Each year numerous students contribute to the CSAFM-SCMAF meetings by presenting research related to their studies. Student award winners are profiled here. Thanks to Campbell Scientific Inc. for sponsoring the Bert Tanner Student Presentation award.

Christopher Schulze, University of Alberta
Campbell Scientific Inc. Bert Tanner Oral Presentation Award

Effects of wildfire and permafrost thaw on ecosystem fluxes from boreal peatland complexes in northwestern Canada

Christopher Schulze, David Olefeldt, Craig Emmerton, Lorna Harris, Natascha Kljun, Laura Chasmer, Chris Hopkinson, Matteo Detto, Manuel Helbig, Gabriel Hould Gosselin, and Oliver Sonnentag

Disturbance regimes including wildfire and permafrost thaw are intensifying across the boreal biome of northwestern Canada in response to climate change. Atmospheric exchanges of potent greenhouse gases such as water vapor, carbon dioxide (CO2), methane (CH4) affected by wildfire and permafrost thaw can potentially drive positive climate system feedbacks. The southern Taiga Plains ecozone is characterized by heterogenous boreal peatland complexes particularly vulnerable to intensifying

disturbance regimes. Here, we present net fluxes of water vapor, CH4, and CO2 measured with the eddy covariance technique at three boreal peatland complexes with different disturbance histories near the southern limit of permafrost. Located in the sporadic permafrost zone, the two recently burned southern sites in Lutose, AB, acted as major CO2 sources. The two other sites mainly differed in permafrost extent, ranging from sporadic to discontinuous, and in peat plateau-to-wetland ratio and corresponding forest cover. CO2 uptake, CH4 emissions, and latent heat fluxes were larger at the sporadic than at the discontinuous permafrost site, whereas both, gross primary productivity and ecosystem respiration, as well as sensible heat fluxes were higher at the northern site. In conclusion, further permafrost thaw will only have negligible effects on the net ecosystem productivity of boreal peatlands in the Taiga Plains’ discontinuous permafrost zone when compared to both, the positive net radiative forcing of increased CH4 emissions following permafrost thaw and the large CO2 losses post wildfire.

Nataša Popović, University of Waterloo
CSAFM-SCMAF Outstanding Oral Presentation Award

Ecosystem-scale surface energy fluxes of a newly constructed boreal upland-fen watershed

Nataša Popović, Richard M. Petrone, Adam Green, Myroslava Khomik, Jonathan S. Price

Obligatory mine reclamation efforts in the Athabasca oil sands region require leased land to be returned to pre-disturbance capacity. This entails the re-establishment of key ecosystem services such as carbon sequestration and water and nutrient cycling. Surface-atmosphere exchanges of energy, heat and moisture are critical to ecohydrological processes and feedbacks that govern ecosystem functionality. Eddy covariance and remote sensing measurements were used to examine the temporal changes in energy partitioning of a reclaimed fen-upland watershed during the initial seven years post-construction (2013 – 2019). Surface conditions evolved from bare ground to robustly vegetated plant communities, altering albedo, surface roughness and plant-mediated shading. Initially in the fen, due to near-surface water table, high soil moisture content and low albedo (0.09) of bare peat, surface evaporation

was high and most energy was partitioned to latent heat (QE ) with a high degree of decoupled surface-atmosphere conditions (surface-atmosphere decoupling factor (Ω) of 0.49). As the plant community developed and stabilized, QE remained the dominant energy flux, but a larger proportion of available energy was partitioned to sensible heat flux (QH) than in previous years. Moreover, effects of plant-mediated sheltering and shading decreased surface evaporative losses and there was an increase in surface-atmosphere coupling (Ω = 0.32), particularly during drier periods. In the drier upland, plant emergence and establishment was slower than in the fen. Initially, bare ground in the upland resulted in higher albedo (0.20-0.25), and QE and Qwere similar. Once understory and treed species matured, QE increased with seasonal trends that mirrored plant phenology. Unlike the fen, the upland continually exhibited a higher degree of surface-atmosphere coupling (Ω ranging between 0.23-0.3). By year seven (2019) energy fluxes and intra- and inter-seasonal trends at the constructed watershed were comparable to natural and post-disturbance boreal landscapes indicating the system is evolving towards a functional, reclaimed ecosystem.

Hayden Tackley, Dalhousie University
CSAFM-SCMAF Outstanding Poster Presentation Award

Impacts of managed dike realignment on soils and shallow groundwater

Coastal defense structures (e.g., dikes, seawalls) protect vulnerable communities along marine coastlines and estuaries from the physical and chemical influences of adjacent water bodies. However, these structures are susceptible to overtopping or breaching by tides and waves potentially amplified by climate change induced sea level rise. Repeated inundation by salinized water can contaminate freshwater resources and salinize soil, impacting land use activities,

including agricultural productivity. Managed ecosystem-based dike realignment approaches and salt marsh restoration can provide alternatives to traditional mitigation approaches; however, these strategies often permanently sacrifice coastal land to marine influence following flooding events. This research assesses the changes to soil salinity at a managed dike realignment project prior to the transformation from diked agricultural land to an estuarine environment. Geophysical surveys, soil/sediment sampling, piezometer wells, UAV surveys, and soil moisture monitoring were used to compare baseline data collected before the realignment and during 8-10 months of intermittent flooding (i.e., spring tides every two weeks). The formerly freshwater environment (bulk electrical conductivity ~300 µS/cm) was shown to transform into a moderately saline environment (~6000 µS/cm) following less than twenty flood events and sediment deposition occurred at high rates with up to 4 cm of marine sediment deposited per flood. The result of this flooding was the transformation from agricultural land to tidal mudflats. The realignment zone serves as an engineered analog of coastal flooding. This study presents an opportunity to examine the impacts of salinized water on agricultural soils, which, due to increased coastal flooding, is expected to impact more areas in the future. Results demonstrate that a small number of flooding events by even moderately saline water can cause root zone salinization and critically reduce crop productivity in coastal areas.

Mariaelisa Polsinelli, McGill University
CSAFM-SCMAF Student Travel Award Winner

A Case Study in Prince Edward Island, Canada: Comparing Process Based Models and Machine Learning for Potato Yield Prediction

Mariaelisa Polsinelli, Morteza Mesbah, Zhiming Qi, Matt Ramsay

Climate change is already affecting the Canadian climate with increases in air temperature, evapotranspiration rate and risk of rainstorms and drought. Potato, a major crop in Prince Edward Island (PEI), Canada, contributed 10.8% of the provinces GDP in 2018. However, potato yields are very sensitive to these changes in weather and an increase of 1−1.4°C could reduce yields by 18-32%. The ability to predict yield before harvest is invaluable information for both decision makers in charge of national food security strategies, and farmers who need to plan their on-farm actions to maximize yield under changing conditions. Process based models and machine learning (ML) are being increasingly applied for yield prediction. The process based model STICS (Simulateur mulTIdisciplinaire pour les Cultures Standard) version 9.2 has been tested to predict yield, total and plant nitrogen, total biomass, and leaf area index, using

multi-year data from research farms in Eastern Canada. This calibrated model was tested to predict yield on commercial farms with multiple field years in Prince County, PEI. The commercial farm data includes detailed management practices and gridded yield map data. A subset of field years with in-field soil samples was taken for STICS and to train Random Forests (RF) for yield prediction. The STICS model required some recalibration but outperformed RF overall with RRMSEs of 12.8% and 16.6% respectively, though the model with the best performance for individual field years varied. Notably, STICS predicted a very dry year accurately and RF was able to predict the yield of a field year using only past-year training data quite well (RRMSE of 11.8%). Though more investigation is needed, these preliminary results show that both crop model and ML can play a part in accurate yield predictions for commercial farms within a changing climate.

Roya Faramarzi, University of Alberta
Campbell Scientific Inc. Bert Tanner Poster Presentation Award

How do management practices across ecoregions impact soil microbial diversity and aggregate stability indicators of soil health?

Roya Faramarzi, M Derek MacKenzie, Monika Gorzelak

Demand for agricultural products with increasing global population drives the use of intensive cultivation practices, which will have negative impacts on soil health and water quality. Understanding the links between indicators of soil health, such as soil aggregate stability, soil organic carbon (TOC), and soil microbial community structure and diversity, are critical for creating sustainable agricultural soil systems. This study will assess how various environmental and land management conditions might influence soil health indicators.

We will evaluate how these management practices affect soil health indicators across agricultural ecoregions in Alberta. The results will inform land users and agricultural stakeholders on more sustainable management of their lands in the future.

Preliminary results showed a significant interaction between management type and soil organic carbon and nitrogen. However, conventional agriculture negatively impacted the TOC and TN, and regenerative agriculture positively influenced the TOC and TN. Soil properties such as soil bulk density (BD) influenced TOC and TN under different management practices. A decrease in BD increased TOC and TN in regenerative agriculture. This could be explained by differences in conventional vs. regenerative farming, where decreased bulk density indicates better aggregation. Also, decreased BD was correlated to an increase in SOC. The silt to Clay ratio and pH had the lowest impact on TOC and TN among soil properties. For the next steps, soil microbial diversity analysis and soil physical properties under different climate conditions will be done, and the results will be discussed.