Cross Regional Eucalyptus Growth and Environmental Data

Description

The dataset is provided in a single .xlsx file named "eucalyptus_growth_environment_data_V2.xlsx" and consists of fifteen sheets: Codebook: This sheet details the index, values, and descriptions for each field within the dataset, providing a comprehensive guide to understanding the data structure. ALL NODES: Contains measurements from all devices, totalling 102,916 data points. This sheet aggregates the data across all nodes. GWD1 to GWD10: These subset sheets include measurements from individual nodes, labelled according to the abbreviation “Generic Wireless Dendrometer” followed by device IDs 1 through 10. Each sheet corresponds to a specific node, representing measurements from ten trees (or nodes). Metadata: Provides detailed metadata for each node, including species, initial diameter, location, measurement frequency, battery specifications, and irrigation status. This information is essential for identifying and differentiating the nodes and their specific attributes. Missing Data Intervals: Details gaps in the data stream, including start and end dates and times when data was not uploaded. It includes information on the total duration of each missing interval and the number of missing data points. Missing Intervals Distribution: Offers a summary of missing data intervals and their distribution, providing insight into data gaps and reasons for missing data. All nodes utilize LoRaWAN for data transmission. Please note that intermittent data gaps may occur due to connectivity issues between the gateway and the nodes, as well as maintenance activities or experimental procedures. Software considerations: The provided R code named “Simple_Dendro_Imputation_and_Analysis.R” is a comprehensive analysis workflow that processes and analyses Eucalyptus growth and environmental data from the "eucalyptus_growth_environment_data_V2.xlsx" dataset. The script begins by loading necessary libraries, setting the working directory, and reading the data from the specified Excel sheet. It then combines date and time information into a unified DateTime format and performs data type conversions for relevant columns. The analysis focuses on a specified device, allowing for the selection of neighbouring devices for imputation of missing data. A loop checks for gaps in the time series and fills in missing intervals based on a defined threshold, followed by a function that imputes missing values using the average from nearby devices. Outliers are identified and managed through linear interpolation. The code further calculates vapor pressure metrics and applies temperature corrections to the dendrometer data. Finally, it saves the cleaned and processed data into a new Excel file while conducting dendrometer analysis using the dendRoAnalyst package, which includes visualizations and calculations of daily growth metrics and correlations with environmental factors such as vapour pressure deficit (VPD).

Steps to reproduce

Eucalyptus plantations are vital for industries globally, providing resources like renewable energy, recyclable fibers, and eco-friendly wood products. To promote sustainable management, ten wireless dendrometer and environmental sensor systems were deployed on Eucalyptus trees—six at Stellenbosch University, South Africa, and four in Leiria, Portugal. These systems measure tree stem growth, air and soil conditions, and transmit data via LoRaWAN to a cloud-based platform (ThingSpeak), with local SD-card backups. Data is collected every 6 minutes by nine systems, and every 11 minutes by one system. The dataset is crucial for understanding forest health, enhancing carbon sequestration, and ensuring sustainable resource use. Managed by the EucXylo Research Chair led by Prof. David Drew and funded by the Hans Merensky Legacy Foundation, the project focuses on the ecophysiology, growth, and wood formation in Eucalyptus trees. The data supports the development of tree growth models and offers high-resolution insights into environmental conditions affecting Eucalyptus growth. Data was collected using ten systems installed on Eucalyptus trees at two locations, measuring tree growth at 1.37m above ground, air temperature and humidity near the canopy, and soil temperature and moisture 5cm below the surface. Data is sent via LoRaWAN to ThingSpeak, with backups on local SD cards. The dataset is stored in the Mendeley Repository and supports studies in sustainable forestry, offering insights into the growth and environmental responses of Eucalyptus trees across different climates. The dataset, Version 5, includes raw data in .xlsx format with sheets for codebook, metadata, and data in CSV format, comprising measurements and metadata for tree growth and environmental conditions. The dataset provides valuable information for sustainable forestry practices, enabling the development of models that simulate tree growth and wood production. It also facilitates comparative studies across regions, helping to optimise forest management strategies. The innovative use of LoRaWAN technology allows for near real-time data transmission, offering almost immediate insights into environmental conditions and tree growth, making it a significant resource for future research in forestry and environmental science.

Institutions

Categories

Funding

Licence