Code for the analysis of Tree-Quest measurements collected in Laxenburg Park

This repository contains code necessary to reproduce the results published in (!!! the link will be added soon !!!):
Manuscript title: Tree-Quest: A Citizen Science App for Collecting Single-Tree Information Journal: Ecological Informatics Status: under review

The input datasets are published and can be downloaded from this Zenodo repository

DBH/TH Analysis

Script #1: analyse_DBH_highQuality.py or analyse_TH_highQuality.py

The script analyzes the performance of Tree-Quest’s High-quality data wrt. FI nd TLS measurements.

• Input Datasets:
  • ARD-LP: ard_gdf_2024_03_07_LaxPark.parquet
  • Quality DBH: quality_{dbh/th}_LaxPark.parquet
• Output Datasets:
  • H-LP: qualHigh_outFiltered_{dbh/th}_LaxPark
• Output Figures:
  • Figure 4 for DBH / Figure 5 for TH in the paper

Script #2: analyse_DBH_OtherQualityCats.py or analyse_TH_OtherQualityCats.py

The script analyzes the performance of Tree-Quest’s High-, Medium- and Low-quality datasets wrt. FI measurements.

• Input Datasets:
  • H-LP: qualHigh_outFiltered_{dbh/th}_LaxPark.parquet
  • Non H-LP: not_qualHigh_{dbh/th}_LaxPark.parquet
• Output Datasets:
  • M-LP: qualMedium_outFiltered_{dbh/th}_LaxPark.parquet
  • L-LP: qualLow_outFiltered_{dbh/th}_LaxPark.parquet
• Output Figures:
  • Figure 6 for DBH / Figure 7 for TH in the paper

Script #3: assign_DBH_UserName.py or assign_TH_UserName.py

This script assigns the citizen category (Expert, Practitioner, Student) to each measurement. The code and the User Info input files for this script cannot be shared, as this data would expose personal user information, which, according to GDPR and the Geo-Quest user agreement, we are not allowed to share. Nevertheless, the anonymized output dataset (HML-LP) is shared to ensure reproducibility of the results.

• Input Datasets:
  • H-LP: qualHigh_outFiltered_{dbh/th}_LaxPark.parquet
  • M-LP: qualMedium_outFiltered_{dbh/th}_LaxPark.parquet
  • L-LP: qualLow_outFiltered_{dbh/th}_LaxPark.parquet
  • User Info files
• Output Datasets:
  • HML-LP: qHML_{DBH/TH}_LaxPark_UserGroups.parquet

Script #4: analyse_DBH_UserGroups.py or analyse_TH_UserGroups.py

The script analyzes the performance of Tree-Quest’s Experts-, Practitioners-, and Students measurements wrt. FI measurements.

• Input Datasets:
  • HML-LP: qHML_{DBH/TH}_LaxPark_UserGroups.parquet
  • Non H-LP: not_qualHigh_{dbh/th}_LaxPark.parquet
• Output Figures:
  • Figure 8 for DBH / Figure 9 for TH in the paper

Script #5: analyse_DBH_Method_highQuality.py

The script analyzes the performance of Tree-Quest’s Automatic and Manual DBH methods, comparing with FI measurements.

• Input Datasets:
  • H-LP: qualHigh_outFiltered_dbh_LaxPark.parquet
• Output Figures:
  • Figure 10 in the paper

Script #6: analyse_DBH_Stadtpark.py or analyse_TH_Stadtpark.py

The script compares the Tree-Quest, GreenLens, Working Trees, and GLOBE Observer measurements wrt. FI measurements.

• Input Datasets:
  • ARD-SP: ard_gdf_2024_04_14_Stadtpark.parquet
• Output Figures:
  • Figure 11 for DBH / Figure 12 for TH in the paper

Datasets Overview

Dataset Name	Test Site	Description	File Name
ARD-LP	Lax. Park	Analysis-ready dataset containing all Tree-Quest observations in Laxenburg Park	ard_gdf_2024_03_07_LaxPark.parquet
Quality Data - LP	Lax. Park	The quality scores (High, Medium, Low) for the Laxenburg park data for DBH and TH measurements	quality_dbh_LaxPark.parquet quality_th_LaxPark.parquet
H-LP	Lax. Park	High-quality Tree-Quest measurements	qualHigh_dbh_LaxPark.parquet qualHigh_th_LaxPark.parquet
Not H-LP	Lax. Park	Tree-Quest measurements other than high-quality measurements	not_qualHigh_dbh_LaxPark.parquet not_qualHigh_th_LaxPark.parquet
M-LP	Lax. Park	Medium-quality Tree-Quest measurements	qualMedium_outFiltered_dbh_LaxPark.parquet
L-LP	Lax. Park	Low-quality Tree-Quest measurements	qualLow_outFiltered_dbh_LaxPark.parquet
HML-LP	Lax. Park	High-, Medium-, and Low-quality measurements with assigned Citizen groups (Experts, Practitioners, and Students)	qHML_DBH_LaxPark_UserGroups.parquet qHML_TH_LaxPark_UserGroups.parquet
ARD-SP	Stadtpark	Analysis-ready dataset containing Tree-Quest, GreenLens, Working Trees, and GLOBE Observer app observations in Stadtpark, Vienna	ard_gdf_2024_04_14_Stadtpark.parquet