http://hdl.handle.net/11375/30348

Hussain, Nur (2024) Evaluating Carbon Dynamics of Agricultural and Managed Forest Ecosystems Using Remote Sensing, MacSphere Open Access Dissertations and Theses, http://hdl.handle.net/11375/30348

Climate change is significantly impacting terrestrial ecosystems through altered environmental conditions and enhanced atmospheric CO2, which has increased by about 40% since the pre-industrial period. Forests, covering about 30% of global land, play a vital role in the carbon (C) cycle by sequestering 15.6 billion tonnes of CO2 equivalent annually from the atmosphere. The agricultural sector is also a major contributor to atmospheric CO2, however, it has a net source of C and emits 9.3 billion tonnes of CO2 equivalent annually, which is about 11% of global emissions. In recent years, advanced remote sensing and eddy covariance (EC) techniques have become vital systems for assessing CO2 exchanges and providing real-time monitoring capabilities for both forest and agricultural ecosystems. Integration of these techniques would enhance our understanding of C exchanges and their major controls in these ecosystems. It will help to explore how these ecosystems may be impacted by climate change and natural and human-induced disturbance events and develop climate-tailored forest and agricultural management practices to increase their carbon sequestration capabilities. In this study, high-resolution remote sensing and EC flux measurements made in an agricultural field in Southern Ontario, Canada were used to determine C sequestration or loss capabilities of different crops in the Great Lakes region and explore how they may be impacted by extreme weather events. This study also explored the best forest management practices that can be adopted to enhance carbon sequestration in the temperature conifer plantation forests in the agricultural landscape of Southern Ontario. Furthermore, it determined how the native (deciduous) forest ecosystems of the Great Lakes region may be impacted by natural disturbances (i.e. insect infestation). These agricultural and forest sites are part of the Turkey Point Environmental Observatory (TPEO) and associated with the Global Water Futures (GWF) program, US-Canada Global Centre for Climate Change and Transboundary Waters, Ameriflux and the global Fluxnet. In the agricultural site, EC fluxes were continuously measured from 2020 to 2023, when the site was planted with corn, sweet potato and tobacco crops. Net ecosystem production (NEP) of the agricultural site was 485 (corn), 249 (corn), -120 (sweet potato) and 7 (tobacco) g C m⁻² yr⁻¹, respectively from 2020 to 2023. The reduction in NEP for corn in 2021 can be attributed to both the drought conditions in May and August, where precipitation was significantly below the 30-year normal (38 mm in May and 46 mm in August), causing stress during critical growth periods, and the inherent differences in carbon dynamics associated with crop types and their responses to climatic extremes. Corresponding annual evapotranspiration (ET) values were 680, 727, 732 and 715 mm yr⁻¹, which accounted for approximately 60%, 72%, 77% and 73% of the annual total precipitation. Study results showed that overall, the site was a net C sink when corn was planted in 2020 and 2021, a net source of C when sweet potatoes were planted in 2022 and C neural when tobacco was planted in 2023. The grain yields (GY) were 537, 491, 118 and 124 g C m⁻² in 2020, 2021, 2022 and 2023 resulting in annual net ecosystem carbon balance (NECB) of -52 (corn), -242 (corn), -238 (Sweet potato) and -117 (tobacco) g C m⁻² yr⁻¹. High-resolution Sentinel-2 satellite (10 × 10 m²) and drone-observed remote sensing data along with EC fluxes were used to evaluate the effect of five different variable retention harvesting (VRH) treatments on the growth and C uptake of a 90-year-old red pine (Pinus resinosa Ait.) plantation (1931) forest, in Southern Ontario, Canada. VRH emulates natural post-disturbance canopy structures to enhance biodiversity and resilience. Treatments included 33% aggregate (33A), 55% aggregate (55A), 33% dispersed (33D), 55% dispersed (55D), and an unharvested control (CN), each replicated four times in 1 ha plots. From 2010 to 2020, mean daily Normalized Difference Vegetation Index (NDVI) values ranged from 0.25 to 0.86, with 55D showing the highest NDVI values. Satellite-derived annual GPP correlated with observed annual GPP (R² = 0.88, p = 0.032) in an adjacent white pine plantation forest. These GPP estimates indicated that VRH treatment with dispersed residual canopies retaining over half the initial basal area (i.e 55D) was the most optimized management strategy that can be deployed for forest growth and C uptake to mitigate climate change. Overall, the mean annual GPP for the 20-ha site was 1651 ± 89 g C m⁻² yr⁻¹, ranging from 1407 to 1864 g C m⁻² yr⁻¹. Finally, high-resolution Sentinel-2 satellite remote sensing and EC observation were employed to investigate the impact of 2021 spongy moth (Lymantria dispar) infestation on forest productivity and C losses in the deciduous and mixed forests across Southern Ontario. Results showed a significant reduction in leaf area index (LAI) and GPP values. Growing season mean LAI values for deciduous (mixed) forests across the region were 3.66 (3.18), 2.74 (2.64), and 3.53 (2.94) m² m⁻² in 2020, 2021, and 2022, respectively, indicating approximately 24 (14)% reduction in LAI compared to pre- and post-infestation years. Similarly, growing season GPP values in deciduous (mixed) forests across the region were 1338 (1208), 868 (932), and 1367 (1175) g C m⁻², respectively in 2020, 2021, and 2022, showing about 35 (22)% reduction in GPP in 2021 compared to pre- and post-infestation years. This infestation-induced reduction in GPP of deciduous and mixed forests, when upscaled to the whole study area (178,000 km²), resulted in 21.1 (21.4) Mt of C loss compared to 2020 (2022), respectively from Sothern Ontario alone. It shows the large scale of C losses caused by 2021 infestation in Canadian Great Lakes region. This dissertation improved our understanding of C exchanges in the forest and agriculture ecosystem within the Great Lakes region of North America. The methods developed in this study offer valuable tools to assess and quantify C uptake capabilities and natural disturbance impacts on the regional C balance of forest ecosystems by integrating field observations, high-resolution remote sensing data and models. Study results will also help in developing sustainable forest management practices to achieve net-zero C emission goals through nature-based climate change solutions.

https://macsphere.mcmaster.ca/bitstream/11375/30348/2/Hussain_Nur_2024September_PhD.pdf