A new approach of the modified method of the analysis of innovative trends (ITA) was applied to the categories of maximum daily rains coming from 41 pluviometric stations of Macta (North West Algeria) for the period 1970-2010. Six categories of maximum daily rainfall were considered: light: 0–4 mm/day (A); mild to moderate: 4–16 mm/day (B); moderate-heavy: 16–32 mm/day (C₁); heavy: 32–64 mm/day (C₂); heavy torrential: 64–128 mm/day (D₁); and torrential: greater than or equal to 128 mm/day (D₂). The results show that during the second period (1990-2010), the increasing trends were detected in the middle categories B (most frequent: 20.78%), C₁ (17.85%), C₂ (20.21%) and D₁ (4.92%). In addition, migratory movements and rainy events towards the higher categories were detected (in particular towards C₁ (8.84%)). These results show the increase in extreme rains in the Macta watershed during the period (1990-2010).

The quantity of precipitation as well as its distribution varies in time, space, and even in small areas. This temporal and/or spatial variability is due to the support of several quite complex climatic and physiographic factors. The description and the prediction of this variability is a fundamental requirement in a wide variety of human activities, as well as in the elaboration and the design of hydraulic projects. The objective of this work is to study the applicability of Kalman filter (KF) technique for the modeling and prediction of annual or monthly rainfall amounts in the Cheliff watershed, as well as the assessment of the prediction error. The major advantage of KF is to provide with the prediction error covariance an indicator of the filter accuracy. In addition, its algorithm works in the temporal domain with a recursive nature and has an optimal estimator in the least squares concept. Another aspect of its optimality is the incorporation of all the available information on the system, measurements and errors in an adaptive operator, which is reset each time as a new measurement is available. For the implementation of this filter, time series of monthly and annual rainfall data registered over a period of 51 years (1959–2009) in 39 precipitation stations are studied in the Cheliff watershed and the obtained results are quite satisfactory.

Floods represent one of the most destructive natural hazards worldwide, causing significant environmental degradation, economic losses, and social disruption. This study aims to delineate flood-susceptible zones within the Mekerra Valley Basin, located in northwestern Algeria, by applying an integrated approach that combines remote sensing techniques, Geographic Information Systems (GIS), and the Analytical Hierarchy Process (AHP), a widely used multi-criteria decision analysis method. Ten conditioning factors influencing flood occurrence were considered, including elevation, slope, precipitation, Topographic Wetness Index (TWI), Normalized Difference Vegetation Index (NDVI), drainage density, distance from streams, Stream Power Index (SPI), soil characteristics, and Land Use/Land Cover (LULC). The AHP method was employed to determine the relative importance of each factor, and the resulting weights were integrated through GIS-based weighted overlay analysis to generate a spatial flood susceptibility map. The findings classified the basin into five susceptibility categories, ranging from very low to very high flood vulnerability. The highest susceptibility areas were mainly associated with low-lying terrain, gentle slopes, high drainage accumulation, and proximity to river networks, whereas elevated regions with steeper slopes exhibited lower flood susceptibility. Model performance assessment using the Receiver Operating Characteristic (ROC) curve yielded an Area Under the Curve (AUC) value of 0.915, indicating a high predictive capability. Furthermore, the reliability of the model was confirmed through validation using historical flood events recorded between 2017–2019 and in 2022, showing a strong spatial correspondence between predicted susceptibility zones and observed flood occurrences. The resulting flood susceptibility map provides an effective decision-support tool for flood hazard assessment, sustainable land-use planning, and the development of targeted disaster risk reduction strategies in the Mekerra Valley Basin.