Mathieu F Bilodeau

Mr. Mathieu F Bilodeau

Dalhousie University
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
UAUsman Ali
Usman Ali
TETravis Esau
Travis Esau
Aitazaz A. FarooqueAitazaz A. Farooque
Aitazaz A. Farooque
QZQamar U. Zaman
Qamar U. Zaman
FAFarhat Abbas
Farhat Abbas
MBMathieu F Bilodeau
Mathieu F Bilodeau
Land use and land cover (LULC) classification maps help understand the state and trends of agricultural production and provide insights for applications in environmental monitoring. One of the major downfalls of the LULC technique is inherently linked to its need for ground truth data to cross-validate maps. This paper aimed at evaluating the efficiency of machine learning (ML) in limiting the use of ground truth data for LULC maps. This was accomplished by (1) extracting reliable LULC information from Sentinel-2 and Landsat-8 s images, (2) generating remote sensing indices used to train ML algorithms, and (3) comparing the results with ground truth data. The remote sensing indices that were tested include the difference vegetation index (DVI), the normalized difference vegetation index (NDVI), the normalized built-up index (NDBI), the urban index (UI), and the normalized bare land index (NBLI). Extracted vegetation indices were evaluated on three ML algorithms, namely, random forest (RF), k-nearest neighbour (K-NN), and k dimensional-tree (KD-Tree). The accuracy of these algorithms was assessed with standard statistical measures and ground truth data randomly collected in Prince Edward Island, Canada. Results showed that high kappa coefficient values were achieved by K-NN (82% and 74%), KD-Tree (80% and 78%), and RF (83% and 73%) for Sentinel-2A and Landsat-8 imagery, respectively. RF was a better classifier than K-NN and KD-Tree and had the highest overall accuracy with Sentinel-2A satellite images (92%). This approach provides the basis for limiting the collection of ground truth data and thus reduces the labour cost, time, and resources needed to collect ground truth data for LULC maps.
1

1 January 2025 in Canadian Journal of Plant Science
Verified
Journal Article
Open access
Mason MacDonaldMason MacDonald
Mason MacDonald
TETravis J. Esau
Travis J. Esau
MBMathieu F Bilodeau
Mathieu F Bilodeau
Balsam fir trees are a specialty agricultural species that are largely used as Christmas trees. Soil or tissue sampling is seldom performed to assess balsam fir nutrition needs, which can contribute to over- or under-fertilization and subsequent environmental challenges. Remote sensing with vegetative indices has had strong relationships with nutrition in other plants. This research aimed to determine whether there is a relationship between normalized difference vegetation index (NDVI) and enhanced vegetation index 2 (EVI2) and nutrition in balsam fir. A random sample of 45 trees was selected in autumn of 2021 and then another 70 trees selected in spring of 2022. Soil and needle tissue samples were analyzed for N, P, K, Ca, Mg, S, Fe, Zn, Mn, B, and Cu. An uncrewed aerial vehicle equipped with a multispectral camera was flown over the orchard to determine the NDVI/EVI2 for each tree. Data were analyzed for correlations through all sampled trees. Separate correlation analyses were performed for mature trees (greater than 5 years old) and young trees (less than 5 years old). Soil nutrients N, P, and K had a significant relationship with NDVI in autumn sampling. Only tissue N was correlated consistently with NDVI in both autumn and spring, with R2 of 37% and 33%, respectively. Relationships were much stronger when only mature trees were considered, increasing to 60% and 51%, respectively. Tissue N was correlated consistently with EVI in both autumn and spring, with R2 of 58% and 40%, respectively, but had no improvement in mature trees. Ultimately both NDVI and EVI2 had predictive power for tissue N that can be useful to producers to help optimize fertilizer input. However, further work is needed to refine these relationships.

8 April 2025 in Agriengineering
Verified
Journal Article
Open access
PDF file
MBMathieu F Bilodeau
Mathieu F Bilodeau
TETravis Esau
Travis Esau
QZQamar U. Zaman
Qamar U. Zaman
BHBrandon Heung
Brandon Heung
Aitazaz A. FarooqueAitazaz A. Farooque
Aitazaz A. Farooque
Excess water in agricultural fields can significantly limit crop productivity. Drone technology offers solutions for identifying and predicting drainage degradation. This study utilized drone-based photogrammetry to create high-resolution elevation models, multispectral imagery for vegetation indices, and flood simulations models to identify zones at risk of poor surface drainage. These models, collected from 2021 to 2023, were used to assess the relationship between poor drainage and corn productivity. The findings revealed a substantial decline in productivity in poorly maintained surface drainage areas, notably a decrease in mean plant height from 1.43 m in 2022 to 0.26 m in flood-prone areas in 2023. Flood-prone zones expanded significantly, from 37% to 61% of the field area between 2022 and 2023, emphasizing the negative cumulative impacts of pre-existing drainage issues. Conversely, fields receiving regular annual maintenance showed an increase in mean plant heights (from 2.23 m to 2.54 m) and NDVI values, reflecting improved drainage conditions. This research demonstrates the effectiveness of drone-derived elevation models for proactively identifying problematic drainage areas, allowing farmers to make informed decisions regarding field maintenance. Implementing these technologies can optimize drainage management practices, enhance agricultural productivity, and increase economic viability in regions that rely on surface drainage.

Verified
Journal Article
Open access
PDF file
MBMathieu F Bilodeau
Mathieu F Bilodeau
TETravis Esau
Travis Esau
Aitazaz A. FarooqueAitazaz A. Farooque
Aitazaz A. Farooque
QZQamar U. Zaman
Qamar U. Zaman
BHBrandon Heung
Brandon Heung
Dykelands are agricultural ground protected from coastal inundation by dyke infra-structure and constitute some of the most agriculturally productive lands in Nova Scotia. Between 2015 and 2019, Canada’s Annual Crop Inventory was used to characterize and estimate hectares of agricultural dykelands cultivated in Nova Scotia. The number of hectares of wheat, barley, corn, forages and soybeans were compiled for each year and compared to the previous year. This was accomplished using GIS software, satellite images, and geodata from the Nova Scotia’s Land Property Database. Results revealed that from 2015 to 2019, an average of 56% of the dykelands’ total surface was dedicated to the production of field crops (wheat, barley, corn, soybeans) and forage. Results also highlighted the importance of forage production on the dykelands. Forage was the largest commodity grown, representing around 80% of the total crop land area of the agricultural dykelands. Corn and soybeans were the second and third crops of abundance, constituting 12 and 4% of the total crop land area, respectively. This study represents the first attempt to document the number of hectares of the principal crops grown on Nova Scotia’s dykelands using crop inventory and property boundaries. Given the predictions of rising sea levels and the overtopping risks that the dykelands face, this study will facilitate more suitable land-use policies by providing stakeholders with an accurate quantitative assessment of the utilization of agricultural dykelands.

1 December 2025 in Environmental Research Communications
Verified
Journal Article
Open access
CDCorey Dawson
Corey Dawson
MBMathieu F Bilodeau
Mathieu F Bilodeau
KZKai Zuo
Kai Zuo
BHBrandon Heung
Brandon Heung
TETravis Esau
Travis Esau
As climate change contributes to higher intensity storms, riparian canopy loss may have geomorphic and ecological consequences on streambank sediment dynamics. This study investigates changes of riparian canopy cover, river morphology responses, and hydrological conditions, following an extratropical cyclone in Atlantic Canada. Using LiDAR data acquired from three timesteps (2013, 2019 and 2024), the spatiotemporal changes in canopy cover and streambank surfaces along a ∼5,100 m segment of two rivers were assessed. The Portapique and Great Village watersheds provide critical habitats for the endangered Inner Bay of Fundy (iBoF) Atlantic salmon and further research is required to support recovering populations. Both rivers experienced canopy cover declines of 19% between 2019 and 2024, following a period of canopy growth—likely driven by Hurricane Fiona in 2022. We applied a DTM of Difference (DoD) technique to investigate streambank changes and found aggradation increased by an average of 91.5% during the 2019–2024 period. Historical hydrometric data was used as reference data to understand how streambank morphology may respond differently to peak flow under canopy loss, compared to a period of growth (2013–2019). Our approach provides a novel framework for assessing climate-driven riparian changes through statistical and spatial analysis to support adaptive planning for salmon recovery under increasing storm intensities. A morphodynamic model was also developed as a proof-of-concept to simulate river morphology responses under peak flow conditions, providing a template for future research where topo-bathymetric data is available.

29 November 2025 in ISPRS Open Journal of Photogrammetry and Remote Sensing
Verified
Journal Article
Open access
MBMathieu F Bilodeau
Mathieu F Bilodeau
TETravis Esau
Travis Esau
Mason MacDonaldMason MacDonald
Mason MacDonald
AFAitazaz A. Farooque
Aitazaz A. Farooque