2026/05/26 by Jeremy B. Yoder, Colin J. Carlson, Christopher W. Callahan · 1 voice
Environmental Science · Agricultural and Biological Sciences · #Species Distribution and Climate Change #Plant Physiology and Cultivation Studies #Remote Sensing in Agriculture
paper · doi:10.64898/2026.05.22.727294
Climate change is expected to touch every ecological community on the planet, but connecting climate change to specific ecological events remains challenging. The emerging practice of climate change attribution can identify how anthropogenic climate change contributes to extreme weather events—but has not yet been applied to unusual ecological events. Here, we demonstrate attribution of a recent, striking ecological anomaly: Joshua trees ( Yucca brevifolia and Y. jaegeriana ) flowering in October 2025, fully four months earlier than normal. We used crowdsourced records of Joshua tree flowering to train a machine learning model that recovers weather triggers of flowering, and successfully predicts regular-seasonal and out-of-season flowering events. We then simulated weather without human-caused climate change using the output of 10 global climate models, and projected Joshua tree flowering under those counterfactual conditions with our trained model. Surprisingly, we found out-of-season blooms were driven by high winter rainfall, not rising temperatures—and therefore, are probably the result of natural weather variability. Our results place out-of-season flowering in context with climate change threats facing Joshua trees, while providing a prototype attribution analysis of an extreme ecological event and identifying priorities for future climate modeling.