Abstract
Phenological synchrony, which isthe alignment in timing and duration of biological events between interdependent species, is crucial for maintaining ecosystem health and functioning. However, ongoing climate change is causing differential shifts in species’ timing; as a result, biotic interactions such as between plants and insects may become asynchronous. Shifts in phenological synchrony can impact ecosystem services such as pollination, nutrient cycling, herbivory and food web stability. I investigated phenological synchrony between vegetation and insect abundance within Pretoria National Botanical Gardens (NBG) using remote sensing, DNA barcoding and advanced statistical modelling. Insect species were sampled biweekly using Malaise traps over a 12-month period, from February 1, 2022, to January 31, 2023, and identified using DNA barcoding. Contemporaneous vegetation phenology was recorded using Sentinel-2 satellite imagery, from which the Normalised Difference Vegetation Index (NDVI) was used to assess greenness and plant productivity. Climate data was sourced from the Agricultural Research Council (ARC). Annual and intra-annual trends in vegetation phenology within Pretoria NBG were described using NDVI coverage maps. Temporal overlap between NDVI and insect abundance was quantified using kernel density estimates, and predictive models were applied to assess the influence of climate on both vegetation phenology and insect activity. A total of 51,326 insect individuals were sampled, of which 83.2% were assigned to order, 31.9% to family, 4.7% to subfamily, 1.7% to tribe, 3.3% to genus and 1.6% to species. Temporal overlap analysis revealed strong synchrony between NDVI and insect abundance (R2 = 0.85). Random Forest and XGBoost SHAP analyses highlighted NDVI as the dominant predictor of insect abundance (R2 = 0.74, RSME = 0.29), and minimum temperature as the most influential climatic driver of both vegetation phenology (R2 = 0.48, RSME = 0.05) and insect abundance (R2 = 0.86, RSME = 0.21). A +1.5°C increase in temperature would decrease vegetation phenology and insect abundance during both early and late growing seasons. These findings highlight the need to monitor phenological synchrony to maintain ecosystem health and further emphasise the value of combining remote sensing and molecular tools with statistical modelling to better understand ecological responses to climate change. Such approaches are essential for informing conservation strategies, enhancing ecosystem resilience, and sustaining plant–insect interactions that support key ecosystem services.