Natural and anthropogenic factors affect forest structure worldwide, primarily affecting forest canopy and its light properties. However, not only stand-replacing events modify canopy structure, but disturbances of lower intensity can also have important ecological implications. To study such effects, we analyzed long-term changes in light properties of a conifer–broadleaf mixed forest in the Southwestern Pyrenees, placed in the fringe between the Mediterranean and Eurosiberian biogeographical regions. At this site, a thinning trial with different intensities (0%, 20%, and 30–40% basal area removed) took place in 1999 and 2009, windstorms affected some plots in 2009 and droughts were recurrent during the sampling period (2003, 2005, 2011). We monitored light properties during 14 years (2005–2019) with hemispherical photographs. We applied partial autocorrelation functions to determine if changes between years could be attributed to internal canopy changes or to external disturbances. In addition, we mapped the broadleaf canopy in 2003, 2008, and 2016 to calculate broadleaf canopy cover and richness at the sampling points with different buffer areas of increasing surface. We applied generalized linear mixed models to evaluate the effects of light variables on canopy richness and cover. We found that light variables had the most important changes during the period 2008 to 2010, reacting to the changes caused that year by the combined effects of wind and forest management. In addition, we found that an area of 4.0 m radius around the sampling points was the best to explain the relationship between light properties and species richness, whereas a radius of 1.0 m was enough to estimate the relationship between light and canopy cover. In addition, light-related variables such as diffuse light and leaf area index were related to species richness, whereas structural variables such as canopy openness were related to canopy cover. In summary, our study demonstrates that non stand-replacing disturbances such as windstorms, thinning, or droughts can have an important role in modifying structural and light-related canopy properties, which in turn may influence natural processes of stand development and ecological succession.

Measuring plant water use is crucial for understanding vegetation roles in hydrological cycles and forest-atmosphere interactions. However, stand-scale, long-term sap flow monitoring is often constrained by high sensor costs, energy requirements, and maintenance. Here, we present a self-built, low-cost, energy-efficient heat pulse method (HPM) sap flow monitoring system comprising 80 sensors and evaluate its multi-year performance and applicability of the dataset, across eight dominant tree species in a subtropical plantation forest in eastern Taiwan equipped with an eddy covariance tower.The compact (600 g), Li-ion battery-driven HPM system consumed 43.3 mW on average, enabling autonomous operation for ~28 days per battery cycle (maximum 44 days) without noticeable loss of measurement precision. Across three years (2022–2024), the 80 sensors successfully captured species-specific and stand-scale sap flow temporal dynamics. Evergreen/semi-deciduous species (Cinnamomum camphora, Zelkova serrata, Bischofia javanica, Fraxinus griffithii) maintained year-round transpiration, while deciduous species (Liquidambar formosana, Koelreuteria elegans, Melia azedarach, Pterocarpus indicus) showed pronounced seasonal dormancy. Data from the HPM and previously installed thermal dissipation method (TDM) sensors were compared with the eddy covariance tower, showing similar correlation patterns (ρ = 0.79–0.91) and confirming the HPM measurements’ reliability. Clustering analyses revealed functional variation within the community. After determining the optimal sample size, sap flow averages from random subsamples also provided physiologically grounded gap-filling of ET data, highlighting potential applications of this system.Overall, this study demonstrates that dense, low-cost, energy-efficient HPM networks can capture multi-species transpiration dynamics, support stand-scale ecohydrological analyses, and provide a practical pathway for investigating ecosystem water use and resilience in high-diversity forests.

Measuring plant water use is crucial for understanding vegetation roles in hydrological cycles and forest-atmosphere interactions. However, stand-scale, long-term sap flow monitoring is often constrained by high sensor costs, energy requirements, and maintenance. Here, we present a self-built, low-cost, energy-efficient heat pulse method (HPM) sap flow monitoring system comprising 80 sensors and evaluate its multi-year performance and applicability of the dataset, across eight dominant tree species in a subtropical plantation forest in eastern Taiwan equipped with an eddy covariance tower.The compact (600 g), Li-ion battery-driven HPM system consumed 43.3 mW on average, enabling autonomous operation for ~28 days per battery cycle (maximum 44 days) without noticeable loss of measurement precision. Across three years (2022–2024), the 80 sensors successfully captured species-specific and stand-scale sap flow temporal dynamics. Evergreen/semi-deciduous species (Cinnamomum camphora, Zelkova serrata, Bischofia javanica, Fraxinus griffithii) maintained year-round transpiration, while deciduous species (Liquidambar formosana, Koelreuteria elegans, Melia azedarach, Pterocarpus indicus) showed pronounced seasonal dormancy. Data from the HPM and previously installed thermal dissipation method (TDM) sensors were compared with the eddy covariance tower, showing similar correlation patterns (ρ = 0.79–0.91) and confirming the HPM measurements’ reliability. Clustering analyses revealed functional variation within the community. After determining the optimal sample size, sap flow averages from random subsamples also provided physiologically grounded gap-filling of ET data, highlighting potential applications of this system.Overall, this study demonstrates that dense, low-cost, energy-efficient HPM networks can capture multi-species transpiration dynamics, support stand-scale ecohydrological analyses, and provide a practical pathway for investigating ecosystem water use and resilience in high-diversity forests.

Measuring plant water use is crucial for understanding vegetation roles in hydrological cycles and forest-atmosphere interactions. However, stand-scale, long-term sap flow monitoring is often constrained by high sensor costs, energy requirements, and maintenance. Here, we present a self-built, low-cost, energy-efficient heat pulse method (HPM) sap flow monitoring system comprising 80 sensors and evaluate its multi-year performance and applicability of the dataset, across eight dominant tree species in a subtropical plantation forest in eastern Taiwan equipped with an eddy covariance tower.The compact (600 g), Li-ion battery-driven HPM system consumed 43.3 mW on average, enabling autonomous operation for ~28 days per battery cycle (maximum 44 days) without noticeable loss of measurement precision. Across three years (2022–2024), the 80 sensors successfully captured species-specific and stand-scale sap flow temporal dynamics. Evergreen/semi-deciduous species (Cinnamomum camphora, Zelkova serrata, Bischofia javanica, Fraxinus griffithii) maintained year-round transpiration, while deciduous species (Liquidambar formosana, Koelreuteria elegans, Melia azedarach, Pterocarpus indicus) showed pronounced seasonal dormancy. Data from the HPM and previously installed thermal dissipation method (TDM) sensors were compared with the eddy covariance tower, showing similar correlation patterns (ρ = 0.79–0.91) and confirming the HPM measurements’ reliability. Clustering analyses revealed functional variation within the community. After determining the optimal sample size, sap flow averages from random subsamples also provided physiologically grounded gap-filling of ET data, highlighting potential applications of this system.Overall, this study demonstrates that dense, low-cost, energy-efficient HPM networks can capture multi-species transpiration dynamics, support stand-scale ecohydrological analyses, and provide a practical pathway for investigating ecosystem water use and resilience in high-diversity forests.