A multi-variable and matrix-based assessment of elevation effects on vegetation dynamics and climate trends in a semi-arid environment

In this paper, we explore the relationships between vegetation dynamics and climatic trends across elevations in the semi-arid Tuzluca district of eastern Türkiye over the period 2002-2023. Using MODIS-derived NDVI and LAI datasets and ERA5 Land derived climate variables (temperature, precipitation, and soil moisture), we applied a matrix-based, multi-variable analytical framework composed of Mann–Kendall (MK) trend tests and Sen's slope estimates across elevation bands of 100 m. While temperature showed strong positive trends across almost all altitude zones during the April–October period, soil moisture exhibited a significant decrease during the same period (May–October) and largely at altitudes above 2400 m. Precipitation, on the other hand, showed a distinctly negative trend in April across all altitudes, while a decreasing trend was observed in July at altitudes above 1800 m and in October at both the lowest and highest altitudes. Vegetation indices have shown that the growing season starts earlier and lasts longer at lower altitudes. Specifically, the growing season spans March to November at altitudes between 900 and 1100 m, while at altitudes above 2800 m, this period is limited to June-September. The record-long vegetation growing season observed in 2010 was accompanied by positive temperature and precipitation anomalies. This suggests that climate extremes can act as a positive driving force for vegetation response in different altitude regions. The study highlights the significance of elevation in regulating vegetation–climate interactions and emphasizes the need to integrate topographic variability into climate adaptation and ecological monitoring strategies in semi-arid mountain regions.