The expansion of African cities leads to the occupation of peripheral urban areas without respecting planning rules. The Greater Lomé (Togo) is no exception to this phenomenon of high-speed horizontal spreading, which causes recurrent flooding. The objective of this research is to understand the spatio-temporal changes in the dynamics of built-up areas in Greater Lomé. The methodology used is based on the analysis of Landsat images from the years 2007, 2012, 2016, and 2020 coupled with direct field observations and a literature review. The results showed an increase in residential areas to the detriment of the other land use/cover types. Estimated at 15,481 ha in 2007, the built-up area reached 35,521 ha in 2020, an increase of 33% to the detriment of vegetation and cultivation areas. This increase was marked by constructions in the floodplain of the Zio River. The field surveys revealed an increase in the density of most of the agglomerations. From 1863 ha in 2007, they increased to 14,485 ha in 2020, an increase of 12,622 ha or approximately 33%. These results indicate that attention needs to be paid to both the planning and control of the development of spaces in the outlying areas of Greater Lomé.

In Togo, and especially in the peri-urban areas of Greater Lomé, flood hazards and their effects on communities and ecosystems have increased in recent years due to global changes. This is predominantly due to increasing urbanization and the changing climate. This study explores the application of remote sensing, geographic information system (GIS) and analytical hierarchy processes (AHP) for flood hazard mapping in Greater Lomé. Maps of seven (7) drivers of flood hazards, namely flow accumulation, rainfall intensity, geology and soil, land use, slope, elevation and distance from the drainage network, were prepared by integrating remote sensing data into a GIS environment. The weight of each factor was estimated via the AHP method and used in the estimation of the flood hazard index (FHI). The results show that there is a high or very high susceptibility of flooding in 53.4% of the Greater Lomé suburbs, while there is a moderate to low susceptibility of flooding in about 46.59% of it, and 0.01% of the study area. The generated map is a useful tool for decision-makers and practitioners in charge of flood risk and disaster management when developing plans to reduce the risk of both current and future floods in the study area.

Climate and land use changes have exacerbated the frequency and intensity of floods in West Africa. In Togo, the communities of the Zio river sub-basin are regularly affected by flood hazard with severe socio-economic consequences. The main objective of this article is to contribute to the improvement of the resilience of these populations through a better understanding of their perceptions and coping strategies to flood hazards. To achieve this goal, a mixed method consisting of household surveys, focus groups and direct observation is used. In addition, the data collected at the household level focused on the perceptions of the flood (origin and cause), the socio-economic impacts as well as the coping strategies. The results indicate that more than 80% of the households are affected by floods and the main negative consequences are the destruction of houses (40.8%), and interruption of socio-economic activities (58.4%). In addition, the efforts to cope with the severe impacts of floods include embankment (71.6%), storage on occasional shelves (52.3%), digging ditches (26.6% and the construction of cement houses with fairly raised foundations (13.1%). This study shows that is important to integrate indigenous knowledge and local population’s opinions in the flood risk management plans in order to increase their resilience.

The Upper Mono Basin Valley (UMBV) in Togo faces recurrent flooding hazards. This study assesses spatial flood susceptibility using an integrated approach combining Geographic Information Systems (GISs), Multi-Criteria Decision Making (MCDM), and the Analytic Hierarchy Process (AHP). Eight factors were weighted according to their influence: accumulation flow, annual precipitation, soil permeability, land use/land cover, slope, elevation, distance from drainage networks, and drainage network density. With a consistency ratio of 0.052, the AHP method proved coherent and enabled the development of a normalized Flood Hazard Index (FHI). Results revealed accumulation flow (weight = 0.33), distance to drainage networks (0.18), and network density (0.16) as the most critical drivers, while precipitation and soil permeability are secondary. Spatial classification revealed heterogeneity: 55% (871,046 ha) of the UMBV has very low susceptibility, while 1% (10,034 ha) is highly vulnerable, mainly in Est-Mono, Ogou, Anié, Tchamba, and Tchaoudjo. In contrast, Blitta and Sotouboua show lower vulnerability due to higher altitudes. This reveals that the UMBV is relatively less prone to flooding. The comparison of data from 28 focus groups in 14 municipalities with the flood susceptibility map shows a strong concordance between local perceptions and the mapping (r = 0.805, p < 0.001). These findings highlight the need for differentiated territorial strategies integrating physical parameters, land use dynamics, and community risk perceptions to strengthen flood risk management in the UMBV.

TAccurate short-term electricity consumption forecasting is essential for operational planning, reserve allocation, and energy management in modern power systems. This study investigates the performance of recurrent and hybrid attention-based deep learning architectures for short-term electricity consumption forecasting, a sase study in Togo. The proposed framework integrates electricity consumption, meteorological, demographic, and temporal information in order to capture both intrinsic temporal dependencies and exogenous influences affecting electricity demand. Four forecasting architectures were evaluated using a weekly temporal window of 168 hours: LSTM-only, LSTM-decoder, LSTM-attention, and a hybrid LSTM--Multi-Head Attention--LSTM model. The experiments included multi-seed evaluation, ablation study, robustness analysis, and statistical comparison using the Wilcoxon signed-rank test. The results show that all models achieved extremely high forecasting accuracy, with coefficients of determination exceeding 0.9998 and MAPE values below 0.004%. The hybrid architectures slightly improved average forecasting performance, while the standalone LSTM model remained highly competitive. The robustness analysis revealed strong sensitivity to noisy inputs but moderate degradation under missing-data conditions. Statistical analysis indicated that the performance differences between architectures were not statistically significant at the 5% level. Overall, the study demonstrates that recurrent deep learning architectures provide reliable and operationally relevant solutions for electricity consumption forecasting in highly structured energy demand series.

Accurate short-term forecasting of meteorological variables remains a major challenge in tropical regions characterized by strong diurnal variability, seasonal transitions, intermittent atmospheric fluctuations, and highly non-stationary temporal dynamics. This study proposes a hybrid deep-learning framework combining the Stationary Wavelet Transform (SWT), Multi-Head Attention (MHA) and Long Short-Term Memory (LSTM) networks’ forecasting layers for multivariate meteorological prediction over selected locations in Togo. The proposed architecture simultaneously forecasts temperature at 2 m, relative humidity, and wind speed using hourly ERA5 atmospheric variables enriched with lagged, rolling-statistical, and cyclic temporal features. SWT decomposition extracts multi-scale atmospheric structures while preserving temporal alignment, the attention mechanism dynamically emphasizes informative temporal sub-sequences, and the LSTM layers improve medium-range temporal dependency modeling. The forecasting task was reformulated under a strictly causal prediction protocol for 24 h and 72 h forecasting horizons. The proposed framework achieved competitive forecasting performance with MAE = 2.99, RMSE = 5.22, and ${\mathrm{R}}^2=0.77$ for the 24 h horizon, and MAE = 3.44, RMSE = 6.11, and ${\mathrm{R}}^2=0.73$ for the 72 h horizon. Temperature forecasting exhibited the highest stability, whereas wind-speed prediction remained more challenging because of stronger atmospheric intermittency. The results demonstrate the relevance of multi-scale temporal representation learning for ERA5-based tropical meteorological forecasting while highlighting important methodological and operational limitations.