Summer Jute Environmental Sensor Dataset from Bangladesh

Description

The Summer Jute Environmental Sensor Dataset from Bangladesh is a field-level agricultural dataset collected using a low-cost Internet of Things (IoT) sensing platform during Kharif-1 (summer) jute cultivation at Sher-e-Bangla Agricultural University, Sher-e-Bangla Nagar, Dhaka, Bangladesh. The dataset contains 5,235 cleaned observations in CSV format. The dataset comprises 11 attributes: DateTime, Location, Season, Temperature, Humidity, Rainfall, Soil_moisture, Type, Sowing, Growth, and Harvest. Environmental measurements were acquired using an ESP32 S3-WROOM-1, DHT22 AM2302, a capacitive soil moisture sensor, and a rain detection sensor, while the remaining attributes describe the cultivation period. The data were collected under natural field conditions through automated sensor monitoring and cleaned by removing missing values, duplicate records, formatting inconsistencies, and inconsistent categorical labels. The dataset is compatible with Python, R, MATLAB, TensorFlow, PyTorch, and scikit-learn. This dataset supports research in smart agriculture, precision farming, environmental monitoring, agricultural IoT, irrigation management, and agricultural data analytics. It can be reused for exploratory data analysis, statistical modelling, environmental pattern analysis, feature engineering, predictive modelling, and benchmarking machine learning methods. Value of the Data: 1. Provides a real-world field-level environmental dataset collected during summer jute cultivation in Bangladesh using a low-cost IoT sensing platform. 2. Supports research in smart agriculture, precision farming, environmental monitoring, irrigation management, agricultural IoT, and machine learning. 3. Enables exploratory data analysis, statistical modelling, feature engineering, environmental pattern analysis, and benchmarking of agricultural methods. 4. Serves as an educational resource for agricultural data analytics, sensor data processing, visualization, and predictive modelling. 5. Facilitates the development and evaluation of IoT-based agricultural monitoring systems using structured environmental sensor data.

Steps to reproduce

The dataset was generated through a reproducible field-level environmental monitoring workflow using a low-cost IoT sensing platform. Researchers who wish to reproduce or extend this dataset may follow the steps below. 1. Select the Monitoring Site Choose an agricultural field used for summer (Kharif-1) jute cultivation or a comparable crop monitoring environment. 2. Prepare the IoT Sensing Platform Assemble an IoT monitoring system consisting of an ESP32 S3-WROOM-1 microcontroller, a DHT22 AM2302 temperature and humidity sensor, a capacitive soil moisture sensor, and a rain and steam detection sensor. 3. Deploy the Sensors Install the soil moisture sensor in the root-zone soil, position the DHT22 sensor above the soil surface to measure ambient environmental conditions, and place the rain sensor in an open location to detect rainfall events. 4. Collect Environmental Measurements Configure the ESP32 to automatically record temperature, humidity, rainfall status, and soil moisture measurements at regular intervals throughout the monitoring period. Store the collected observations together with cultivation-related information, including DateTime, location, season, crop type, sowing period, growth period, and harvest period. 5. Validate and Clean the Data Review the recorded observations and remove duplicate records, missing values, formatting inconsistencies, communication errors, and inconsistent categorical labels. Standardize column names and data formats to create a structured dataset. 6. Export the Dataset Export the validated observations as a CSV file containing the following attributes: DateTime, Location, Season, Temperature, Humidity, Rainfall, Soil_moisture, Type, Sowing, Growth, and Harvest. Following these steps will enable researchers to reproduce a comparable field-level agricultural environmental dataset or extend the dataset for additional crops, cultivation seasons, geographical regions, or IoT-based smart agriculture applications.

Institutions

• American International University-Bangladesh

  Dhaka Division, Dhaka
• Sher-e-Bangla Agricultural University

  Dhaka Division, Dhaka

Categories

Licence