SMARTER WATER MONITORING WITH IOT RURAL POND AND AQUACULTURE MONITORING EXTENDED SENSING WITH OFFLINE RESILIENCE CALIBRATION SUPPORT AND PWQI VALIDATION

Abstract

Even where villages depend on local ponds for daily needs, checking their condition still relies on rare hand-collected samples - methods that lag behind actual changes due to delays and expense [1], [2], [3]. While some smart setups using ESP32 chips and budget sensors now allow live tracking, they miss key factors like oxygen levels in water, fail during network blackouts, offer unclear feedback, and skip upkeep planning - especially problematic for users with minimal reading skills [2], [4], [5], [6]. Here comes a redesigned buoy-style device measuring five aspects: acidity, total dissolved solids, cloudiness, oxygen presence, and heat; powered by sunlight and conserving power through timed wake-ups; storing data locally while syncing later when connection returns; switching between long-range radio and cellular networks as needed; prompting recalibration at intervals; plus computing its own simplified Pond Water Quality Index assigning one of four practical labels based on suitability. Tested against more than four thousand records from fish farming sites showing clear chemical links between acid levels and oxygen (correlation r = −0.709), it sorted conditions correctly 95.7% of the time, scoring averages of 80.4 (best), 61.1 (suitable), and 33.5 (unsuitable). When lined up beside newer models published in IEEE journals from 2024 onward, this version covers more variables, handles disconnection better, and speaks clearer to people managing resources day-to-day