Nationwide operational mapping of grassland first mowing dates combining machine learning and Sentinel-2 time series

Grassland dynamics are modulated by management intensity and impact overall ecosystem functioning. In mowed grasslands, the first mowing date is a key indicator of management intensification. The aim of this work was to assess several supervised regression models for mapping grassland first mowing d...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Remote sensing of environment Jg. 315; S. 114476
Hauptverfasser: Rivas, Henry, Touchais, Hélène, Thierion, Vincent, Millet, Jerome, Curtet, Laurence, Fauvel, Mathieu
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Inc 15.12.2024
Elsevier
Schlagworte:
agroecology
biodiversity
data collection
Deep-learning models
ecosystems
environment
France
Grassland management intensification
grasslands
Machine Learning
Mowing dates mapping
neural networks
prediction
Regression
Satellite image time series
Statistics
time series analysis
France
Mowing dates mapping
Satellite image time series
Regression
Deep-learning models
Grassland management intensification
ISSN:0034-4257, 1879-0704
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Grassland dynamics are modulated by management intensity and impact overall ecosystem functioning. In mowed grasslands, the first mowing date is a key indicator of management intensification. The aim of this work was to assess several supervised regression models for mapping grassland first mowing date at national-level using Sentinel-2 time series. Three deep-learning architectures, two conventional machine learning models and two threshold-based methods (fixed and relative) were compared. Algorithms were trained/calibrated and tested from field observations, using a spatial cross-validation approach. Our findings showed that time aware deep-learning models – Lightweight Temporal Attention Encoder (LTAE) and 1D Convolutional Neural Network (1D-CNN) – yielded higher performances compared to Multilayer Perceptron, Random Forest and Ridge Regression models. Threshold-based methods under-performed compared to all other models. Best model (LTAE) mean absolute error was within six days with a coefficient of determination of 0.52. Additionally, errors were accentuated at extreme (late/early) mowing dates, which were underrepresented in the data set. Oversampling techniques did not improve predicting extreme mowing dates. Finally, the best prediction accuracy was obtained when the number of clear dates surrounding the mowing event was greater than 2. Our outputs evidenced time aware deep-learning models’ potential for large-scale grassland first mowing event monitoring. A national-level map was produced to support bird-life monitoring or public policies for biodiversity and agro-ecological transition in France. [Display omitted] •Estimation of grassland first mowing date using regression algorithm.•Time aware deep-learning architectures were the most accurate models.•Threshold-based methods under-performed compared to all supervised models.•LTAE performed reliably across all unknown sites, demonstrating transferability.•Oversampling techniques did not improve predictions accuracy of extreme mowing dates.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0034-4257
1879-0704
DOI:10.1016/j.rse.2024.114476