Rice is a commodity that plays an essential role in the Indonesian economy and in general, around 90% of the Indonesian population consumes rice as a staple food (Donggulo et al., 2017). Rice production in Indonesia from 2020 to 2023 tends to decline, based on data from the Central Statistics Agency (BPS) of the Republic of Indonesia in 2023. Rice production in 2020 was 37.7 million tons, decreased by 200,000 tons in 2021 to 34.5 million tons. In 2022, domestic rice production decreased again to 34.4 million tons and then in 2023, Rice production in Indonesia reached 30.9 million tons. It was caused by the drought that hit Indonesia that year. Many efforts have been made to produce rice with efficient use of water.
       
The International Rice Research Institute uses genetic engineering, breeding and integrated resource management to increase rice production by reducing water requirements in rice fields. Water-saving irrigation, such as saturated soil cultivation and alternating wetting and drying, can drastically reduce unproductive water flows and increase water productivity. However, this technology predominantly causes decreased yields in current lowland rice varieties (Smith, 2020).
       
Other new approaches are being researched to improve water management without sacrificing yields. One of them is using molecular biotechnology to increase drought stress tolerance and aerobic rice development to achieve high and sustainable yields on non-flooded soil. Through water-saving irrigation technology, rice fields will change from continuously anaerobic to partially or even wholly aerobic conditions (Brown, 2022). However, the obstacles faced xplained that the soil is dry and weeds are often an obstacle to production and compete with rice plants for sunlight, water, CO₂, O₂ and nutrients.
       
Regarding flooded or dry rice fields, there are also problems regarding the availability of nitrogen (N). The N element is leachable and easily leached in the soil, so the availability of N in the soil is low. Nitrogen can also be supplied from natural sources by making synthetic fertilizers. Urea is the most widely used N fertilizer to provide N nutrients for plants, including when rice fields are flooded. Urea fertilizer is a highly water-soluble nitrogen fertilizer that immediately dissolves in the soil solution after application. Once in the soil, urea begins to hydrolyze as it absorbs water and reacts with the soil urease enzyme, initiating the transformation of nitrogen into forms available for plant uptake. This rapid dissolution and subsequent hydrolysis make urea one of the most efficient and widely used nitrogen fertilizers in agricultural production (Zhao et al., 2022).
       
On the other hand, in dry conditions, nitrogen is lost from the soil through evaporation (volatilization). Based on the description of the agricultural sector’s condition, one effort can be made as an alternative method for estimating nitrogen levels in plants, such as applying technology, which has great potential for assessing high levels of precision at low cost. This method is considered productive, adequate in data acquisition or retrieval and efficient in terms of time and operational costs, including the application of technology to obtain information from easy and more accurate images, such as the physical condition of plants (Yu et al., 2022).
       
Technological development of smart farming for various applications is a feasible method. The development of technology for rice-based cultivation, the Internet of Thinking (IoT), is an effort to increase rice productivity. One of the uses of IoT is unmanned aerial vehicle (UAV), often referred to as a drone. Drones equipped with multispectral cameras are highly recommended for use as monitoring and prediction tools, providing information regarding crop status through images. It can simplify the evaluation process on a broader scale. Information that can be obtained from using UAVs for agriculture is normalized difference vegetation index (NDVI) to determine plant health conditions, plant nutrient absorption, plant density and monitoring plant needs. Meanwhile, normalized difference water index (NDWI) was used to determine water availability in rice cultivation and the water found in the soil.
       
One advantage of remote sensing technology is that it can monitor objects over a large area. The presence of drones also supports precision agriculture because it is an alternative with low costs and efficiency (Norasma et al., 2019). Apart from this, remote sensing technology is a technology for estimating N content from the results of vegetation index analysis from aerial photos using a multispectral camera.

The research will be conducted on intensive agricultural land in Calaccu Hamlet, Mattirowalie Village, District. Maniangpajo District. Wajo (Fig 1). Rice plants are planted in November 2023 and harvested in March 2024.


       
The research utilized a UAV from the DJI Phantom 4 series equipped with a multispectral camera and a Chlorophyll Content Meter 200 Plus (CCM-200 Plus). The materials used included Mekongga variety seeds, husk soil, Urea, SP-36, KCl and Pespticide fertilizers. Additionally, various materials were used for soil sample analysis in the laboratory, such as soil and plant tissue samples, concentrated sulfuric acid, NaOH, H₂SO₄, H₃BO₃ and distilled water (aquades).
       
This research employed a split-plot design based on urea fertilizer application and water management using intermittent irrigation. The sub-plot treatments consisted of different urea fertilizer doses: P0 (control, no fertilizer), P1 (150 kg/ha of urea), P2 (250 kg/ha of urea) and P3 (350 kg/ha of urea). The main plot treatments for water management were G1 (4 days of inundation and 4 days of dry) and G2 (7 days of inundation and 7 days of dry). Each plot measured 3 m ×  4 m and the experimental treatments were repeated three times, resulting in a total of 24 research plots. The legowo 2:1 planting system was implemented and NDWI-based analysis was conducted to assess the effectiveness of water on rice plants. The results of the observations were analysed using variance (ANOVA) to determine whether there was a real influence on the treatment given. Further testing was done to determine whether the treatment had fundamental differences. A correlation test will also be used to determine the relationship between methods.
       
Aerial photo recording using the drone application was carried out 2 times, namely when the rice plants were 35 day after planting (DAP) in the experimental map. The flying height of the drone used is 25 M Flying Height (TB) In previous research, it was shown that there was a good drone flying height of 50 meters, Jamisyah (2022). This shows that the higher the drone flys, the less accurate the quality of the information obtained. This is in line with what was conveyed,stated that the higher the drone flies, the smaller the number of objects produced, so that it has a singnifiable effect on the number of objects from the image that is classified as if the information obtained is inaccurate. Added by Shofiyanti (2011), the quality and resolution of the image produced by the drone depends on the flight altitude.
 
Data analysis
 
This analysis method uses  the Normalized Difference Vegatation Index (NDVI). Vegetation index analysis using the AgisoftMetashape application The mathematical transformations used are:

Where,
NDVI = Normalized difference water index.
NIR = Infrared near.
R = Red.
       
NDVI values range from -1 to 1. Higher values tend to indicate healthier plant growth and more leaves. Data segmentation applies a threshold to the NDVI value to identify areas that have better or worse plant growth. Monitoring changes during the growing season, perform repetitive monitoring using imagery taken at specific time intervals. Compare NDVI changes over time to observe plant growth and health.
       
To see how water management monitoring in rice plants is by means  of NDWI data collection. First of all, collect satellite image data of drone images that cover the rice planting area that you want to monitor. Ensure that the image obtained has at least two necessary spectral bands, namely the near-infrared band (NIR) and the visible band (e.g., green or blue). The NDWI calculation method is calculated using the following formula:

Where,
NDWI = Normalized difference water index.
NIR = Near infrared band.
Blue = Pixel value in the blue or green band.
       
Interpretation of NDWI values: After calculating the NDWI for each pixel in the image, it will get a value range between -1 to 1. A positive NDWI value (close to 1) indicates the presence of high water content in the plant. A negative NDWI value (close to -1) indicates an area that tends to have no water content.

The rice plants cultivated at the research site are a new type of rice with high productivity but require good water management and balanced fertilization, From the observations of rice plants at the research site, such as plant height, number of leaves (Fig 2), number of tillers, which are respectively at the age of 20 and 40 DAP and plant chlorophyll index (Fig 3). It was found that the results of the experiment on the fertilization treatment at a dose of 350 kg/ha (P3) The flooding and fertilizer doses did not significantly affect rice plant parameters (Fig 2). However, the treatment that tended to produce the highest average value for rice plant height at 20 days after planting (DAP) was the 350 kg/ha urea fertilizer treatment (P3), at 43.83 cm, and the lowest value was the treatment without urea fertilizer (P0), at 40.28 cm. 40 days after planting = and for rice plant height at 40 days after planting, the treatment that tended to produce the highest average value was the 350 kg/ha fertilizer treatment with 7 days of flooding and 7 days of dryness (P3G2), at 72.9 cm, and the lowest value was the treatment without fertilizer with 4 days of flooding and 4 days of dryness (P0G1), at 57.4 cm showed the best results although the tendency had no significant effect on the treated maps, while for the inundation experiment, the 4- and 7-day treatments did not show a significant difference. This data shows that the intermittent irrigation process tends to have a good impact on the optimization of fertilizer absorption, so that a small dose of fertilizer such as in (P1) can provide optimal growth and is not much different from the highest dose (P3), this can be a recommendation for rice cul. Fig 2(a) showed that the treatment of urea at a dose of  The treatment that tended to produce the highest average number of leaves in rice plants 20 days after planting was the urea fertilizer dose of 350 kg/ha with 7 days of flooding and 7 days of drying (P3G2), with 44.22 leaves. The lowest value was found in the unfertilized treatment with 4 days of flooding and 4 days of drying (P0G1), with 17.67 leaves. For the number of leaves in rice plants 40 days after planting, the treatment that tended to produce the highest average number was the 350 kg/ha fertilizer dose with 4 days of flooding and 4 days of drying (P3G1), with 203.33 leaves. The lowest value was found in the unfertilized treatment with 4 days of flooding and 4 days of drying (P0G1), with 60.56 leaves. For the leaf number parameter, the treatment with the highest leaf count was 203.33 leaves at a fertilizer dose of 350 kg/ha with 4 days of flooding and 4 days of drying (P3G1), while the lowest value was 60.56 leaves at 40 days after planting. This occurs because chlorophyll This is because of the nitrogen content in urea fertilizer which is useful as a plant growth booster. This is in accordance with Adzima (2022) which states that element N is a macronutrient that is useful for plant growth, which is generally needed during the vegetative period of plants for an increase in root length and an increase in plant height.  350 kg/ha (P3) provided the best average value in the treatment of 4 days of inundation and 4 days of dry and 7 days of inundation and 7 days of dry. The application of fertilizer to plants is closely related to the essential nutrients needed by plants, N fertilizer is one of the macronutrients for rice plants. Nitrogen fertilization is related to the increase in plant height, leaf area and number of plant leaves (Ueda, 2017). Fig 2(b) shows the long treatment of inundation and dryness of 7 days each also affects the increase in plant height. In dry conditions, the length of plant roots increases which helps plants to absorb fertilizer optimally and respirate aerobically also the irrigation management was significant in shoot dry weight value (Bian, 2023). According to Sagar et al., (2024) about Nitrogen (N) is one of the macronutrients for plant growth, which is generally indispensable for vegetative growth of plants such as roots, stems and leaves.




       
Fig 2(c to d) shows the increase in the number of leaves of rice plants is influenced by the availability of N nutrients in urea fertilizer. The development that occurs in the vegetative phase of rice is determined by several factors, such as fertilizer efficacy, nutrient availability and the ability of plants to obtain and process nutrients. Nitrogen is the main nutrient that promote vegetative growth in plants, such as increasing the number of seedlings, developing in the number and area of leaves (Ueda, 2017). Periodically dry soil conditions are expected to make plants grow roots as a response to survive. Long roots can help plants reach water sources and absorb nutrients that are more optimal for plant growth.
       
Fig 3(a) shows the number of tillers per clump shows a positive increase, this is because the nitrogen contained in the fertilizer can stimulate the roots to grow, which further affects the increase in the number of saplings. Roots play a role in plant growth to provide nutrients and water for plant needs in plant physiological processes (Salimah and Wahdah, 2015). Fig 3(b) shows that the intermittent irrigation treatment did not give a very significant effect on the increase in the number of tillers, this is because the number of tillers in the urea treatment at a dose of 350 kg/ha (P3) in conditions of 7 days of inundation and 7 days of dry was only 0.78 with 4 days of inundation and 4 days of dry.
       
Fig 3(c to d) shows the plant chlorophyll index was carried out to determine the absorption and chlorophyll content obtained from the nutrient N. The addition of N nutrients can increase the amount of chlorophyll pigment in rice plants, this is supported by a report from Ueda (2017) which states that nitrogen fertilization correlates with photocinate net, chlorophyll pigment amount, leaf area, seed weight, number of seeds and plant biomass. It is also supported by the statement of Soepriyanto (2021) which states that the higher the degree of solubility of a fertilizer given, the easier it is absorbed by plants and affects the efficiency of nitrogen absorption by plants, so that the amount of chlorophyll produced increases, or vice versa.
 
UAV-multispectral data analysis
 
UAV observations were carried out to see the normalized difference water index (NDWI) and normalized difference vegetation index (NDVI) values in rice plantations, this observation was made in the vegetative phase of rice plants (35 DAP). Based on the results of NDWI analysis (Fig 4), the 7-day inundation and 7-day dry (G2) treatment gave the highest value compared to the 4-day inundation and 4-day dry (G1) treatment. Meanwhile, the results of NDVI analysis (Fig 4) showed that the treatment of urea fertilizer dosage of 350 kg/ha (P3) gave the highest value compared to other urea fertilizer dose treatments. NDWI and NDVI data can be seen in the following figuretivation applying intermittent irrigation.


       
Normalized difference water index (NDWI) this study, the treatment of 7-day inundation and dry gave the highest value compared to other treatment interactions, which at 35 DAP 7-day inundation obtained a value of 0.507 and was significantly different from 4-day inundation with a value of 0.306. This result is in accordance with Gulácsi and Kovács (2018) which stated that the NDWI class of 0.5-0.6 is included in the category of medium moisture content. Water that inundates the land increases the moisture of the soil and the surrounding plants. NDWI is very sensitive to water content, so the NDWI value increases if waterlogging or humidity is relatively high. Another thing is also because rice plants have the ability to absorb water from stagnant soil, thereby increasing the moisture content in plant tissues. The light reflectance in the near-infrared (NIR) spectrum and the green spectrum will differ under these conditions, resulting in higher NDWI values. In accordance with Lu and Fricke (2023) who stated that excessive use of chemical fertilizers can cause plant stress or affect the osmotic balance of plant roots, reducing the plant’s ability to absorb water. According to Ashraf et al., (2025), who stated that moisture stress directly inhibits nitrogen uptake in rice plant because low soil moisture reduces nutrient solubility and root activity. When chemical fertilizers are reduced, plants are healthier and able to absorb water more efficiently, with less chemical fertilizers, soil structure tends to be more stable, allowing plant roots to absorb water better. This condition contributes to the increase in NDWI scores.
       
The highest normalized difference vegetation index (NDVI) was the dose of urea fertilizer of 350 kg/ha (P3) which at 35 DAP obtained the highest value of 0.23 and was significantly different from other treatments except for the treatment of 250 kg/ha (Fig 5). This is in accordance who stated that in the initial vegetative phase the NDVI value ranged from 0.2-0.4, while in the flowering and panicle formation phase it ranged from 0.7-0.9. This is also in line with the chlorophyll index and also productivity, where laboratory analysis shows relatively similar results that a dose of 350 kg/ha obtains the highest chlorophyll value. This Is because the greener a plant, the higher the chlorophyll of the plant, so it is highly recommended in large-scale rice cultivation. This is supported by Adzima (2022) in his research stating that the NDVI index will be directly proportional to plant chlorophyll. According to Tamilmounika et al., (2024) which stated that the NDVI derived from drone imagery showed a significant positive relationship with SPDA values (Soil Plant Analysis Develop-ment), which serve as an indicator of leaf chlorophyll content. The greenness of the leaves shows that a plant has sufficient nutrients where the nutrient N has an important role in the greenery of the leaves, the reflection of light in the near-infrared spectrum (NIR) will increase according to the level of greenness of the leaves, which chemical fertilizers support higher photosynthesis because it is easier for plants to get the nutrients they need, with the increase in photosynthesis, the chlorophyll produced also increases,  thereby increasing the NDVI value (Istiqomah, 2020).

Based on the results of the study, it can be concluded that there is an interaction between the urea fertilizer dose of 350 kg/ha with a 7-day flooding period and a 7-day dry treatment period (G2P3) on plant height parameters (72.9 cm) and the total amount of chlorophyll (330.06 UNITS). There is an interaction between the 350 kg/ha dose treatment with 4 days of flooding and 4 days of dry treatment (P3G1) on the number of plant leaves (203.33 strands). The highest NDWI value is at a 7-day flooding period and a 7-day dry treatment period (G2) with a value of 0.51 and the highest NDVI value is at a 350 kg/ha urea fertilizer dose treatment (P3) with a value of 0.24.
       
Based on the results of the study, it can be concluded that there is an interaction between the urea fertilizer dose of 350 kg/ha with a 7-day flooding period and a 7-day dry treatment period (G2P3) on plant height parameters (72.9 cm) and the total amount of chlorophyll (330.06 UNITS). There is an interaction between the 350 kg/ha dose treatment with 4 days of flooding and 4 days of dry treatment (P3G1) on the number of plant leaves (203.33 strands). The highest NDWI value is at a 7-day flooding period and a 7-day dry treatment period (G2) with a value of 0.51 and the highest NDVI value is at a 350 kg/ha urea fertilizer dose treatment (P3) with a value of 0.24.

The authors declare no conflict of interest.