Skip to main content Skip to local navigation

Ecosystem productivity responses in the Bruce Peninsula over the past 20 years

Ecosystem productivity responses in the Bruce Peninsula over the past 20 years

Lord-Emmanuel Achidago at Bruce Peninsula

by Lord-Emmanuel Achidago and Richard Bello

The Bruce Peninsula, renowned for its expansive forest and UNESCO World Biosphere Reserve designation, faced a significant setback with a major wildfire in 1908. Since then, vegetation regrowth has produced trees of up to 114 years old. Concerns about the Peninsula's ability to sequester CO2 amid climate change highlight the need for ecosystem productivity estimates.

In this project, we evaluated gross primary productivity (GPP) models for each season over the past two decades, comparing growth among coniferous forests, croplands, and deciduous forests, representing 55%, 16%, and 14% of the total land area, respectively.


We evaluated two light-use efficiency (LUE) models of Gross Primary Productivity (GPP) dating back to 2000. The first model utilizes data from the Global Orbiting Carbon Observatory-2 Sunlight-Induced Fluorescence (GOSIF) at approximately 5000m, leveraging sunlight-induced fluorescence. The second model, MOD17A2HGF from MODIS (Moderate Image Spectroradiometer), operates at 500m and relies on normalized reflectance indices.

Fig 1. GPP comparisons at 3 locations for each land cover type.

Correlations between modeled GPP and five environmental variables from the ERA-5 weather model (31km) are examined to identify any climate-induced GPP trends over time and the sensitivity of GPP to inter-annual variability, using Pearson correlation coefficients as a standardized measure of their impact. These variables include vapor pressure deficit, volumetric soil water, air temperature, solar radiation, and precipitation. Climate change trend analysis employs non-parametric statistics.

GPP comparisons are conducted across three distinct land cover types: coniferous forests, deciduous forests, and croplands, identified using Land Cover 2020 data (30m). Figure 1 illustrates GPP comparisons at 3 locations for each land cover type. We present seasonal GPP averages for winter, spring, summer, and fall from 2000 to 2020 in Figure 2 (below), alongside corresponding 20-year seasonal means.


The annual average Gross Primary Productivity (GPP) for coniferous forests, croplands, and deciduous forests, as per GOSIF data, are 3.32 gCm-2d-1, 4.17 gCm-2d-1, and 4.84 gCm-2d-1 respectively. Conversely, according to MODIS data, the annual productivity for coniferous forests is 4.03 gCm-2d-1, for croplands is 3.74 gCm-2d-1, and for deciduous forests is 4.21 gCm-2d-1.

Fig 2. Seasonal GPP averages for winter, spring, summer, and fall from 2000 to 2020, alongside corresponding 20-year seasonal means.

When weighted based on the land cover distribution in the Bruce Peninsula, GOSIF indicates sequestration rates of 1.83 gCm-2d-1 for coniferous forests, 0.67 gCm-2d-1 for croplands, and 0.67 gCm-2d-1 for deciduous forests. MODIS data shows sequestration rates of 2.22 gCm-2d-1 for coniferous forests, 0.60 gCm-2d-1 for croplands, and 0.58 gCm-2d-1 for deciduous forests over the past two decades. The estimated annual CO2 uptake for the Peninsula is 1501.2 gCm-2yr-1 by GOSIF and 1457.3 gCm-2yr-1 by MODIS.

Compared with the literature, GPP values in the Bruce Peninsula are slightly larger than other reported ecosystems, likely due to the forests' youth and vigorous growth stage. The ecosystems exhibit significant inter-annual variability in response to temperature and vapour pressure deficit, with coniferous forests showing the most consistent and strongest responses. Seasonal differences in GPP responses are observed, with spring and fall showing the strongest responses across all ecosystems.

Despite any trends in climate, the ability of ecosystems to sequester CO2 over time remains relatively stable. However, significant seasonal GPP responses to climate interannual variability are evident, highlighting the sensitivity of ecosystems to changing environmental conditions.


The ecosystems of the Bruce Peninsula demonstrate performance on par with or exceeding that reported in existing literature. Both models yield similar annual productivity results, with solar-induced chlorophyll fluorescence offering a more direct measure of photosynthesis while models utilizing reflectance indices provide higher spatial resolution. Discrepancies between the models arise predominantly during summertime, coinciding with full foliage, necessitating further investigation to reconcile these disparities. It is noteworthy that satellite sensor-based models are hindered under overcast conditions and necessitate gap-filling methods.

Regarding their ability to sequester atmospheric CO2, the ecosystems exhibit stability over time amidst the impacts of climate change. The lack of statistically significant Gross Primary Productivity (GPP) trends stems from substantial inter-annual variability in climate variables, obscuring any discernible long-term patterns over the past two decades. Notably, coniferous forests display the most robust responses to internal climate variability, particularly outside of winter. Should trends in environmental factors persist during future spring and fall seasons, these forests may gain a competitive advantage over other land cover types.

The overall impact of these ecosystems on CO2 uptake hinges on net ecosystem productivity (NEP), incorporating respiratory losses from vegetation and soils across the Bruce Peninsula, presenting a critical area for future research and investigation.


Lord-Emmanuel Achidago is a master's degree graduate in Physical Geography at EUC. His master's research was on "Quantifying the effect of stressors on vegetation communities in the Bruce Peninsula aimed at assessing the impact of climate stresses on plant communities in the national park." Also a musician who enjoys producing and vegetable gardening, he has been part of the Oral History, Food Justice and Food Making project team led by Professor Honor Ford-Smith. He also worked with Professor Tarmo Remmel on a project that developed R code to identify and characterize land cove patterns that act as fire boundaries from several fire events and is currently working with Professor Adeyemi Olusola in his Humber River Catchment project and as Ravine Team Lead with the Toronto Region Conservation Authority.

Richard Bello is Associate Professor Emeritus at EUC. His research focuses on global/climate change, geography, climate science, northern environments, carbon dynamics, Hudson Bay Lowlands and Toronto's urban environment. He recently developed a new research program examining the hydrology and carbon dynamics of Eastern White Cedar forests of the Bruce Peninsula. Read also his article on Mapping the future of research in the Canadian North.