Classifying and Mapping Cultural Ecosystem Services Using Artificial Intelligence and Social Media Data.

Quantifying and mapping cultural ecosystem services are complex because of their intangibility. Data from social media, such as geo-tagged photographs, has been proposed for mapping cultural use or appreciation of ecosystems. However, manual content analysis and classification of large numbers of photographs is time-consuming. The potential of deep learning for automating the analysis of crowdsourced social media content is still being explored in CES research. Here, we use a new deep learning model for automating the classification of natural and human elements relevant to CES from Flickr images. This approach applies a convolutional neural network architecture to analyze over 29,000 photographs from the Lithuanian coast and uses hierarchical clustering to group these photographs. The accuracy of the classification was assessed by comparison with manual classification. Over 37% of the photographs were taken for the landscape appreciation class, and 28% of the photographs were taken of nature, of animals or plants, which represent the nature appreciation class. The main clusters were identified in urban areas, more precisely in the main coastal cities of Lithuania. The distribution of the nature photographs was concentrated around particular natural attractions, and they were more likely to occur in parks and natural reserves with high levels of vegetation and animal cover. This approach that was developed for clustering the photographs was accurate and saved approximately 100 km of manual work. The method demonstrates how analyzing large numbers of digital photographs expands the analytical toolbox available to researchers and allows the quantification and mapping of CES at large geographical scales. Automated assessment and mapping of cultural ecosystem services could be used to inform urban planning and improve nature reserve management.

Harnessing LSTM and XGBoost algorithms for storm prediction.

Storms can cause significant damage, severe social disturbance and loss of human life, but predicting them is challenging due to their infrequent occurrence. To overcome this problem, a novel deep learning and machine learning approach based on long short-term memory (LSTM) and Extreme Gradient Boosting (XGBoost) was applied to predict storm characteristics and occurrence in Western France. A combination of data from buoys and a storm database between 1996 and 2020 was processed for model training and testing. The models were trained and validated with the dataset from January 1996 to December 2015 and the trained models were then used to predict storm characteristics and occurrence from January 2016 to December 2020. The LSTM model used to predict storm characteristics showed great accuracy in forecasting temperature and pressure, with challenges observed in capturing extreme values for wave height and wind speed. The trained XGBoost model, on the other hand, performed extremely well in predicting storm occurrence. The methodology adopted can help reduce the impact of storms on humans and objects.

Monitoring rural-urban transformation in the coastal region of Rabat-Sale-Kenitra, Morocco.

Worldwide urbanization drives rural-urban transformation (RUT) which has major consequences in many countries of the Global South where there is an urgent need to better understand and manage the underlying processes and consequences for ecosystem services. To fill existing knowledge gaps on the extent and time course of RUT in Morocco, this study focused on (i) analyzing the spatial patterns of rural-urban transformation in the Rabat-Sale-Kenitra (RSK) region from 1972 to 2020, (ii) identifying key mechanisms of change, and (iii) defining the main driving forces behind the spatial transformation patterns. To this end, we processed data of the Landsat free archive, historical grayscale Corona images, and nighttime lights datasets on Google Earth Engine (GEE) using machine learning classifiers and LandTrendr spectral-temporal segmentation algorithms. With an overall accuracy (OA) ranging from 88-95%, the results revealed that during the study period the RSK region experienced a 473% growth of horizontal built-up reflected in an area increase from 63.4 km2 to 299.9 km2. The main changes occurred along the Kenitra-Rabat-Temara axis and in central cities connected to the main road network. The horizontal expansion of large and medium-sized cities led to the formation of a Rural-Urban Interface (RUI) on the outskirts. The urban sprawl of some cities has affected the surrounding rural lands within the RUI. Environmental, social, economic, and political forces have interacted in shaping the changes in rural-urban landscapes.

Environmental, Climatic, and Parasite Molecular Factors Impacting the Incidence of Cutaneous Leishmaniasis Due to Leishmania tropica in Three Moroccan Foci.

Cutaneous leishmaniasis (CL) occurring due to Leishmania tropica is a public health problem in Morocco. The distribution and incidence of this form of leishmaniasis have increased in an unusual way in the last decade, and the control measures put in place are struggling to slow down the epidemic. This study was designed to assess the impact of climatic and environmental factors on CL in L. tropica foci. The data collected included CL incidence and climatic and environmental factors across three Moroccan foci (Foum Jemaa, Imintanout, and Ouazzane) from 2000 to 2019. Statistical analyses were performed using the linear regression model. An association was found between the occurrence of CL in Imintanout and temperature and humidity (r2 = 0.6076, df = (1.18), p-value = 3.09 × 10-5; r2 = 0.6306, df = (1.18), p-value = 1.77 × 10-5). As a second objective of our study, we investigated the population structure of L.tropica in these three foci, using the nuclear marker internal transcribed spacer 1 (ITS1). Our results showed a low-to-medium level of geographic differentiation among the L.tropica populations using pairwise differentiation. Molecular diversity indices showed a high genetic diversity in Foum Jemaa and Imintanout; indeed, 29 polymorphic sites were identified, leading to the definition of 13 haplotypes. Tajima's D and Fu's F test statistics in all populations were not statistically significant, and consistent with a population at drift-mutation equilibrium. Further analysis, including additional DNA markers and a larger sample size, could provide a more complete perspective of L. tropica's population structure in these three regions. In addition, further research is needed to better understand the impact of climatic conditions on the transmission cycle of Leishmania, allowing both for the development of effective control measures, and for the development of a predictive model for this parasitosis.