Dept of IDE http://localhost:8080/xmlui/handle/123456789/19 2026-08-04T12:50:38Z 2026-08-04T12:50:38Z ANILA J W GAUTHAM M R ANURAG P Praveena K K, (Guide) http://localhost:8080/xmlui/handle/123456789/2413 2026-07-20T10:34:19Z 2026-01-01T00:00:00Z

Title: PREDICTING SOIL WATER DEFICIT IN COCONUT PLANTATIONS USING SATELLITE DATA AND MACHINE LEARNING Authors: ANILA J W; GAUTHAM M R; ANURAG P; Praveena K K, (Guide) Abstract: Irrigation scheduling for coconut (Cocos nucifera L.) traditionally relies on generalized crop coefficients and theoretical water balance computations, which fail to capture real-time crop water status. This study, conducted at the Central Plantation Crops Research Institute (CPCRI), Kasaragod, Kerala, developed a satellite-driven machine learning framework to predict daily soil water deficit (SWD) for mature coconut and to schedule irrigation accordingly. Data were collected across 39 grid points within the plantation for the period 2019–2026, combining meteorological data from NASA POWER with satellite-derived vegetation and moisture indices. Reference evapotranspiration and crop evapotranspiration were computed using the FAO-56 Penman-Monteith method, and a daily soil water balance was used to derive SWD and binary irrigation labels from irregular satellite passes. Random Forest, XGBoost, and LSTM models were trained and compared on two input configurations: Dataset A (full meteorological variables) and Dataset B (satellite-centric, with leakage-prone variables removed), allowing the same three algorithms to be evaluated for their reliance on satellite signal versus meteorological arithmetic. On Dataset A, Random Forest and XGBoost achieved near-perfect regression accuracy (R² = 0.9983 and 0.9957, respectively) and classification accuracy of 1.00, but this was traced to target determinism, the models reconstructing the deterministic water-balance formula rather than learning genuine satellite-based crop-water signals. A Cascade architecture, which thresholded XGBoost's continuous SWD output at 60 mm to obtain irrigation decisions, failed entirely on the minority Irrigate class (recall = 0.00). In contrast, LSTM, trained directly as a classifier on the satellite-centric Dataset B, achieved an R² of 0.8117 (RMSE 11.76 mm), an overall classification accuracy of 0.9927, and a recall of 0.9766 with an F1-score of 0.9653 for the Irrigate class, correctly identifying 167 of 171 irrigation events. These findings show that headline regression or classification accuracy alone is not a reliable indicator of a model's field usefulness for satellite-based irrigation scheduling, and that LSTM offers a more robust and field-deployable alternative to tree-based models trained on leakage-prone meteorological inputs. The study demonstrates the feasibility of a satellite-only 54 irrigation scheduling framework for mature coconut and provides a basis for future work using a clean, fully satellite-derived dataset.

2026-01-01T00:00:00Z ATHIRA KRISHNAN FASNA SHERIN P Anu Varughese, (Guide) http://localhost:8080/xmlui/handle/123456789/2410 2026-07-20T09:40:30Z 2026-01-01T00:00:00Z

Title: IOT‑BASED FERTIGATION AUTOMATION AND HYDRAULIC PERFORMANCE EVALUATION OF DRIP SYSTEM AND FERTIGATION EQUIPMENTS Authors: ATHIRA KRISHNAN; FASNA SHERIN P; Anu Varughese, (Guide) Abstract: The study was taken up to develop and evaluate an automated, IoT-based drip irrigation and fertigation system for a vertical farming structure. First, the hydraulic performance of three fertigation equipment units was tested; the ¾-inch Venturi injector achieved a maximum suction rate of 3.6 to 3.7 L/min at 0.6 kg/cm² pressure, outperforming the positive-displacement flow restrictions. Second, testing the vertical drip system hydraulics showed that increasing operating pressure from 0.2 to 1.0 kg/cm² gave an emission uniformity in the range 60.1% to 69.3% across the vertical tiers. Third, to prepare for smart dosing, the 3-in-1 NPK sensor was calibrated using soil samples analyzed by the District Soil Testing Laboratory, Malappuram, generating highly accurate correction equations (R2 = 0.9706 for N, R2 = 0.8812 for P, and R2 = 0.8817 for K). Finally, the fully integrated ESP32 automation system successfully monitored a microclimate environment of 27.2°C temperature and 86% humidity. It demonstrated precise closed-loop control by automatically activating the main water pump, while successfully keeping all three nutrient dosing valves turned off when the real-time soil fertility readings remained safely above the minimum programmed baseline thresholds and turned on when it was below the threshold levels. Thus an Internet of Things (IoT)-based fertigation system was developed and tested in the field to enable automated and precise application of irrigation water and fertilizers based on real- time field conditions.

2026-01-01T00:00:00Z SUHANA MOL RASHEED SONA MOHAN KRISHNA K P Sajeena S, (Guide) http://localhost:8080/xmlui/handle/123456789/2406 2026-07-20T07:54:00Z 2026-07-01T00:00:00Z

Title: PERFORMANCE EVALUATION OF ARDUINO-CONTROLLED DYNAMIC LED GROW LIGHT SYSTEMS FOR INDOOR PRODUCTION OF AMARANTHUS Authors: SUHANA MOL RASHEED; SONA MOHAN; KRISHNA K P; Sajeena S, (Guide) Abstract: The increasing adoption of LED grow lighting has created new opportunities to optimize plant growth through precise manipulation of light spectrum and intensity. Stage-specific spectral management has gained considerable attention as an effective strategy for improving crop growth, productivity, and resource-use efficiency. The present study was conducted to evaluate the effect of dynamic red and blue LED spectral treatments on the growth performance of red amaranthus (Amaranthus tricolor L., cv. Vyga) under an indoor cultivation system. A dual-rack LED lighting system was developed using Arduino Uno, PWM-based intensity control, MOSFET driver circuits, potentiometers, and a 16×2 I²C LCD display for real-time monitoring and regulation of light intensity. The photosynthetic photon flux density (PPFD) was calculated and maintained according to the crop growth stages. Two dynamic spectral treatments were evaluated: Treatment 1 (T1) with a Blue to Red spectral progression and Treatment 2 (T2) with a Red to Blue spectral progression, while maintaining stage-specific PPFD requirements during germination, seedling, and vegetative growth. Biometric observations, including plant height, stem girth, number of leaves, and leaf area, were recorded at seven-day intervals to assess plant growth under both treatments. The experimental data were analysed using two-way Analysis of Variance (ANOVA) to determine the effects of LED treatment, growth period, and their interaction on plant growth. The results showed that all biometric parameters increased progressively with crop age under both treatments. However, Treatment 1 (Blue to Red progression) consistently outperformed Treatment 2 (Red to Blue progression) by recording greater plant height, higher stem girth, increased number of leaves, and larger leaf area. The two-way ANOVA further confirmed that both LED treatment and growth stage had significant effects on plant growth, demonstrating the importance of stage-specific spectral management. Overall, Treatment 1 (Blue to Red progression) was found to be the most effective lighting strategy for promoting the growth and development of red amaranthus under controlled indoor conditions. The developed Arduino-based LED lighting system provides an efficient, economical, and adaptable approach for indoor farming and vertical farming, with potential for application in the cultivation of other leafy vegetable crops

2026-07-01T00:00:00Z ANASWARA C N ANJALI M R HIBA FATHIMA Sarathjith M C, (Guide) http://localhost:8080/xmlui/handle/123456789/2405 2026-07-20T07:48:58Z 2026-07-01T00:00:00Z

Title: IDENTIFICATION OF GROUNDWATER POTENTIAL ZONES IN BHARATHAPUZHA RIVER BASIN USING GIS BASED AHP TECHNIQUE Authors: ANASWARA C N; ANJALI M R; HIBA FATHIMA; Sarathjith M C, (Guide) Abstract: Identification of GWPZs is essential for sustainable groundwater management in the BRB. This study aim to delineate GWPZs in the BRB by integrating GIS and AHP. Eight thematic layers namely geomorphology, geology, land use/land cover, soil, slope, drainage density, lineament density and rainfall were selected based on their influence on groundwater recharge and occurrence. A weighted overlay analysis was performed in the GIS environment to generate the groundwater potential index map and it was reclassified into five categories; very poor, poor, moderate, high and very high groundwater potential zones. Validation was carried out by overlaying observation well locations. Furthermore, to evaluate the influence of rainfall variability, additional analyses were conducted under extreme rainfall conditions minimum non-zero annual rainfall and maximum annual rainfall while keeping other thematic layers constant. The study highlights the spatial distribution of groundwater potential and emphasizes the significant role of rainfall in groundwater recharge dynamics in BRB.

2026-07-01T00:00:00Z