SNDT WOMEN'S UNIVERSITY

BMK Knowledge Resource Centre

Vithaldas Vidyavihar, Juhu Tara Road,
Santacruz (West) Mumbai - 400049

Machine learning-based verification of satellite weather alerts for heavy rainfall in Ladakh

By: Contributor(s): Description: pp63-69Subject(s): In: Current ScienceSummary: The present study investigates heavy rainfall patterns and their variability in Ladakh by using K-means clustering with the Apriori algorithm, which uncovers the co-occurrence pattern of heavy rainfall alerts obtained from the ISRO MOSDAC portal. The proposed algorithms reveal that location (lat.: min. = 32.82, max. = 35.71; long.: min. = 77.59, max. = 79.91) and location (lat.: min. = 33.11, max. = 35.76; long.: min. = 75.82, max. = 77.75) exhibit the highest confidence (79.55%) and frequency (48.61%) among all patterns, indicating strong interdependencies. This suggests that the alerts in one location can potentially impact the other, offering actionable guidance for disaster preparedness. The present study highlights a significant match between predicted alerts and actual heavy rainfall events, underscoring the utility of machine learning in refining weather alert systems
Tags from this library: No tags from this library for this title. Log in to add tags.
Star ratings
    Average rating: 0.0 (0 votes)
Holdings
Item type Current library Call number Vol info Status Barcode
Journal Article SNDT Juhu Available jp861.5
Periodicals SNDT Juhu P505/CS (Browse shelf(Opens below)) Vol. 129, No. 1 (01/07/2025) Available JP861

The present study investigates heavy rainfall patterns
and their variability in Ladakh by using K-means clustering with the Apriori algorithm, which uncovers
the co-occurrence pattern of heavy rainfall alerts obtained from the ISRO MOSDAC portal. The proposed
algorithms reveal that location (lat.: min. = 32.82,
max. = 35.71; long.: min. = 77.59, max. = 79.91)
and location (lat.: min. = 33.11, max. = 35.76; long.:
min. = 75.82, max. = 77.75) exhibit the highest confidence (79.55%) and frequency (48.61%) among all
patterns, indicating strong interdependencies. This
suggests that the alerts in one location can potentially impact the other, offering actionable guidance
for disaster preparedness. The present study highlights a significant match between predicted alerts and
actual heavy rainfall events, underscoring the utility
of machine learning in refining weather alert systems

There are no comments on this title.

to post a comment.