Next Article in Journal
Mapping the Distribution of Viruses in Wild Apple Populations in the Southeast Region of Kazakhstan
Previous Article in Journal
Tracing Dam Impacts on Braided Riverbank Vegetation: A Spatiotemporal Analysis of Cover Dynamics and Hydrological Drivers
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 16
Issue 7
10.3390/f16071118
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Phenophases of Mixed-Forest Species Are Regulated by Photo-Hydro-Thermal Conditions: An Approach Using UAV-Derived and In Situ Data

by
Marín Pompa-García
1,*,
Eduardo Daniel Vivar-Vivar
1,
Andrea Cecilia Acosta-Hernández
1 and
Sergio Rossi
2
1
Laboratorio de Dendroecología, Facultad de Ciencias Forestales y Ambientales, Universidad Juárez del Estado de Durango, Río Papaloapan y Blvd. Durango s/n, Col. Valle del Sur, Durango 34120, Mexico
2
Laboratoire sur les Écosystèmes Terrestres Boréaux, Département des Sciences Fondamentales, Université du Québec à Chicoutimi, Chicoutimi, QC G7H 2B1, Canada
*
Author to whom correspondence should be addressed.
Forests 2025, 16(7), 1118; https://doi.org/10.3390/f16071118
Submission received: 8 June 2025 / Revised: 29 June 2025 / Accepted: 3 July 2025 / Published: 6 July 2025
(This article belongs to the Section Forest Meteorology and Climate Change)
Browse Figures
Review Reports Versions Notes

Abstract

Severe drought events have raised concerns regarding their effects on the phenological cycles of forest species. This study evaluates the correspondence between in situ phenophases and those detected by an unmanned aerial vehicle (UAV) in tree species coexisting within a mixed forest, with particular attention to their relationship with climatic variables. Based on 12 consecutive monthly field observations, we compared phenological developments with UAV-derived normalized difference vegetation index (NDVI) values, which were then correlated with environmental variables. The analysis revealed a convergence of inflection points and seasonal phenological shifts, likely driven by climatic factors, although distinct patterns emerged between coniferous and broadleaf species. Photoperiod (PP), vapor pressure deficit (VPD), maximum temperature (TMAX), and, to a lesser extent, precipitation (P) were the primary environmental variables influencing NDVI results, used here as a proxy for phenology. Photothermal conditions revealed seasonal asynchrony in NDVI responses between coniferous and broadleaf species, exerting a positive influence on conifers during summer, while having a negative impact on broadleaf species in spring. Validation of in situ observations with UAV-derived data demonstrated a biological correlation between canopy dynamics and NDVI values, supporting its use as a proxy for detecting phenophases at the level of individual trees.

Graphical Abstract

1. Introduction

Currently, exacerbated climatic variations are impacting vegetation phenological cycles, evident in early canopy activity in spring and extensions in autumn [1]. It is widely accepted that such variations have repercussions on the distribution of carbohydrates within the tree, productivity rates and carbon cycles [2,3]. Therefore, generating scientific evidence of the appearance, magnitude and duration of phenophases and their changing climatic interrelations is a constant concern for the scientific community [4].
Modern phenology has rapidly expanded the use of direct local observations [5], combining these with modern remote sensing techniques on a global scale and developing sophisticated modeling strategies [6,7]. In particular, the advent of unmanned aerial vehicles (hereafter, UAV) improves near real-time monitoring, identifying the normalized difference vegetation index (hereafter, NDVI) as a temporal proxy with which to distinguish phenophases [8]. These series demonstrate recognizable seasonal trends, capturing defined plant phenophases such as leaf-out (green-up), peak vegetation greenness (maturity) and the decline of vegetative activity (senescence). Despite its refined spatial scale, the adjustment between seasonal NDVI inflection points and their temporal correspondences with in situ phenological data remains to be verified [9]. Thus, integrating direct ground observations with UAV-derived high-resolution imagery can enhance the detailed analysis of phenological metrics at the individual tree level, thereby helping to address current challenges [10]. For instance, accurately capturing the phenological dynamics of mixed species is challenging and merits further research [5,7]. Likewise, our knowledge of the complex interrelationships of local meteorological elements that potentially influence the phenophases of mixed species is still fragmented.
Given the strong evidence that increased photoperiod and temperature can play a very important role in phenological responses [11], it is appropriate to conduct these experiments in drought-affected sites [12] where the interspecific spatial distribution of trees is a dominant factor in the phenological variability of each forest stand [5]. The diverse forests of northern Mexico, where a range of species coexist, face episodes of severe drought [13], which affect them with varying degrees of temporal and spatial intensity. Furthermore, drought seasonality is predicted to increase seasonal dryness severity through increased evapotranspiration rates [12], which may drive resource availability over time, leading to different phenophase patterns. Therefore, this seasonal drought-prone area represents an ideal opportunity to better decode the covariant meteorological elements that regulate phenology, since seasonal water shortages exert a key control over canopy activity. Besides its ecological and economic importance and the availability of local in situ data providing comparative advantages [13], this ecosystem serves as an ideal natural laboratory for generating knowledge, since its fluctuating and harsh conditions contribute to our increased understanding of the mechanisms of species coexistence in diverse forests.
The objective of this study is therefore to describe the visible variations in phenology among species cohabiting mixed forests and to determine if there is a link between UAV-derived NDVI values and direct observations of species phenology, as well as their relationship to climate variability that forms the basis for the effects of climate on phenological changes. We addressed the following specific questions: How do the monthly phenophases transition within tree species? Is there a biological correspondence between ground data of phenological events and NDVI values derived from UAVs? How are these driven by climate data? We hypothesized that there is substantial differentiation between species showing temporal coupling across ground-data phenology and those detected by UAVs, which is in turn driven by intra-annual photo-thermic and hydroclimatic drivers.

2. Materials and Methods

2.1. Study Area and Data Collection

As a strategy to achieve the experimental goal, we chose a site with a convergence of mixed-forest species (107.1114° W and 27.1492° N), which presented well-differentiated growth seasons in terms of both hydroclimatic factors and drought sensitivity (ref. [13], see Figure 1).
The study area is located in the Sierra Tarahumara in the southeast of the state of Chihuahua, Mexico, and corresponds to a permanent experimental plot for forest research. The dominant genera in the area are Pinus, Juniperus, Quercus and Arbutus, with an arboreal stratum characterized by a regenerative or juvenile state, with few individuals in high diameter classes. It contains basal area (BA) and volume stocks of 25.16 m2 and 158.86 m3 per hectare. It is a site with considerable diversity (Shannon–Wiener Diversity Index of 1.16) and good species richness (Margalef Index of 1.04). It presents values of aboveground biomass and carbon of 142.78 Mg ha−1 and 71.39 Mg ha−1, respectively, with densities of 2165 trees ha−1. The predominant shrubs and herbs are Ceanothus buxifolius Willd. ex Schult. f., Bouvardia ternifolia (Cav.) Schltdl., Houstonia rubra Cav., Eryngium heterophyllum Engelm., Dysphania graveolents (Lag. & Rodr.) Mosyakin & Clemants, Bouteloua gracilis (Kunth) Lag. ex Griffiths and Cyperus esculentus L. This vegetation is supported by gentle topography with low amounts of stoniness and a temperate climate (more details in ref. [13]). The size of the site corresponds to 1 hectare and contains mixed individuals of (a) Arbutus bicolor S. González, M. González & P.D. Sørensen; (b) Juniperus deppeana Steud.; (c) Pinus engelmannii Carr.; and (d) Quercus grisea Liebm. (see details of the ecology of these species in Appendix A).

2.2. Transitional Phenophases: Ground and Remote Sensing Detection

The trees were preferably selected under the criteria of good formation with no apparent damage, and the sampling characteristics are documented in Table 1. For five trees per species, we photographed the same sun-exposed branch on the same individual trees monthly with an Apple iPhone 11 with a resolution of 1792 × 828 pixels at 326 ppi and a 12 MP camera (Apple Inc., Cupertino, CA, USA). According to [14], these devices produce valuable data for investigating key phenological events. The photographs were taken under the same configuration, always seeking the same azimuthal orientation, exposure and time of day to ensure as consistent a color balance as possible [15] (Figure 1c).
The phenological study period spanned from November 2023 to October 2024, aiming to encompass the complete annual phenological cycle. A Davis Vantage Vue wireless weather station (model 6252M, Davis Instruments Corp., Hayward, CA, USA) was calibrated and positioned 400 m from the central point of the study site. The recorded variables included maximum temperature (TMAX, °C), minimum temperature (TMIN, °C), and precipitation (P, mm). The device provides real-time data updates every 2.5 s, with an accuracy of ±0.5 °C for temperature and ±4% for precipitation, covering a range of −40 °C to 65 °C and up to 6553 mm of rainfall.
To verify and ensure the consistency of the instrumental climate data, historical daily climatic records from the NASA Prediction of Worldwide Energy Resources (POWER, version 2.4.14, at a 0.5 × 0.5-degree grid) [16] were incorporated. Photoperiod (PP, hours) data were obtained from NOAA [17].

2.3. Environmental and Phenology Relationships

On the same day the tree structures were photographed, we conducted a UAV flight mission at around 13:00 h with a clear sky using a DJI Phantom 4 multispectral drone (P4M) (Dajiang Innovation Technology Co., Shenzhen, China) (Figure 1c). These cameras covered the blue (450 nm ± 16 nm), green (560 nm ± 16 nm), red (650 nm ± 16 nm), red edge (730 nm ± 16 nm) and near-infrared (840 nm ± 26 nm) wavelength bands. The quadcopter was also equipped with a 2 MP global shutter [18]. The UAV was programmed to fly at a height of 60 m above ground level with a frontal and lateral overlap of 80%. Given the challenge of separating the reflectance of foliage and canopy to avoid confusion [19], we manually digitized the tree crowns. We calculated the NDVI at the center of each tree crown for the study period using the following expression:
N D V I = N I R − R N I R + R
where NDVI is the normalized difference vegetation index, and NIR and R are near-infrared and red bands (840 ± 26 nm and 650 ± 16 nm, respectively). To reduce reflectance errors, we standardized NDVI values at the same time (13:32 h) to enhance consistency [20]. The P4M is equipped with a sunlight sensor that enhances radiometric accuracy and consistency. When used in conjunction with the ‘radiometric-calibration’ tool in the OpenDroneMap software (version 2.8.4; Cleveland Metroparks, Cleveland, OH, USA) [21], it efficiently standardizes NDVI values.
To find the NDVI dynamics per species, we initially performed an analysis of variance (ANOVA) between the temporal series of NDVI values of the studied species to determine whether their differences were significant using the stats package of the R studio software, version 4.4.2 (R Core Team [22] Vienna, Austria; Supplementary Materials S1). A Tukey means comparison analysis for monthly NDVI values by species was also performed using the rstats package of the aforementioned software.
Subsequently, based on the maximum and minimum values of the temporal NDVI dynamics, inflection points were established, resulting in seasonal oscillatory patterns associated with the vegetation dynamics detected by the UAV. Once the inflection points were verified with field records, we interpreted them as spectral phenological phases. For these phases, we analyzed the correlations of each NDVI phenophase with its monthly values of maximum temperature (TMAX, °C), photoperiod (PP, sum of hours), average vapor pressure deficit (VPD, kPa) and accumulated precipitation (P, mm) using Spearman’s coefficient [23]. Similarly, from scatter plots between the NDVI and climate variables, simple linear relationships were explored to explain the temporal variability of the NDVI in phenophases.

3. Results

3.1. Transitional Phenophases: Ground and Remote Sensing Detection

According to the visual detection of phenophases on the ground, we found a divergent temporal pattern between conifers and broadleaves (Figure 2). In conifers, the primordia began their development in spring, maintaining their evergreen foliage, while broadleaves kept their foliage green during winter, followed by leaf dehiscence in spring, with buds appearing in summer.
In the conifers, P. engelmannii was notable for evidencing a bud dormant state in winter (November to January), followed by bud swelling in early spring (February to April). The complete appearance of reproductive buds occurred at the end of spring (May and June), while pollen traces and the appearance of new needles were evident at the beginning of summer, with accelerated growth continuing during this season (August). A pronounced differentiation and growth of new twigs followed during autumn (October), entering a state of maturity in November and beginning dormancy with the onset of winter (See Figure 2). In J. deppeana, only expansions in the leaf structure were observed, and changes in the organs were not visible (Figure 2).
For the broadleaf species, Q. grisea showed fruits in autumn (November) that matured in winter (December). At the beginning of spring (February to March), the leaves began to change color, withering and shedding in May–June in Q. grisea and June–July in A. bicolor. Leaf primordia were visible during summer, with Q. grisea presenting early in July and A. bicolor in August. In both cases, rapid leaf expansion occurred, followed by leaf hardening as a sign of maturation until autumn (October) (See Figure 2).
Analysis of the dynamics detected through the NDVI from the UAV confirmed two differential seasonal oscillatory patterns in terms of magnitude and modality between the conifers and broadleaves, consistent with the ANOVA values (see Appendix B). It was confirmed with the variance analysis that the differences between NDVI values over time were only statistically significant between conifers and broadleaves, but not within these two groups (Appendix B, Figure A1, Table A1). The temporal coincidence in inflection points was useful to differentiate three clearly visible periods that we interpreted as spectral phenophases. The first was from 15 November 2023 to 15 February 2024 (late autumn-winter), in which the NDVI in the canopy seems to decrease in angiosperms and gymnosperms. The second was from 15 February 2024 to 14 June 2024 (late winter-spring), with an evident increase in NDVI values, more pronounced in broadleaves. The third phenophase was from 14 June 2024 to 14 October 2024 (summer-early autumn), showing a subsequent increase in the NDVI preceded by the lowest values recorded in the absence of foliage, mainly accentuated in the broadleaves (Figure 3).

3.2. Environmental and Phenology Relationships

The statistical associations between monthly NDVI and daily environmental data reveal their close temporal relationship, although it was seasonally asynchronous between gymnosperms and angiosperms. Mainly, the modality in the second phenophase was more conspicuous in broadleaves than in conifers, with maximum values in spring (March) and minimum values in summer (June) during the driest season. In contrast, climate data showed their unimodal trends (see Figure 1d), characterized by a well-defined seasonality between dry and wet seasons, as documented in Silva-Avila et al. [13].
The phenophases detected through the UAV NDVI are predominantly associated with light-hydro-thermal variables and less so with precipitation, at least during the analyzed time window.
In conifers, all climate variables except P showed a significant association with the NDVI throughout the studied period. A high positive correlation of PP (r = 0.8, p < 0.001) was notable during summer–autumn (August–September), the period of elongation and development of foliar primordia, which contrasted with a negative correlation in autumn–winter (r = −0.65, p < 0.001), when the trees entered dormancy and the foliage degraded. TMAX (r = 0.78, p < 0.001) and VPD (r = 0.75, p < 0.001) were also positively associated with summer NDVI results in the rainy season during foliage elongation and formation.
For broadleaves, all climate variables maintained a close relationship with the NDVI values of phenophases. PP was notable with a negative effect from autumn in Arbutus (r = −0.88, p < 0.001), intensifying in spring when the leaves finished maturing and began to degrade (Quercus, r = −0.91, p < 0.001). TMAX was mainly associated negatively with the NDVI of Quercus (r = −0.85, p < 0.001), followed by VPD (r = −0.82, p < 0.001), which corresponded with foliage decay during spring. P was positively associated during summer in the rainy season when the foliar structures developed (r = 0.54, p < 0.001).
Regarding the explanatory linear models of the phenophases, the predictive capacity of PP, TMAX and VPD variables was notable, confirming the role of the obtained correlations. Their covariation could be positive or negative, depending on phenophase and species (Table 2).

4. Discussion

4.1. Transitional Phenophases: Ground and Remote Sensing Detection

Approaching multi-species phenology through direct observations validated with remote sensing values and explained by climate produced an element with which to validate our hypothesis. Our study presented similar conditions in terms of photoperiod, water and nutrient availability, and pedological factors within the microsite for all the studied species. Moreover, we studied mature trees in a natural environment, unlike in traditional manipulative experiments, which take place in controlled environments with saplings [24,25] and where underestimations have been observed compared to the results of longer-term studies [26]. Our results allowed for the differentiation of greater phenological details in the search to optimize phenological monitoring performance [4]. Most models are based on remote sensor observations that still show deficiencies in resolution [5,7] and are limited in studies such as ours that include biological data taken directly in the field and linked with local climatic data. It has been reported that, while boreal forest species are showing increases in their growth, tropical forests present declining growth [4]. An intermediate forest such as the one studied here can improve our understanding of the relationships between phenology and environmental variables.
Since the basic unit was the tree, and this has largely been the case since the beginning of phenology [27], these results could strengthen the argument to scale local results to ecosystem levels [28], since the mix of species studied here contributed knowledge that is traditionally difficult to obtain in heterogeneous forests [29], in contrast to evergreen forests, which tend to produce spectral uniformity. Although the results correspond to a local area, it is plausible that they can have an impact on the biogeography of other ecosystems, given the relationship of climate to phenology across bioclimatic gradients [30].
It is recognized that structural heterogeneity plays an important role in monitoring phenological variation [31]. However, in situ observation proved fundamental and demonstrates that it cannot be dispensed with, although improvements to photographic monitoring protocols are still required. The higher the temporal, spatial and spectral resolution, the better the understanding of phenology. Above all, the analysis of organs expanded our knowledge relative to those that habitually only consider foliage analysis [10], or those observing only two seasons: spring (bud break and leaf elongation) and autumn (coloration and leaf fall) [32].
The combination with the use of UAV technology proved to be a useful alternative for monitoring phenology, the evidence of which is supported by Figure 3, in line with findings in recent studies [9]. The opportunity it provides users to obtain data at the desired temporal and spatial detail level is a great advantage over traditional satellite images in which these scales cannot be controlled [6], and it constitutes an intermediate tool between ground observation and satellite imagery. According to Polgar and Primack [33], satellite imagery has traditionally been used to monitor phenology. However, individual tree detection significantly enhances our ability to characterize crown-level phenology in both spatial and temporal dimensions. This ‘near-surface’ remote sensing approach offers improved spectral and spatial resolution compared to conventional satellite-based methods. In summary, UAVs represent a promising strategy for disentangling the effects of changing species composition and monitoring climate impacts in mixed stands spanning only a few hectares [8,34].
The identification of temporal peaks or inflection points coincided with findings for ponderosa pine communities [35]. The determination of phenometrics with the NDVI identified the magnitude and temporality that corresponded with phenological seasons. Unlike these, our results allowed a finer scale of analysis in the presence of heterogeneous forests. Validation of the NDVI with field data allowed an explicit connection to be made between remote sensing and phenology [35]. These patterns are consistent with those reported in previous studies [9,36].
As expected, the species of the mixed forest presented different phenophases and temporality in their appearance, duration and termination. Predominantly, differentiation between conifers and broadleaves depends on their seasonality and on eco-physiological mechanisms that remain unclear [37]. Although the preformation and neoformation of organs cannot be determined further with the current temporal analysis window, it is evident that in the case of conifers, the buds are formed from the previous year, and their elongation occurs in spring–summer (Figure 2), regulated by a polycyclic and seasonally staggered behavior [38]. In contrast, the phenophases of angiosperms showed organ development during the same year as their initial appearance, presumably attributed to their hydraulic functioning as an anisohydric rather than isohydric species. That is, the former have an advantageous strategy over conifers in terms of photosynthesizing structural carbohydrates when their foliage is present and, at the same time, saving respiration costs when they lose that foliage in the dry season. In a disadvantageous behavior, conifers increase their competition for assimilates with repercussions of carbohydrate deficit, with an emphasis on bud and brachyblast development. According to Bosio et al. [11], freshly formed leaves are photosynthetically active, but their respiration needs exceed photosynthate production, and the tree must usually balance this deficit from vascular cambium reserves, with a consequent effect on radial growth. Interconnections were evident in a study of Pinus longaeva [39]: a slight decrease in the bud length rate coincided with cambial activation, while a decrease in needle growth rate occurred simultaneously with the end of pollination. The maximum needle length occurred with the cessation of cambial growth, and bud development corresponded with a slight temporary increase in stem width. These mechanisms seem to configure the differences in the timing of phenophases between conifers and broadleaves, which aligns with [40].

4.2. Environmental and Phenological Relationships

The influence of environmental variables on NDVI results, considered here as a proxy with which to identify phenophases, has provided statistical arguments to confirm our hypothesis. As expected, seasonal climatic conditions regulate the spectral phenophases of NDVI, reducing uncertainty in quantifying their seasonal role in phenophases, in agreement with Hover et al. [38]. We have shown that phenology is seasonally constrained by photo-hydro-thermal conditions in a contrasting manner between conifers and broadleaves.
Conifers are modulated by PP in an oscillatory manner in two directions. In the summer, it is associated with the vegetative development of new foliar structures and reproductive organs while, in the winter, it affects the aerial structure causing leaf fall. These phenological phases, represented by marked inflection points in the phenological processes of temperate forests, have been reported to be dependent on photoperiod and temperature [41]. A longer P, in combination with an increase in TMAX, promotes bud elongation as seen in P. tabuliformis [42], the photosynthetic activation of which is reflected in an increase in the NDVI during early spring, but is most evident during the summer, along with high VPD values, when rains occur (Figure 3). In Pinus longaeva, bud development is associated with radial growth, but it does not correspond to a precipitation event but rather to hydration following the season of dormancy [39]. Bud opening and pollination in this species have been related to accumulated heat, but they do not affect cambial growth in the same way. In other words, the increase in temperatures is first manifested in phenological structures before radial growth.
Indeed, research on Pinus tabuliformis indicates that an extension of the growing season does not translate into greater radial growth [42], a parameter that is dependent to a greater extent on moisture availability.
Similarly, during winter, physiological mechanisms in phytohormones, phytochromes and carbohydrate metabolism are activated under the influence of longer PP and warmer conditions [41]. It seems that warm conditions can extend the duration of the growing season but not necessarily its productivity (see Man et al. [42]), since the dynamics depend on winter moisture conditions [43]. Thus, water reserves created during the previous winter play a crucial role in maintaining the optimal metabolic functioning of the transition from endo- to eco-dormancy in these conifers [41].
The species Q. grisea and A. bicolor are climatically regulated by a negative influence of photo-thermal conditions and the VPD as precursors of leaf fall due to the high evaporation rates they experience during the driest months. Indeed, Quercus has been very responsive to temperatures [41], putting into perspective the shortening of winter, or occurrence of an early spring linked with a late autumn, which can pose risks to the plant due to lack of soil moisture. Furthermore, an extension of the photoperiod can have an adverse effect in areas prone to drought, such as the study site, as low water availability can lead to stress and hydraulic failure, with consequent changes in phenology.
Interestingly, the emergence of leaf primordia produced before the rains begin indicates that precipitation during that year is not decisive for bud opening, and their appearance is potentially associated with the exploitation of previously generated reserves of water (see Acosta-Hernández et al. [44]). However, we lack further data to test the magnitude of the retrospective seasonal “memory” that these species have for using their resources, so a further and more profound analysis of the temporal window is necessary. When the rains begin, the relative humidity in the environment favors photosynthetic activity, thus reaching the maximum NDVI (Figure 3) attributed to the evident accelerated elongation of shoots and buds. Monsoonal rains could consequently alleviate drought stress, but the lag between water shortage and NDVI values still makes it challenging to test this hypothesis, with an emphasis on the wide variety of species present.

4.3. Final Remarks

The asynchronous oscillation in the influence of environmental variables on the studied species is very important. In the rainy summer, the NDVI of conifers is positively associated with the combination of PP, VPD and TMAX: however, the effect is negative on the foliage activity of broadleaves. The release of winter dormancy was mediated by intraspecific responses to the covariation of PP, temperature and VPD. The direct relationship of PP with the emergence of primordia is explained because light perception by phytochrome provides a temporal mechanism that ensures bud opening at the required time [11]. Similarly, there are indications that the maximum growth rates of vegetative organs are controlled by an adequate balance of PP, temperature, VPD and precipitation [41].
The development of phenophases therefore seems to depend on the covariation of these variables, and the complexity is accentuated by the heterogeneity of the species composition.
The onset and duration of phenological phases thus have implications for the carbohydrate reserves the tree will subsequently utilize. For example, if summers and winters are exacerbated as warm and dry, the tree phenology may be compromised due to the need for water and structural carbohydrates for subsequent phenophases. Some implications include changes in the timing of phenophases and potentially lead to decline, loss of productivity, and even mortality [45,46], as well as ecological mismatches with the insects and animals that depend on the phenological changes (e.g., for pollination, fruiting).
Our results concur with those of other studies showing the dependence on variables such as TMAX, VPD and PP, but which have mainly been conducted using larger-scale satellite and phenocamera technologies [7]. In contrast, our study explicitly represents responses at a micro-scale that is usually ignored, and the findings could contribute to adjusting larger-scale models.
Seasonal environmental conditions therefore have implications for the acclimatization of species to climate change. Above all, the high susceptibility to drought episodes in the study area, as shown recently (see [13,44]), and particularly those preceding summer, could be the response to polycyclism in conifers as an adaptation to drought. In our case, phenophases in Pinus are dissociated mainly due to environmental factors. This behavior makes it possible to regulate growth according to the climatic conditions of the present and past year. That is, during an adverse year, limitations in resources for bud formation or structures can be expected in the following year, and vice versa.
Finally, explanatory models of phenophases confirm the linear predictive capacity of climatic variables, with emphasis on the photoperiod effect for broadleaves during spring, but positive for conifers in summer. However, these results should be treated with some caution, as biological non-linearity cannot be considered constant.

4.4. Limitations

Although our experiment constitutes a natural laboratory, the findings cannot be considered universally conclusive since we only explored one year of variability, which is inadequate to explain long-term plant responses to environmental changes. As is well known, this area is experiencing increasing aridification trends [44]; however, there is a lack of temporal data to assess the specific role of drought episodes, including long-term NDVI series. While the characterization of fine-scale phenological variation was acceptable, the temporal resolution of the sampling still merits adjustment to enable detection of processes shorter than a month [5]. Moreover, it is essential to consider the link with explicit in situ photosynthesis data, ecophysiology, hydraulic conductivity and wood formation, among other modeling techniques [47], utilizing, for example, phenocameras and LiDAR technology [48], since we cannot yet classify climate vulnerability interspecifically or inter-seasonally. This could make a valuable contribution. According to Hover et al. [38], seasonal analysis for more than one year is very necessary to understand the plasticity of species in response to hydroclimatic variations.
Furthermore, multiple factors shape tree responses, including thermal acclimatization, water and nutrient availability, as well as ontogenetic aspects that remain to be investigated [49]. Above all, climate data for more than one year preceding major phenological phases are required because the temporal window of analysis may not be adequate, leading to discrepancies [42]. The connections between phenological stages also deserve attention (see Fu et al. [50]), since mechanisms of “memory” are reported in trees as strategies for dormancy cycles, endodormancy, ecodormancy and paradormancy (see Fadón et al. [51]). It is advisable to include physiological and morphological aspects to complement the representation of underlying mechanisms in phenology. The study of the role of phytohormones also merits further investigation, especially in light of the advancement in the onset of phenological states, which even vary among dimensions and ages of trees, and ontogeny [41]. Similarly, tree social status is a factor to consider in phenology monitoring, especially in suppressed trees.
Although the studied trees correspond to a variable structure, it is recognized that structural heterogeneity plays an important role, as does social status, in the distribution of the photoperiod and the microclimate of the stand [52]. Trees respond based on size and social status (especially in suppressed trees) [53]. In our case, it would be advisable to model by age and social position within the stand. It is also recommended to study the ideal sample size and analyze root phenological relationships [54], given their significant role in tree response mechanisms.

5. Conclusions

The monthly field observations proved to be an appropriate link for monitoring phenophases, which showed divergences of phenological patterns between conifers and broadleaves. In the case of the former, the buds were formed from the previous year, and their elongation occurred in spring–summer. In contrast, the phenophases of the latter showed organ development in the same year as their initial appearance, but the underlying mechanisms remain unclear. The conifers exhibited development of primordia and growth beginning in early spring, while the broadleaves began in the summer. However, it is advisable to expand monitoring of the roots, which remain largely unexplored.
Validation of in situ data with those derived from the UAV demonstrated a biological correlation between canopy dynamics and the NDVI as a proxy for detecting phenophases in individual trees. Thus, the UAV constitutes a valuable tool for improved monitoring of vegetation dynamics.
Phenophases are controlled by the magnitude of the photoperiod and concurrent seasonal thermal variables but are temporally differentiated between the conifers and broadleaves. Variables related to drought have a negative impact on foliage dynamics in all species, although monsoonal rains benefit broadleaf species almost immediately.
Although this study contributes to our understanding of phenology in mixed forests, it is advisable to reduce the temporal scale of monitoring to less than one month to perceive the phenological dynamics in greater detail, including their association with morphological, biometric, ecophysiological, phytohormonal, xylogenesis and ontogenetic data. It is also advisable to conduct long-term monitoring of remote sensing data.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/f16071118/s1, Supplementary Material S1. R script used in the development of the manuscript.

Author Contributions

Conceptualization, M.P.-G. and S.R.; methodology, M.P.-G.; formal analysis, E.D.V.-V. and A.C.A.-H.; investigation, M.P.-G.; data curation, M.P.-G., E.D.V.-V. and A.C.A.-H.; writing—original draft preparation, M.P.-G., E.D.V.-V., A.C.A.-H. and S.R.; project administration, M.P.-G. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Appendix A.1. Ecology of the Studied Species

Appendix A.1.1. Arbutus bicolor S. González, M. González & P.D. Sørensen

Arbutus bicolor develops in temperate forests associated with conifers, pines and oaks, at between 1000 and 2400 m above sea level. This species prefers well-drained, slightly acidic soils, with temperate and humid climates. It has morphological adaptations (thick, leathery leaves) that help reduce transpiration and conserve water under seasonal drought conditions. The clustering of its flowers favors insect pollination. The wood and branches of the plant provide shelter and food for some animals. The species regenerates through seeds or vegetative propagation and, once established, is resistant to environmental changes [55].

Appendix A.1.2. Juniperus deppeana Steud.

Juniperus deppeana develops in temperate forests and shrublands at elevations of 1500 to 3000 m above sea level in mountainous regions of Mexico. The species prefers well-drained, often rocky or calcareous soils, and adapts to semi-arid or arid climates. It has morphological adaptations, such as scaly leaves that minimize water loss, allowing it to withstand prolonged droughts. The flowers are small and produce cones that are dispersed by birds and mammals. Its wood is used by wildlife for nesting. Regeneration occurs mainly through seeds, and once established, the species is resistant to water stress and extreme temperatures [56].

Appendix A.1.3. Pinus engelmannii Carr.

Pinus engelmannii is found in temperate mountain forests, associated with other species of pines, oaks and firs, at elevations of 2000 to 3300 m above sea level in the northern mountainous regions of Mexico. The species prefers well-drained and slightly acidic soils in areas with cool and humid climates, with temperatures between 10 and 20 °C. It has morphological adaptations, such as needle-like leaves and thick bark, which allow it to endure frost and water loss in dry conditions. Its cones open when mature, releasing seeds that are mainly dispersed by the wind. Regeneration occurs through seeds and sometimes by shoots from the trunk base. Once established, it is resistant to fires and moderate droughts [57,58].
Appendix A.1.4. Quercus grisea Liebm.
Quercus grisea is found in temperate forests, associated with other species of oaks and pines, at elevations of 1800 to 3000 m above sea level in the mountains of Mexico. The species prefers well-drained, often calcareous or rocky soils, and adapts to cool and semi-arid climates. It has morphological adaptations, such as leathery leaves that reduce water loss and withstand seasonal drought. Its flowers are small, and its fruits, acorns, are mainly dispersed by mammals and birds. The species regenerates through seeds and can withstand conditions of water stress and low temperatures, being resistant to fires and moderate climate changes [59].

Appendix B

Table A1. Results of ANOVA for fixed effects on NDVI values from mixed-effects models adjusted by tree genus.
Table A1. Results of ANOVA for fixed effects on NDVI values from mixed-effects models adjusted by tree genus.
EffectFDfDf.resPr(>F)Signif.Codes
Genus35.3873520<2.2 × 10−16***
Date481.4514422717<2.2× 10−16***
Interaction Genus × Date155.50613222717<2.2× 10−16***
Where: F = F value, D.f. = degrees of freedom, Df.res = residual degrees of freedom, Pr(>F) = probability value of finding a value greater than F, Signif.Codes = significance codes: *** p < 0.001
The analysis shows that both genus and date significantly influence the NDVI. Furthermore, the interaction between both factors indicates that each genus responds differently depending on the date of measurement.
Table A2. Paired NDVI comparisons between genera with Bonferroni adjustment.
Table A2. Paired NDVI comparisons between genera with Bonferroni adjustment.
Comparisonn1n2Statisticdfpp.adjSignificance
Arbutus–Juniperus54023103.226941.00× 10−038.00 × 10−03**
Arbutus–Pinus54017,439−7.355517.35 × 10−134.41 × 10−12***
Arbutus–Quercus5403128−3.998167.29 × 10−054.37 × 10−04***
Juniperus–Pinus231017,439−28.126987.96 × 10−1534.78 × 10−152***
Juniperus–Quercus23103128−12.953631.95 × 10−371.17 × 10−36***
Pinus–Quercus17,43931285.9434283.10 × 10−091.86 × 10−08***
Asterisks denote significance levels as follows: p < 0.01, p < 0.001. Where: Significance = ** p < 0.01, *** p < 0.001.
Forests 16 01118 g0a1
Figure A1. Boxplot of NDVI of genus. Significance codes = ** p < 0.01, *** p < 0.001.
Figure A1. Boxplot of NDVI of genus. Significance codes = ** p < 0.01, *** p < 0.001.
Forests 16 01118 g0a1
Table A3. Results of fixed effects ANOVA on the NDVI values from mixed effects models adjusted by tree class.
Table A3. Results of fixed effects ANOVA on the NDVI values from mixed effects models adjusted by tree class.
EffectFDfDf.resPr(>F)Signif.Codes
Class (Conifer or Broadleaf)5.541215220.01894*
Date458.04854422,805<2.2 × 10−16***
Interaction Class × Date416.6864422,805<2.2 × 10−16***
Where: F value = F value, D.f. = degrees of freedom, Df.res = residual degrees of freedom, Pr(>F) = probability value of finding a value greater than F, Signif.Codes = significance codes: * p < 0.05, *** p < 0.001.
Forests 16 01118 g0a2
Figure A2. Boxplot of NDVI of conifers and broadleafs. Significance code = * p < 0.05.
Figure A2. Boxplot of NDVI of conifers and broadleafs. Significance code = * p < 0.05.
Forests 16 01118 g0a2

References

  1. Ault, T.R.; Schwartz, M.D.; Zurita-Milla, R.; Weltzin, J.F.; Betancourt, J.L. Trends and natural variability of spring onset in the coterminous United States as evaluated by a new gridded dataset of spring indices. J. Clim. 2015, 28, 8363–8378. [Google Scholar] [CrossRef]
  2. Keenan, T.F.; Gray, J.; Friedl, M.A.; Toomey, M.; Bohrer, G.; Hollinger, D.Y.; Munger, J.W.; O’Keefe, J.; Schmid, H.P.; Wing, I.S.; et al. Net carbon uptake has increased through warming-induced changes in temperate forest phenology. Nat. Clim. Change 2014, 4, 598–604. [Google Scholar] [CrossRef]
  3. Piao, S.; Liu, Z.; Wang, T.; Peng, S.; Ciais, P.; Huang, M.; Ahlstrom, A.; Burkhart, J.F.; Chevallier, F.; Janssens, I.A.; et al. Weakening temperature control on the interannual variations of spring carbon uptake across northern lands. Nat. Clim. Change 2017, 7, 359–363. [Google Scholar] [CrossRef]
  4. Piao, S.; Liu, Q.; Chen, A.; Janssens, I.A.; Fu, Y.; Dai, J.; Liu, L.; Lian, X.; Shen, M.; Zhu, X. Plant phenology and global climate change: Current progresses and challenges. Glob. Change Biol. 2019, 25, 1922–1940. [Google Scholar] [CrossRef] [PubMed]
  5. Klosterman, S.; Melaas, E.; Wang, J.A.; Martinez, A.; Frederick, S.; O’Keefe, J.; Orwig, D.A.; Wang, Z.; Sun, Q.; Schaaf, C.; et al. Fine-scale perspectives on landscape phenology from unmanned aerial vehicle (UAV) photography. Agric. For. Meteorol. 2018, 248, 397–407. [Google Scholar] [CrossRef]
  6. Park, J.Y.; Muller-Landau, H.C.; Lichstein, J.W.; Rifai, S.W.; Dandois, J.P.; Bohlman, S.A. Quantifying leaf phenology of individual trees and species in a tropical forest using Unmanned Aerial Vehicle (UAV) images. Remote Sens. 2019, 11, 1534. [Google Scholar] [CrossRef]
  7. Thapa, S.; Millan, V.E.G.; Eklundh, L. Assessing Forest Phenology: A multi-scale comparison of near-surface (UAV, spectral reflectance sensor, PhenoCam) and satellite (MODIS, Sentinel-2) remote sensing. Remote Sens. 2021, 13, 1597. [Google Scholar] [CrossRef]
  8. Fawcett, D.; Bennie, J.; Anderson, K. Monitoring spring phenology of individual tree crowns using drone-acquired NDVI data. Remote Sens. Ecol. Conserv. 2020, 7, 227–244. [Google Scholar] [CrossRef]
  9. Berra, E.F.; Gaulton, R.; Barr, S. Assessing spring phenology of a temperate woodland: A multiscale comparison of ground, unmanned aerial vehicle and Landsat satellite observations. Remote Sens. Environ. 2019, 223, 229–242. [Google Scholar] [CrossRef]
  10. Delpierre, N.; Vitasse, Y.; Chuine, I.; Guillemot, J.; Bazot, S.; Rutishauser, T.; Rathgeber, C.B.K. Temperate and boreal forest tree phenology: From organ-scale processes to terrestrial ecosystem models. Ann. For. Sci. 2016, 73, 5–25. [Google Scholar] [CrossRef]
  11. Bosio, F.; Rossi, S.; Marcati, C.R. Periodicity and environmental drivers of apical and lateral growth in a Cerrado woody species. Trees 2016, 30, 1495–1505. [Google Scholar] [CrossRef]
  12. Seager, R.; Ting, M.; Davis, M.; Cane, M.; Naik, N.; Nakamura, J.; Li, C.; Cook, E.; Stahle, D.W. Mexican drought: An observational modeling and tree ring study of variability and climate change. Atmósfera 2009, 22, 1–31. [Google Scholar]
  13. Silva-Ávila, N.; Pompa-García, M.; Martínez-Rivas, J.A. Kawí Tamiruyé: Un acceso universal al conocimiento del bosque. Una revisión documental. Nat. Soc. Desafíos Medioambient. 2025, 11, 60–75. [Google Scholar] [CrossRef]
  14. Klinger, Y.P.; Eckstein, R.L.; Kleinebecker, T. iPhenology: Using open-access citizen science photos to track phenology at continental scale. Methods Ecol. Evol. 2023, 14, 1424–1431. [Google Scholar] [CrossRef]
  15. Richardson, A.D.; Braswell, B.H.; Hollinger, D.Y.; Jenkins, J.P.; Ollinger, S.V. Near-surface remote sensing of spatial and temporal variation in canopy phenology. Ecol. Appl. 2009, 19, 1417–1428. [Google Scholar] [CrossRef]
  16. NASA POWER Data Access Viewer. Available online: https://power.larc.nasa.gov/data-access-viewer/ (accessed on 4 April 2025).
  17. NOAA Solar Calculator. Available online: https://gml.noaa.gov/grad/solcalc/ (accessed on 4 April 2025).
  18. Support for P4 Multispectral. Available online: https://www.dji.com/mx/support/product/p4-multispectral (accessed on 4 June 2025).
  19. Lin, C.; Chen, S.-Y.; Chen, C.-C.; Tai, C.-H. Detecting newly grown tree leaves from unmanned-aerial-vehicle images using hyperspectral target detection techniques. ISPRS J. Photogramm. Remote Sens. 2018, 142, 174–189. [Google Scholar] [CrossRef]
  20. Li, D.; Chen, J.M.; Zhang, X.; Yan, Y.; Zhu, J.; Zheng, H.; Zhou, K.; Yao, X.; Tian, Y.; Zhu, Y.; et al. Improved estimation of leaf chlorophyll content of row crops from canopy reflectance spectra through minimizing canopy structural effects and optimizing off-noon observation time. Remote Sens. Environ. 2020, 248, 111985. [Google Scholar] [CrossRef]
  21. OpenDroneMap. A Command Line Toolkit to Generate Maps, Point Clouds, 3D Models and Dems from Drone, Balloon or Kite Images. Available online: https://github.com/OpenDroneMap/ODM (accessed on 20 February 2025).
  22. R Core Team. R: A Language and Environment for Statistical Computing; The R Project for Statistical Computing: Vienna, Austria, 2025; Available online: https://www.R-project.org/ (accessed on 4 June 2025).
  23. Myers, L.; Sirois, M.J. Spearman Correlation Coefficients, Differences between. Wiley StatsRef Stat. Ref. Online, 2014; epub ahead of print. [Google Scholar] [CrossRef]
  24. Primack, R.B.; Laube, J.; Gallinat, A.S.; Menzel, A. From observations to experiments in phenology research: Investigating climate change impacts on trees and shrubs using dormant twigs. Ann. Bot. 2015, 116, 889–897. [Google Scholar] [CrossRef]
  25. Beaton, J.; Perry, A.; Cottrell, J.; Iason, G.; Stockan, J.; Cavers, S. Phenotypic trait variation in a long-term multisite common garden experiment of Scots pine in Scotland. Sci. Data 2022, 9, 671. [Google Scholar] [CrossRef]
  26. Wolkovich, E.M.; Cook, B.I.; Allen, J.M.; Crimmins, T.M.; Betancourt, J.L.; Travers, S.E.; Pau, S.; Regetz, J.; Davies, T.J.; Kraft, N.J.B.; et al. Warming experiments underpredict plant phenological responses to climate change. Nature 2012, 485, 494–497. [Google Scholar] [CrossRef]
  27. Delpierre, N.; Guillemot, J.; Dufrêne, E.; Cecchini, S.; Nicolas, M. Tree phenological ranks repeat from year to year and correlate with growth in temperate deciduous forests. Agric. For. Meteorol. 2017, 234–235, 1–10. [Google Scholar] [CrossRef]
  28. Stoy, P.C.; Williams, M.; Disney, M.; Prieto-Blanco, A.; Huntley, B.; Baxter, R.; Lewis, P. Upscaling as ecological information transfer: A simple framework with application to Arctic ecosystem carbon exchange. Landsc. Ecol. 2009, 24, 971–986. [Google Scholar] [CrossRef]
  29. Klosterman, S.T.; Hufkens, K.; Gray, J.M.; Melaas, E.; Sonnentag, O.; Lavine, I.; Mitchell, L.; Norman, R.; Friedl, M.A.; Richardson, A.D. Evaluating remote sensing of deciduous forest phenology at multiple spatial scales using PhenoCam imagery. Biogeosciences 2014, 11, 4305–4320. [Google Scholar] [CrossRef]
  30. Vitasse, Y.; Signarbieux, C.; Fu, Y.H. Global warming leads to more uniform spring phenology across elevations. Proc. Natl. Acad. Sci. USA 2018, 115, 1004–1008. [Google Scholar] [CrossRef]
  31. Jiang, D.; Xu, Z.; Nie, T. Advancements in monitoring tree phenology under global change: A comprehensive review. Forests 2025, 16, 771. [Google Scholar] [CrossRef]
  32. Richardson, A.D.; Black, T.A.; Ciais, P.; Delbart, N.; Friedl, M.A.; Gobron, N.; Hollinger, D.Y.; Kutsch, W.L.; Longdoz, B.; Luyssaert, S.; et al. Influence of spring and autumn phenological transitions on forest ecosystem productivity. Phil. Trans. R. Soc. B 2010, 365, 3227–3246. [Google Scholar] [CrossRef]
  33. Polgar, C.A.; Primack, R.B. Leaf-out phenology of temperate woody plants: From trees to ecosystems. New Phytol. 2011, 191, 926–941. [Google Scholar] [CrossRef]
  34. Vitasse, Y.; Bresson, C.C.; Kremer, A.; Michalet, R.; Delzon, S. Quantifying phenological plasticity to temperature in two temperate tree species. Funct. Ecol. 2010, 24, 1211–1218. [Google Scholar] [CrossRef]
  35. Walker, J.J.; Soulard, C.E. Phenology patterns indicate recovery trajectories of Ponderosa pine forests after high-severity fires. Remote Sens. 2019, 11, 2782. [Google Scholar] [CrossRef]
  36. White, K.; Pontius, J.; Schaberg, P. Remote sensing of spring phenology in northeastern forests: A comparison of methods, field metrics and sources of uncertainty. Remote Sens. Environ. 2014, 148, 97–107. [Google Scholar] [CrossRef]
  37. Ishii, H.; Asano, S. The role of crown architecture, leaf phenology and photosynthetic activity in promoting complementary use of light among coexisting species in temperate forests. Ecol. Res. 2010, 25, 715–722. [Google Scholar] [CrossRef]
  38. Hover, A.; Buissart, F.; Caraglio, Y.; Heinz, C.; Pailler, F.; Ramel, M.; Vennetier, M.; Prévosto, B.; Sabatier, S. Growth phenology in Pinus halepensis Mill.: Apical shoot bud content and shoot elongation. Ann. For. Sci. 2017, 74, 39. [Google Scholar] [CrossRef]
  39. Hallman, C.; Arnott, H. Morphological and physiological phenology of Pinus longaeva in the White Mountains of California. Tree-Ring Res. 2015, 71, 1–12. [Google Scholar] [CrossRef]
  40. Burdon, R.D. Shoot phenology as a driver or modulator of stem diameter growth and wood properties, with special reference to Pinus radiata. Forests 2023, 14, 570. [Google Scholar] [CrossRef]
  41. Basler, D.; Körner, C. Photoperiod and temperature responses of bud swelling and bud burst in four temperate forest tree species. Tree Physiol. 2014, 34, 377–388. [Google Scholar] [CrossRef]
  42. Man, Z.; Zhang, J.; Liu, J.; Liu, L.; Yang, J.; Cao, Z. Process-based modeling of phenology and radial growth in Pinus tabuliformis in response to climate factors over a cold and semi-arid region. Plants 2024, 13, 980. [Google Scholar] [CrossRef]
  43. Clark, J.S.; Salk, C.; Melillo, J.; Mohan, J. Tree phenology responses to winter chilling, spring warming, at north and south range limits. Funct. Ecol. 2014, 28, 1344–1355. [Google Scholar] [CrossRef]
  44. Acosta-Hernández, A.C.; Pompa-García, M.; Martínez-Rivas, J.A.; Vivar-Vivar, E.D. Cutting the Greenness Index into 12 monthly slices: How intra-annual NDVI dynamics help decipher drought responses in mixed forest tree species. Remote Sens. 2024, 16, 389. [Google Scholar] [CrossRef]
  45. Gunderson, C.A.; Edwards, N.T.; Walker, A.V.; O’Hara, K.H.; Campion, C.M.; Hanson, P.J. Forest phenology and a warmer climate—Growing season extension in relation to climatic provenance. Glob. Change Biol. 2011, 18, 2008–2025. [Google Scholar] [CrossRef]
  46. Luo, Y.; Zohner, C.; Crowther, T.W.; Feng, J.; Hoch, G.; Li, P.; Richardson, A.D.; Vitasse, Y.; Gessler, A. Internal physiological drivers of leaf development in trees: Understanding the relationship between non-structural carbohydrates and leaf phenology. Funct. Ecol. 2024, 1–14, early view. [Google Scholar] [CrossRef]
  47. Zhang, J.; Zhai, J.; Wang, J.; Si, J.; Li, J.; Ge, X.; Li, Z. Interrelationships and environmental influences of photosynthetic capacity and hydraulic conductivity in desert species Populus pruinose. Forests 2024, 15, 1094. [Google Scholar] [CrossRef]
  48. Gray, R.E.J.; Ewers, R.M. Monitoring forest phenology in a changing world. Forests 2021, 12, 297. [Google Scholar] [CrossRef]
  49. Kellomäki, S. Physiology, growth, and acclimatizing of boreal tree species to climate. In Management of Boreal Forests, 1st ed.; Springer: Cham, Switzerland, 2022; pp. 139–216. [Google Scholar] [CrossRef]
  50. Fu, Y.H.; Piao, S.; de Beeck, M.O.; Cong, N.; Zhao, H.; Zhang, Y.; Menzel, A.; Janssens, I.A. Recent spring phenology shifts in western Central Europe based on multiscale observations. Glob. Ecol. Biogeogr. 2014, 23, 1255–1263. [Google Scholar] [CrossRef]
  51. Fadón, E.; Fernandez, E.; Behn, H.; Luedeling, E. A conceptual framework for winter dormancy in deciduous trees. Agronomy 2020, 10, 241. [Google Scholar] [CrossRef]
  52. Wu, X.; Niu, C.; Liu, X.; Hu, T.; Feng, Y.; Zhao, Y.; Liu, S.; Liu, Z.; Dai, G.; Zhang, Y.; et al. Canopy structure regulates autumn phenology by mediating the microclimate in temperate forests. Nat. Clim. Change 2024, 14, 1299–1305. [Google Scholar] [CrossRef]
  53. Vitasse, Y.; Lenz, A.; Hoch, G.; Körner, C. Earlier leaf-out rather than difference in freezing resistance puts juvenile trees at greater risk of damage than adult trees. J. Ecol. 2014, 102, 981–988. [Google Scholar] [CrossRef]
  54. Radville, L.; Post, E.; Eissenstat, D.M. On the sensitivity of root and leaf phenology to warming in the Arctic. Arct. Antarct. Alp. Res. 2018, 50, S100020. [Google Scholar] [CrossRef]
  55. González-Elizondo, M.S.; González-Elizondo, M.; Sørensen, P.D. Arbutus bicolor (Ericaceae, Arbuteae), a new species from Mexico. Act. Bot. Mex. 2012, 99, 55–72. [Google Scholar] [CrossRef]
  56. Herrerías-Mier, L.G.; de Pascual Pola, M.C.C.N. Características estructurales y demográficas de Juniperus deppeana Steud. en dos localidades del estado de Tlaxcala. Rev. Mex. Cienc. For. 2021, 11, 124–151. [Google Scholar] [CrossRef]
  57. Jiménez-Salazar, M.A.; Méndez-González, J. Distribución actual y potencial de Pinus engelmannii Carriére bajo escenarios de cambio climático. Madera Bosques 2021, 27, e2732117. [Google Scholar] [CrossRef]
  58. Pavek, D.S. Pinus engelmannii. Fire Effects Information System [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). 1994. Available online: https://www.fs.usda.gov/database/feis/plants/tree/pineng/all.html (accessed on 9 May 2025).
  59. Pavek, D.S. Quercus grisea. Fire Effects Information System [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). 1994. Available online: https://www.fs.usda.gov/database/feis/plants/tree/quegri/all.html (accessed on 9 May 2025).
Forests 16 01118 g001
Figure 1. (a) Location of the study area (base map from the ‘ESRI Satellite’ service, accessed via the QuickMapServices plugin in QGIS version 3.40.6), (b) view of the ecosystem (photograph taken by the authors on 15 February 2024), (c) schematic representation of the monthly process of capturing phenological ground images and aerial images using UAVs and (d) daily dynamics of climatic variables from November 2023 to November 2024. Months in lowercase denote the year 2023, while those in uppercase pertain to 2024. (P: Precipitation; TMAX: maximum temperature; VPD: vapor pressure deficit; PP: photoperiod; source: created by the authors). The climate data used were collected hourly with a Davis Vantage Vue wireless weather station located in the study area.
Figure 1. (a) Location of the study area (base map from the ‘ESRI Satellite’ service, accessed via the QuickMapServices plugin in QGIS version 3.40.6), (b) view of the ecosystem (photograph taken by the authors on 15 February 2024), (c) schematic representation of the monthly process of capturing phenological ground images and aerial images using UAVs and (d) daily dynamics of climatic variables from November 2023 to November 2024. Months in lowercase denote the year 2023, while those in uppercase pertain to 2024. (P: Precipitation; TMAX: maximum temperature; VPD: vapor pressure deficit; PP: photoperiod; source: created by the authors). The climate data used were collected hourly with a Davis Vantage Vue wireless weather station located in the study area.
Forests 16 01118 g001
Forests 16 01118 g002
Figure 2. Phenology in the field of four species in a mixed forest: (a) Juniperus deppeana; (b) Pinus engelmannii; (c) Arbutus bicolor; (d) Quercus grisea. The foliage is evergreen in (a,b), contrasting with (c,d), which lose their leaves from March to June, with foliage formation reactivating from July onward.
Figure 2. Phenology in the field of four species in a mixed forest: (a) Juniperus deppeana; (b) Pinus engelmannii; (c) Arbutus bicolor; (d) Quercus grisea. The foliage is evergreen in (a,b), contrasting with (c,d), which lose their leaves from March to June, with foliage formation reactivating from July onward.
Forests 16 01118 g002
Forests 16 01118 g003
Figure 3. Spectrally extracted phenophases obtained through the UAV NDVI, differentiated by dotted vertical lines, showing the oscillation of the NDVI (0–1) on the left Y-axis. The correlation values of climate variables are denoted by colored horizontal lines. The asterisks represent statistical significance as p < 0.0001. Seasons are shown in shades of blue to white. The months in lowercase denote the year 2023, while those in uppercase pertain to 2024.
Figure 3. Spectrally extracted phenophases obtained through the UAV NDVI, differentiated by dotted vertical lines, showing the oscillation of the NDVI (0–1) on the left Y-axis. The correlation values of climate variables are denoted by colored horizontal lines. The asterisks represent statistical significance as p < 0.0001. Seasons are shown in shades of blue to white. The months in lowercase denote the year 2023, while those in uppercase pertain to 2024.
Forests 16 01118 g003
Table 1. Dasometric characteristics of the sampled trees.
Table 1. Dasometric characteristics of the sampled trees.
SpeciesVariablenMinMaxMeanSDAge
A. bicolorBD (cm)516.3031.5021.665.83
DBH (cm)59.8021.9014.204.80
CH (m)50.871.861.560.3946 ± 4
TH (m)54.017.575.701.30
J. deppeanaBD (cm)511.4019.1014.803.35
DBH (cm)58.1012.009.861.89
CH (m)51.952.061.990.0545 ± 3
TH (m)54.074.674.270.27
P. engelmanniiBD (cm)520.7033.8028.746.17
DBH (cm)516.3027.5023.205.1860 ± 2
CH (m)52.493.022.750.21
TH (m)57.3112.2010.141.93
Q. griseaBD (cm)523.7040.3033.546.67
DBH (cm)519.2032.6026.125.19
CH (m)52.463.002.770.2242 ± 1
TH (m)59.7112.8211.651.20
Key: n = number of trees; BD = basal diameter; DBH = diameter at breast height; CH = commercial height; TH = total height; min = minimum; max = maximum; mean = average; SD = standard deviation.
Table 2. Selected linear models of NDVI explaining phenophases as a function of climatic variables.
Table 2. Selected linear models of NDVI explaining phenophases as a function of climatic variables.
PhenophasePeriodVariablesModelR2
115 November 2023 to 15 February 2024
(Late autumn–winter)		Juniperus–PPNDVI = −0.08 × PP + 1.090.63
Pinus–PPNDVI = −0.06 × PP + 1.080.62
215 February 2024 to 14 June 2024
(Late winter–spring)		Quercus–PPNDVI = −0.14 × PP + 2.080.80
Quercus–TMAXNDVI = −0.02 × TMAX + 0.710.66
Quercus–VPDNDVI = −0.22 × VPD + 0.560.64
314 June 2024 to 14 October 2024
(Summer–early autumn)		Juniperus–PPNDVI = 0.08 × PP − 0.680.93
Juniperus–TMAXNDVI = 0.01 × TMAX + 0.0030.58
Juniperus–VPDNDVI = 0.09 × VPD + 0.240.57
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Pompa-García, M.; Vivar-Vivar, E.D.; Acosta-Hernández, A.C.; Rossi, S. The Phenophases of Mixed-Forest Species Are Regulated by Photo-Hydro-Thermal Conditions: An Approach Using UAV-Derived and In Situ Data. Forests 2025, 16, 1118. https://doi.org/10.3390/f16071118

AMA Style

Pompa-García M, Vivar-Vivar ED, Acosta-Hernández AC, Rossi S. The Phenophases of Mixed-Forest Species Are Regulated by Photo-Hydro-Thermal Conditions: An Approach Using UAV-Derived and In Situ Data. Forests. 2025; 16(7):1118. https://doi.org/10.3390/f16071118

Chicago/Turabian Style

Pompa-García, Marín, Eduardo Daniel Vivar-Vivar, Andrea Cecilia Acosta-Hernández, and Sergio Rossi. 2025. "The Phenophases of Mixed-Forest Species Are Regulated by Photo-Hydro-Thermal Conditions: An Approach Using UAV-Derived and In Situ Data" Forests 16, no. 7: 1118. https://doi.org/10.3390/f16071118

APA Style

Pompa-García, M., Vivar-Vivar, E. D., Acosta-Hernández, A. C., & Rossi, S. (2025). The Phenophases of Mixed-Forest Species Are Regulated by Photo-Hydro-Thermal Conditions: An Approach Using UAV-Derived and In Situ Data. Forests, 16(7), 1118. https://doi.org/10.3390/f16071118

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop