Weather radars are vital for precipitation nowcasting, offering high-resolution inputs. This study assesses radar-based rainfall nowcasting over the Core Monsoon Zone using a dual-polarization C-band radar at Silkheda. Utilizing the Python framework of Short-term Ensemble Prediction System (PySTEPS), we apply three optical flow algorithms—Lucas-Kanade (LK), Dynamic and Adaptive Radar Tracking of Storms (DARTS), and Variational Echo Tracking (VET)—to assess nowcasting accuracy under five different rain categories based on daily rain accumulation: light, moderate, heavy, very heavy, and extremely heavy, as defined by the India Meteorological Department. Probabilistic nowcasts were generated using a set of 24 ensemble members. The Relative Operating Characteristic (ROC) curve and Fraction Skill Score (FSS) are used as verification methods. The ROC reveals almost similar values for all three optical methods, with DARTS outperforming for extremely heavy rainfall with a lead time of up to +3 hr. It is observed that FSS reduces with higher intensity thresholds and increases with higher spatial scales, indicating that nowcasting skill reduces with increased intensity and finer spatial resolution. In general, even for a spatial scale of 1 km and rain rate threshold of 0.1 mm/h, the nowcasting lead time for light and moderate rain days is up to 1 hr, while for other rain categories, it is up to 2-2.5 hr.

This study introduces a modified radar-based technique to estimate planetary boundary layer (PBL) height using dual-polarization radar observations. Synergistic measurements from C-band dual-polarization radar and GPS radiosondes collected during July–September 2023 at the Atmospheric Research Testbed–Central India (ART–CI), Silkheda (23.57°N, 77.24°E; 563 m amsl), were used to validate the method and examine the diurnal variation of the PBL. Utilizing a sequence of radar volume scans, Quasi-Vertical Profiles (QVPs) were generated to construct height–time cross-sections. The PBL is delineated by identifying the sharp gradient in the differential reflectivity (ZDR) ratio to spectral width—a distinct signature attributed to Bragg scattering at the interface between the boundary layer and the free troposphere. Validation against concurrent GPS radiosonde data reveals a striking correlation (r = 0.96) with a root mean square error (RMSE) of ~149 m. Additionally, our estimation of the entrainment zone depth, which shows a moderate correlation (r = 0.70, RMSE ~155 m) with radiosonde measurements, underscores the robustness of the approach. This radar-based technique offers strong potential for routine application across operational weather radar networks, enabling high-resolution spatiotemporal monitoring of PBL dynamics under varying meteorological conditions.

Multifrequency radar retrievals of precipitation characteristics are more accurate than single-frequency radars. This study focuses on retrieving the raindrop size distribution parameters (e.g., median volume drop diameter;$D_{\mathbf {0}}$and normalized intercept parameter,$N_{\mathbf {w}}$) from X- and Ka-band radars using the dual-frequency ratio (DFR) technique and quantitatively assesses the retrieved parameter with co-located Joss-Waldvogel Disdrometer (JWD) and global precipitation measurement (GPM) observations. Both radars are collocated at Mandhardev ($18.04~^{\circ }$N,$73.86~^{\circ }$E, ~1.3 km AMSL), a high-altitude site in the Western Ghats (WGs), India. The T-matrix scattering simulations are performed at X- and Ka-band to establish the relationship between DFR and$D_{\mathbf {0}}$. Using$D_{\mathbf {0}}$, an iterative calculation is performed to calculate the$N_{\mathbf {w}}$. The$D_{\mathbf {0}}$values derived from DFR demonstrate good agreement with the disdrometer measurements, while compared to GPM observations, they reveal similar vertical variations but differing magnitudes.

A generic architecture for autonomous agents is presented. In common with other current proposals the agent is capable of reacting to and reasoning about events which occur in its environment, executing actions and plans in order to achieve goals in its environment, and communicating with other agents. The work described here proposes certain advances on other systems, notably the ability to reason about and make decisions under uncertainty, including decisions about competing beliefs and alternative actions. The framework is grounded in a non-classical decision model, the ‘domino’ model. This is formalized to ensure continuity with classical decision theory and avoid ad hoc features. The domino model is embodied in a well-defined knowledge representation language, R2L, which explicitly supports the central concepts of decisions and plans, and associated constructs of goals, arguments, commitments, obligations and constraints. The availability of such a language provides a sound basis for building knowledge-based agents for practical applications. A major issue for such applications, however, is how to ensure their safe operation. This is a central issue whether the agents are used in an advisory role (e.g. decision support systems) or an autonomous one (e.g. in a robot). Techniques for explicit management of safety are described and some broader theoretical implications are discussed.

A supervised classifier is described where a systematic search through a training database produced an efficacious set of prototypes for its operation. Using this technique, two studies, one dealing with phone recognition and another with large-vocabulary speech recognition, were conducted with data from several speakers. The superiority of the supervised classifier design is demonstrated by comparative performance analysis.>