A Libelium-Based Multisensor Dataset for Precision Soil, Microclimate, and Plant Monitoring in an Olive Grove in Goraj, Kucove, Albania

Description

This dataset contains environmental, soil, and plant-related measurements collected using Libelium Waspmote Plug & Sense units installed in an olive grove in the village of Goraj, Municipality of Kucove, Albania. The sensor configuration comprised 9370-P, TEROS 11, PHYTOS 31, Apogee SQ-100X-SS, and a DC3 dendrometer. Data were recorded between 21 February and 13 May 2026 using two sensing units identified as Goraj-1 and Goraj-2. Goraj forms part of the Perondi Administrative Unit and is situated approximately 2.5 km northeast of the city of Kuçovë. The village has approximately 2,000 inhabitants and is characterized by the close relationship between agricultural activity and local water resources. Its hills are extensively covered by olive plantations, locally known as "Ullishta", and vineyards. The surrounding terrain consists mainly of gently rolling hills and plains suitable for agricultural production. The Goraj reservoir, formed by the valley and the Brashnik stream basin, provides an important source of irrigation for agricultural land in Goraj and Perondi. Its presence contributes to humid microclimatic conditions in the valley, particularly through evaporation during summer. The wider Kuçovë area has a continental Mediterranean climate, with very hot, dry summers and cold, wet winters, and holds Albania's national record for the highest recorded temperature of 44.4 °C. These environmental and microclimatic characteristics make the location relevant for continuous monitoring of atmospheric, soil, water-related, and plant conditions in olive cultivation. The deposited files include 9,561 original timestamped hexadecimal frames, 9,435 decoded and quality-controlled sensor measurements, 126 system events, a rejected-row record, a quality-control summary, and the Python script used for decoding and validation. Measured variables include battery charge, air temperature, relative humidity, atmospheric pressure, leaf wetness, dendrometer readings, photosynthetically active radiation, soil volumetric water content, dielectric permittivity, electrical conductivity, and soil temperature. Values outside the defined validation ranges were converted to missing values without deleting complete observations. Quality-control fields identify inactive sensor configurations, temporal gaps, sequence discontinuities, and out-of-range measurements. The original hexadecimal frames are retained to support transparency and reproducibility.

Steps to reproduce

Download and extract the dataset ZIP file. Ensure that Python 3 is installed. No additional Python libraries are required. Open a terminal in the extracted dataset folder. Run the following command: python libelium_cleaner.py libelium_frames_raw.csv -o reproduced_output. The script will generate the quality-controlled measurements, decoded events, rejected-row record, and quality-control summary. The default threshold for identifying large temporal gaps is 20 minutes and can be modified using the --gap-minutes argument. The generated files reproduce the processed data included in the published dataset.

Institutions

• Polytechnic University of Tirana

  Tirana, Tirana

Categories

Funders

• National Agency for Scientific Research and Innovation (NASRI)

  Grant ID: PIKSH No.788/1 Prot.,date 29.04.2025

Licence