INTRODUCTION Water use is becoming one of the more worrisome environmental issues in many parts of the world as the planetary population continues to expand and resources become more constrained. The amount of water used to keep the fields in optimum condition can occasionally be wasted by ineffective irrigation practices used on farms, especially when it comes to rice. It is a fallacy that rice can only be produced in submerged environments. Farmers in India and many other countries across the world regularly pond their rice crop for over 80% of the time they are in cultivation. Previously rice was cultivated in submerged conditions to control weeds but over time it has become mandatory. A switch from a continuously submerged system to an alternate wetting and drying system is therefore necessary. Continuous flooding can be avoided by AWD, which also conserves water. There is an added benefit if AWD can be automated using IoT (Internet of Things). Thus, the primary goal is to optimize the use of water in agricultural irrigation systems by applying artificial intelligence techniques and thereby reducing the quantity of water lost with conventional irrigation systems. Modern farmers are using cutting-edge technology to track crop productivity, gather data on crop growth, and collect weather data. IoT can connect all of them, and automated irrigation was once the place to start. By using IoT farmers can regulate the irrigation water as per crop needs. As a result, irrigation efficiency and water productivity are improved while ensuring logical water distribution. Because so much nitrogen is lost from the soil through leaching and denitrification, rice has an extremely low nitrogen utilization efficiency of less than 30 to 40 %. The efficiency of added fertilizer N in rice is influenced by N sources, application techniques, N rates, and management techniques (Wang et al., 2011). For researchers and producers, effective management of fertilizers, particularly N, remains a significant concern. Therefore, implementing intelligent management strategies at the right moment according to the crop's needs is absolutely essential. Utilizing precise nitrogen management strategies, such as LCC, green seeker, and nutrient expert-based nitrogen management, can optimize and recommend nitrogen and meet crop nutrient needs while causing minimal environmental damage. Keeping the views above the experiment was planned to study the response of rice (Oryza sativa L) to irrigation regimes and precise nitrogen management. MATERIALS AND METHODS During the kharif season of 2021, the field experiment was held at the Water Technology Centre, College Farm, College of Agriculture, Rajendranagar, Hyderabad, Telangana. The experimental site was located at 17°32' N Latitude, 78°40' E Longitude, and an altitude of 542.6 m above mean sea level. The experiment field had a sandy clay texture, a pH of 7.6, and an EC of 0.86 dS m-1. It was low in organic carbon (0.45%) and available nitrogen (242 kg ha-1) but high in available phosphorous (34.5 kg ha-1) and potassium (197.6 kg ha-1). KNM-118 paddy variety was transplanted 30 days after sowing with 2-3 seedlings per hill with a spacing of 15cm × 15cm. The experiment was laid out in strip-plot design with three irrigation methods as main plot treatments viz., I1-Alternate wetting and drying by the sensor-based semi-automatic - Irrigation up to 5 cm depth throughout the crop growth when the water level drops 5 cm below ground level, I2- Alternate wetting and drying by manual - Irrigation up to 5 cm depth throughout the crop growth when the water level drops 5 cm below ground level, I3-Flooded transplanted rice- 2-5 cm of water depth as per crop growth stage (control) and four nitrogen treatments as subplot treatments viz., N1- Basal 1/3rd of N as soil application + top dressing of nitrogen as per LCC values, N2 - Basal 1/3rd of N as soil application + top dressing of nitrogen as per green seeker values, N3 - Nutrient expert-based nitrogen management, N4- Recommended dose of fertilizer (120-60-40 kg NPK ha-1). Each plot was separated by providing buffer channels for proper maintenance of the treatments. The amount of water applied to the field is measured with the help of a water meter. In irrigation methods, AWD by manual irrigation the water level in the field is observed using a field water tube and when the water level reaches 5cm below ground level, the field was irrigated until 5cm above the ground and in AWD by sensor-based irrigation, water was applied to the field based on the indication from the sensor.For this study, we used the Smart Paddy Internet of Things technology, which applies to rice fields with fully automated and autonomous behaviour. Sensor can be automated in such a way that when the water level falls below 5cm below ground level, the motor turns on, and when the water level rises above 5cm above ground level, the motor turns off, preventing over-application of water.The farmer can control flows, volumes, and water levels in the fields using a special website, mobile app, or SMS service by manually operating when necessary or by configuring the irrigation program in accordance with the water requirements of various water-saving strategies. In nitrogen management treatments nitrogen fertilizer was applied based on the LCC(Leaf colour chart is an easy-to-use tool for monitoring the relative greenness of a rice leaf as an indicator of the plant N and urea was applied when leaf matches LCC-3), and Green seeker (Green seeker handheld sensor is an easy-to-use optical sensor that instantly measures plant height and vigor in terms of NDVI readings and urea was applied when NDVI value is less than 0.7) in N1 and N2 respectively, whereas in N3 nitrogen fertilizer was applied in accordance with Nutrient expert software and in N4 as per recommended dosage(120-60-40 Kg NPK ha-1). RESULTS AND DISCUSSION A. Yield Attributes Number of Panicles (m-2). The number of panicles m-2 was higher in I3 (308 m-2) than in I1 (266 m-2) and I2 (273 m-2) (Table 1). The lower number of panicles m-2 observed during the later water regime may be due to the fact that plants under moisture stress could not extract more nutrients from the deeper soil layer due to moisture deficit conditions. This ultimately resulted in poor growth, fewer tillers, and subsequently fewer panicles. Kumar et al. (2014); Sandhu et al. (2012) reported similar findings. The number of panicles m-2 was much higher in the N1 (304 m-2) than in the N3(261 m-2) and N2 (271 m-2) and were on par with the N4 (293 m-2) (Table 1). The lowest number of panicles m-2in N3may be the result of insufficient nitrogen for crops to grow and develop more effectively (Sandya Rani, 2012). The findings concur with those of Sun et al. (2012); Ali Abdalla and Abou-Khalifa (2012). Panicle Length(cm). Panicle length was higher in I3 (22.8 cm) than in I1 (21.0 cm) and I2 (21.3 cm) (Table 1). The lower panicle length in I1 and I2could be due to thedue to lower water level might have caused moisture stress to rice plants.These findings are in consistent with those of Azarpour et al. (2011); Rahaman and Sinha (2013). The panicle length was much higher in the N1 (22.8 cm) than in the N3 (20.4 cm) and N2 (21.5 cm) and on par with the N4 (22.2 cm) (Table 1). It is viable that the increased availability and uptake of N, which is a substrate for the synthesis of organic compounds that comprise protoplasm and chlorophyll, caused an increase in cell division and enlargement at higher nitrogen doses (Avijit et al., 2011). The findings of this study are similar to those of Malik et al. (2014); Ali Abdalla and Abou-Khalifa (2012); Debnath and Bandyopadhyay (2008). Test Weight (g). In contrast to I1 (21.6 g) and I2 (23.8 g), I3 had a higher test weight (21.9 g) (Table 1). Whereas I1 and I2 were on par with each other. According to the findings of several researchers, test weight can increase when crops receive enough water without being stressed during the flowering and grain development stages (Pandey et al., 2010; Rahaman and Sinha, 2013; Srinivasulu, 2017). Test weight (g) was substantially greater in N1 (24.1 g) than in N2 (22.1 g), N3 (20.6 g), and was on par with the N4 (23.0 g) (Table 1). The increased test weight could result from higher nitrogen application levels than lower nitrogen application levels transferring more carbohydrates to grain. Bhavana et al. (2020) reported findings that were comparable to these findings. Grain Yield (kg ha-1). In comparison to I1 (5532 kg ha-1) and I2 (5582 kg ha-1), I3 had a higher grain production of 6302 kg ha-1(Table 1). While I1 and I2 were on par with each other. The improved performance of the crop plants was aided by a favourable soil water balance under saturation, which may have contributed to the higher seed yield observed in the flooded transplanted rice. This favourable vegetative growth and development under an adequate and sufficient moisture regime were maintained throughout the crop growth. As a result, crop plants that received regular flooding irrigation when ponded water disappeared from the surface of the ground produced more tillers. Similar findings had been made by several other researchers Pandey et al. (2010); Kumar et al. (2013); Srinivasulu et al. (2017). The grain yield in N1 (6125 kg ha-1) was significantly greater than that in N3 (5390 kg ha-1) and was comparable to that in N2 (5792 kg ha-1) and N4 (5915 kg ha-1) (Table 1). The enhancement of yield characteristics, which in turn increased the yield, was presumably caused by an adequate N supply during the reproductive growth phase (Duttarganvi et al., 2014). Kenchaiah et al. (2000) also found higher grain yield under LCC-based N management than the blanket recommendation. Similar findings were reported by Manjappa et al. (2006); Houshmandfar and Kimaro (2011); Sui et al. (2013); Bhavana et al. (2020). Harvest Index. A higher harvest index was recorded in I2 (47.9) than in I3 (46.9) and was on par with I1 (47.5) (Table 1). A significantly higher harvest index was observed in N2 (49.1) than in N3 (45.5) and was on par with N1 (48.5) and N4 (46.7) (Table 1). Timely release of nitrogen in a sustained manner to absorb and translocate sufficient quantities of photosynthates to the sink, resulted in the production of elevated yield structure and yield in top dressing of N as per LCC. This study findings concur with those of Moharana et al. (2017); Tauseeef Ahmad (2014). Water Saving. Amount of water saved in AWD by sensor-based semi-automatic irrigation and AWD by manual irrigation is 25.2% and 23.3 % higher than flooded transplanted rice.