids colonies in Cuba from 1980 to 2020. Then, we predicted potentially suitable breeding sites for larids and assessed their breeding macrohabitat selection, using deep and machine learning algorithms respectively. Using a convolutional neural network and Landsat satellite images we predicted the suitability for nesting of non-monitored sites of this archipelago. Furthermore, we assessed the relative contribution of 18 land- and marine-based environmental covariates describing macrohabitats at three spatial scales (i.e. 10, 50 and 100 km) using random forests. Convolutional neural network exhibited good performance at training, validation and test (F1-scores >85%). Sites with higher habitat suitability (p > .75) covered 20.3% of the predicting area. Larids breeding macrohabitats were sites relatively close to main islands, featuring sparse vegetation cover and high chlorophyll-a concentration at sea in 50 and 100 km around colonies. Lower sea surface temperature at larger spatial scales was determinant to distinguish the breeding from non-breeding sites. A more comprehensive understanding of the seabird breeding macrohabitats selection can be reached from the complementary use of convolutional neural networks and random forest models. Our analysis provides crucial knowledge in tropical regions that lack complete and regular monitoring of seabirds' breeding sites.

Quantitative imagery analysis of spot patterns for the three-haplogroup classification of Triatoma dimidiata (Latreille, 1811) (Hemiptera: Reduviidae), an important vector of Chagas disease.

BACKGROUND: Spots and coloring patterns evaluated quantitatively can be used to discriminate and identify possible cryptic species. Species included in the Triatoma dimidiata (Reduviidae: Triatominae) complex are major disease vectors of Chagas disease. Phylogenetic studies have defined three haplogroups for Mexico and part of Central America. We report here our evaluation of the possibility of correctly discriminating these three T. dimidiata haplogroups using the pattern of the dorsal spots. METHODS: Digital images of the dorsal region of individuals from the three haplogroups were used. Image processing was used to extract primary and secondary variables characterizing the dorsal spot pattern. Statistical analysis of the variables included descriptive statistics, non-parametric Kruskal-Wallis tests, discriminant function analysis (DFA) and a neural classification network. RESULTS: A distinctive spot pattern was found for each haplogroup. The most differentiated pattern was presented by haplogroup 2, which was characterized by its notably larger central spots. Haplogroups 1 and 3 were more similar to each other, but there were consistent differences in the shape and orientation of the spots. Significant differences were found among haplogroups in almost all of the variables analyzed, with the largest differences seen for relative spot area, mean relative area of central spots, central spots Feret diameter and lateral spots Feret diameter and aspect ratio. Both the DFA and the neural network had correct discrimination values of > 90%. CONCLUSIONS: Based on the results of this analysis, we conclude that the spot pattern can be reliably used to discriminate among the three haplogroups of T. dimidiata in Mexico, and possibly among triatomine species.

[Population size and behavior pattern of Grus canadensis nesiotes (Aves: Gruidae) in two localities of Cuba]. / Tamaño poblacional y patron de conducta de Grus Canadensis (Aves: Gruidae) en dos localidades de Cuba.

The availability of information on species abundance in the Neotropic is insufficient, and this prevents the execution of precise analysis and the definition of adequate conservation strategies for endemic and threatened species. This study aimed to analyze the population size of the endemic and threatened subspecies Grus canadensis nesiotes. For this, a simultaneous census was undertaken in 24 count stations in Isla de la Juventud (IJ) and 32 stations in Ciego de Ávila (CA), Cuba, during two consecutive days between 2008 and 2010. Abundance and behavior pattern (instantaneous method) were analyzed by habitat type, to help understand how cranes modify their behavioral pattern when the natural habitat is changed. Flocks in IJ had three individuals, and between 1.9 ± 1.5 and 2.8 ± 1.5 in CA. Population size in IJ was 164 individuals, and in CA of 137, 141 and 168 individuals for the 2008-2010 period, respectively. The counting efficacy was high (IJ: 91 %; CA: 81-87 %) and the numerical concordance was intermediate (IJ: 45.4 %; CA: 72 %). When comparing the habitat type, the abundance was higher in natural savannahs (83), followed by coastal flats (59), pines (23) and cattle pastures (7) in IJ; while in CA, marsh grasslands hosted the greatest abundance for the three years period (130; 120; 112), followed by grassland with palms (2; 17; 51) and cattle pastures (5; 4; 5). The cranes were fed more in cattle pastures and were more alert in natural savannas and marsh grasslands. The frequency of feeding and alert behaviors was different from the natural savannah/coastal flats and natural savannah/cattle pastures combinations in IJ. For CA, differences were found between marsh grasslands and marsh grasslands with palms. The population size increased by management strategies adopted in CA; nevertheless, might be affected by habitat loss associated with invasive alien plants in IJ. We propose the maintenance of prescribed fire in marsh grasslands under protection regime, as a strategy for long-term management to contribute with population growth.