Next Article in Journal
Pine Wood Nematode’s Migration and Defense Mechanism of Highly Resistant and Susceptible Pinus massoniana
Previous Article in Journal
Spatiotemporal Variation and Pattern Analysis of Air Pollution and Its Correlation with NDVI in Nanjing City, China: A Landsat-Based Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 14
Issue 10
10.3390/f14102107
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 thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Long Term Seasonal Variability on Litterfall in Tropical Dry Forests, Western Thailand

by
Dokrak Marod
1,2,
Tohru Nakashizuka
3,
Tomoyuki Saitoh
4,
Keizo Hirai
3,
Sathid Thinkampheang
1,
Lamthai Asanok
5,
Wongsatorn Phumphuang
6,
Noppakun Danrad
1,* and
Sura Pattanakiat
7,*
1
Department of Forest Biology, Faculty of Forestry, Kasetsart University, Bangkok 10900, Thailand
2
Cooperate Centre of Thai Forest Ecological Research Network, Faculty of Forestry, Kasetsart University, Bangkok 10900, Thailand
3
Forestry and Forest Products Research Institute, Tsukuba 305-8687, Japan
4
Tohoku Research Center, Forestry and Forest Products Research Institute, Morioka 020-0123, Japan
5
Department of Agroforestry, Maejo University, Phrae Campus, Phrae 54140, Thailand
6
Department of National Parks, Wildlife and Plant Conservation, Bangkok 10900, Thailand
7
Faculty of Environment and Resource Studies, Mahidol University, Nakhonpathom 73170, Thailand
*
Authors to whom correspondence should be addressed.
Forests 2023, 14(10), 2107; https://doi.org/10.3390/f14102107
Submission received: 27 September 2023 / Revised: 18 October 2023 / Accepted: 18 October 2023 / Published: 20 October 2023
(This article belongs to the Topic Litter Decompositions: From Individuals to Ecosystems)
Browse Figures
Review Reports Versions Notes

Abstract

:
Nutrient recycling is one of the most important services that supports other processes in ecosystems. Changing litterfall patterns induced by climate change can cause imbalances in nutrient availability. In this study, we reported the long-term (28-year) interplay between environmental factors and variability among litterfall fractions (leaves, flowers, and fruit) in a tropical dry forest located in Kanchanaburi, Thailand. A long-term litter trap dataset was collected and analyzed by lagged generalized additive models. Strong seasonality was observed among the litter fractions. The greatest leaf and flower litterfall accumulated mostly during the cool, dry season, while fruit litterfall occurred mostly during the rainy season. For leaf litter, significant deviations in maximum temperature (Tmax), volumetric soil moisture content (SM), and evapotranspiration (ET) during the months prior to the current litterfall month were the most plausible factors affecting leaf litter production. Vapor pressure deficit (VPD) and ET were isolated as the most significant factors affecting flower litterfall. Interestingly, light, mean temperature (Tmean), and the southern oscillation index (SOI) were the most significant factors affecting fruit litterfall, and wetter years proved to be highly correlated with elevated fruit litterfall. Such environmental variability affects both the triggering of litterfall and its quantity. Shifting environmental conditions can therefore alter nutrient recycling rates through the changing characteristics and quantity of litter.

1. Introduction

Litterfall is a fundamental ecological process that influences carbon and nutrient cycling as a result of the decomposition of aboveground litter into soil [1,2]. Litter is considered essential for the maintenance of soil fertility in terrestrial ecosystems [3,4]. The seasonality of litterfall is tightly related to plant vegetative phenology (i.e., leaf emergence, development, aging, and abscission) because most of the litterfall necromass is composed of leaves [5]. Climate change is a major driving force of variability in litterfall production, its spatiotemporal patterns, and the dynamics of productivity and nutrient balance in forest ecosystems [6,7]. Significant variables that affect litterfall dynamics include temperature [8,9], precipitation [10,11], temperature-precipitation interactions [12], photoperiod [13,14], and topography, in conjunction with soil water retention capacity [15].
By reducing relative humidity, wind can dry the soil (especially at high temperatures) and increase the VPD [16]. These conditions can increase rates of leaf and branch fall, which can be seen as adaptations to avoid transpiration water losses and cope with water stress [17]. However, litterfall seasonality varies among forest ecosystems and tree species [18,19], and can display unimodal, bimodal, or irregular patterns through time [3,20]. Climate change can also affect tree phenology by stimulating the irregular production of flowers and fruits, which in turn can increase the variability of total leaf litterfall [21,22].
Tropical dry forests (TDFs) are characterized by prolonged dry seasons [23,24]. They experience high intra-annual variations in climate and usually experience less than 100 mm during the months of the dry season [25]. Tropical dry forests exhibit distinct differences in patterns of vegetative and reproductive phenology when compared to other biomes [26]. Moreover, phenological events in TDFs are mostly caused by seasonal variations in rainfall, rather than temperature [27,28]. Strong water deficits are a crucial trigger for leaf shedding, and their peak often occurs during the dry season [20,29]. Up to 95% of leaves may be lost from living trees during the dry season [30], which causes a massive accumulation of litter necromass on the soil surface [10]. In TDFs, leaf shedding by deciduous species usually peaks during the dry season, whereas litterfall in evergreen species does not generally show any obvious seasonal pattern [20,31].
Broad variations in resource acquisition and reproductive strategies among TDF tree species could be associated with interspecific differences in phenological responses to seasonal cycles and to phase changes in the southern oscillation index (SOI) [32]. The SOI is one measure of the large-scale fluctuations in air pressure occurring between the western and eastern tropical Pacific during El Niño and La Niña episodes. Recent studies have shown that litterfall seasonality is associated with climate seasonality [5,20,33,34]. Some seasonal features of litterfall can be captured through short-term (less than 1 year) datasets with a minimal variability of 10% [5], but long-term monitoring is necessary to detect spatiotemporal variations in litterfall that are caused by interannual climate variability. Moreover, identifying the factors that govern litterfall production can help us understand more fully the response of forest ecosystems to climate change [35,36]. Most studies on litterfall production conducted in TDFs have been short-term in nature [37,38], particularly in Southeast Asia, compared to Central Africa and South America [33]. Investigations of the long-term dynamics of litterfall fractions, such as the 30-year study by Detto et al. [32], have been relatively rare [39].
In this study, we investigated interannual variation in aboveground litterfall over a 28-year period in a tropical dry forest (TDF) in western Thailand. Our objectives were to clarify (1) what climatic variables determine the seasonal variation of litterfall production and (2) investigate whether litterfall production varies over the years during long-term monitoring.

2. Materials and Methods

2.1. Study Site

The study was conducted at Mae Klong Watershed Research Station (MKWRS) in Thong Pha Phum district, western Thailand (Figure S1). Litterfall data were collected between January 1993 and December 2021 in litter traps that were placed in a 4-ha (200 m × 200 m) forest dynamic plot (FDP) of TDF. The TDF was located between 14°30′ and 14°45′ N and from 98°45′ to 99° E. The mean monthly temperature is approximately 27.5 °C, with a maximum of 39.1 °C in April and a minimum of 14.6 °C in December. The climate is affected by monsoons, and annual rainfall usually exceeds 1650 mm, most of which falls during the rainy season from April to October. The altitude of the study site ranged from 150 to 350 m a.s.l. Soil was classified as an alfisol derived from sedimentary rock, gneiss, and limestone. Topsoil was approximately 20 cm deep, and its texture varied from sandy clay to sandy clay loam. The subsoil was approximately 2 m in depth and was composed of clay. The site was moderately well-drained [40].
On the permanent plot, it is mostly covered by mature MDF, nearly 30 m tall, with some patches of deciduous dipterocarp forest (DDF) on mountain ridges and dry evergreen forest (DEF) along the creeks. The tree density was 171 individuals ha−1, and the basal area was 17.2 m2 ha−1. The dominant tree species were mostly deciduous and included Pentacme siamensis, Xylia xylocarpa var. kerrii, and Pterocarpus macrocarpus. Some evergreen species, such as Dipterocarpus alatus, grew on the valley floor [41]. In MDF, bamboos dominate the middle layer of the canopy profile and achieve heights of 10–15 m. Fires are frequent during the dry or summer seasons [42]. Three dominant species, D. alatus (Dipala), P. macrocarpus (Ptemac), and P. siamensis (Pansia), which are representative of DEF, MDF, and DDF, respectively [41], were analyzed to detect seasonal variations in litterfall.

2.2. Litterfall Collection

In 1993, a 1-ha (100 m × 100 m) area within the 4-ha plot was selected for the intensive monitoring of litterfall. One hundred litter traps were arranged in a regular matrix at 10 m × 10 m spacing. Each litter trap had an opening of 0.5 m2 (0.7 m × 0.7 m) and was created from a stainless-steel net (mesh size, 2 mm) to ensure protection from fire. Traps were set approximately 1 m above the forest floor. Litter was collected monthly over a period of 28-year between January 1993 and December 2021. Litter samples were oven-dried at 70 °C for 48 h and sorted into leaves, flowers, fruits, and other components (bark, twigs, and branches), based on the methods of [43]. Dry biomass of each component was identified by species and weighed using an analytical balance with a resolution of 0.01 g. The sum of the component dry weights was referred to as the total litterfall.

2.3. Climate Data

The daily weather data during the study period was measured locally from the MKWRS weather station and included evaporation (Evap, mm), rainfall (Rain, mm), maximum temperature (Tmax, °C), mean temperature (Tmean, °C), minimum temperature (Tmin, °C), and relative humidity (RH, %). Vapor pressure deficit (VPD, kPa), windspeed (Vs, ms−1), light (light, MJ m−2 day−1), volumetric soil moisture content (SM, %), Palmer drought index (PI, unitless), and evapotranspiration (ET, mm) were collected from the TerraClim database (URL: https://www.climatologylab.org/terraclimate.html, assessed on 7 March 2023).

2.4. Data Analysis

To characterize the temporal variability in litterfall, excluding bamboo litterfall, we calculated monthly and annual means and standard deviations for the 28-year study period. To determine the contribution of every component (leaf, flower, fruit, and other components), the respective percentages were first calculated and were then averaged to generate monthly and yearly litterfall variations as a fractional contribution to the total litter pool. Additional exploratory analyses were conducted to determine whether trends in litterfall correlated with the values of the climate variables.
The determination of underlying links between litterfall and climatic variables can have both spatial and temporal components and requires a precise understanding of the estimated “instantaneous” values of the variables [44]. As mentioned above, the interest in modeling litterfall through a generalized additive model (GAM) was primarily to detect any significant effect of environmental variables on litterfall types and to determine the potential influence of lagged environmental variables. We investigated the environmental drivers influencing litterfall components using GAM. Over the linear models, GAM has the added advantage of capturing any nonlinear temporal features in a litterfall time series and can be used to model the relationship between the mean response variable and ‘smoothed’ functions of the explanatory variables [45,46]. In the current analysis, we used the monthly and annual scales of resolution as fixed factors, with the Gaussian distribution as the identity link function.
Similar GAM structures were used to model monthly litterfall components as a function of environmental variables. Collinear environmental variables were identified by calculating variance inflation factors between variable pairs and were removed from the analysis. Model fits were used to examine the effects of climate variables on each litter fraction, as indicated by significant p-values (at a 95% confidence interval). We also tested the explanatory power of environmental variables at various lags, ranging from zero (simultaneous with the current litterfall month) to five months before a given litterfall measurement using a lagged GAM approach.
The lagged GAM model was used to plot the temporal variability of statistically significant environmental variables and to test whether incremental changes in the values of these variables could increase or decrease the probability of litterfall (as quantified by the odds ratio). Environmental variables during the months leading up to and during litterfall have been reported to correlate with leaf phenology in tropical forests [34]. We therefore used a window spanning the current litterfall month to five months prior to current litter production to test for significant environmental forcing. All the analyses were performed using R statistical software version 3.6.1 [47], and GAMs were modeled using the “mgcv” package [48]. All analyses were performed using R statistical software version 3.6.1 [47], and GAMs were modeled using the “mgcv” package [48].

3. Results

3.1. Climate Regime

Twenty-eight years of climatic data (1993–2021) showed that MKWRS experiences a monsoon climate with marked seasonal variations in temperature and rainfall. The average annual rainfall was 1615 ± 236 mm, and no significant temporal rainfall trend was detected. Rainfall seasonality was distinct throughout the study period (Figure S2a,b), and the rainy period generally began in late April and extended until late October. Peak rainfall occurs from July to September and accounts for approximately 53% of the annual total. The rainfall was very low from November to March, although small rainfall events (<10 mm) occurred occasionally. Depending on the intra-annual variation in temperature, three seasons could be observed in the area, namely winter (November–January), summer (February–April), and the rainy season (May–October). The cold or winter season starts in November, lasts until February, and experiences lower temperatures (15–19 °C) and significantly less rainfall. During the summer, temperatures range from 35 to 38 °C (Figure S2b).
When considering the potential climate change signature, we found that maximum and mean temperatures displayed significant decreasing trends (p < 0.001) over the 28-year period (Figure 1a,b). A contrasting trend was found for evaporation (Figure 1c), indicating that increased evaporation might lead to droughts in the future.

3.2. Litter Production

Total annual litter production was 9.46 ± 1.56 Mg ha−1 yr−1 (range = 6.79–12.22 Mg ha−1 yr−1). Annual litter production varied among study years, and a slightly decreasing trend was found (p < 0.001, Figure 2). The greatest litterfall was recorded in 2002 (12.22 Mg ha−1 yr−1) and the lowest in 1999 (6.79 Mg ha−1 yr−1). The monthly variation of litterfall components is shown in Figure 3. Peaks in leaf litter and other fractions were uniformly distributed during the 28-year study period and were distributed unimodally across seasons. Flower and fruit litter production decreased when the first 10-year period (1993–2003) and the last 20-year period (2004–2021) were compared, and leaf litter displayed a slightly decreasing trend (Figure 3b,c).
When split into their litter components, mean leaf litterfall was 6.26 ± 1.14 Mg ha−1 yr−1, mean flower litter was 0.16 ± 0.15 Mg ha−1 yr−1, mean fruit litter was 0.41 ± 0.37 Mg ha−1 yr−1 for fruit litter, and other components (bark, twigs, and branches) contributed 2.63 ± 0.72 Mg ha−1 yr−1. Leaf litter represented 67% of the total litterfall, while 26% was contributed by the category of other components. Fruit litterfall accounted for 5% of the total, and flower litterfall contributed the remaining 2%. Annual variations in the various litterfall components are shown in Figure 4a. In Figure 4b, leaf litter is shown to have reached its maximum during the cool dry season (November–February) and lowest during the rainy season (May–August). Flower litter followed a similar pattern, with peak contribution during November–February, while the contribution of fruit to the litter pool was highest between March and September, indicating that fruit fall followed flower and leaf fall. The “other” component of litter followed a similar trend to fruit fall.
Litterfall trends among evergreen D. alatus, deciduous P. macrocarpus, and P. siamensis are plotted in Figure 5. Although temporal patterns of litterfall components varied among species, peak leaf fall for all three occurred during the cool, dry season, and only slight differences were recorded among them. Peak litterfall for D. alatus occurred in November, which was earlier than the peaks for other species, which took place in January (Figure 5a). Leaf litterfall was highest for P. siamensis (0.37 ± 0.11 Mg ha−1 yr−1, 5.89% of total leaf litterfall), followed by D. alatus (0.22 ± 0.05 Mg ha−1 yr−1. 3.58% of the total), and P. siamensis (0.11 ± 0.05 Mg ha−1 yr−1, 1.77% of the total). Flowers and fruit litterfall were similar in magnitude (Figure 5b,c). Peak flower fall for D. alatus took place during the same period as leaf litterfall. However, fruit and flower peaks for the two deciduous species took place slightly after peak leaf fall during the hot, dry season. Fruit fall for all three species peaked during the hot, dry (March–April) season. The apparent peaks of leaf and flower fall for evergreen D. alatus occurred during the cool, dry season, with the fruit fall peaking two months later. However, in P. siamensis and P. macrocarpus, deciduous species, the peak leaf fall occurs two months before the flower and fruit fall.

3.3. Relationship between Litterfall Seasonality and Climate Change

A GAM was used to fit the litterfall components pooled across species. The major variables affecting the litterfall components are listed in Table 1, with only leaf, flower, and fruit fractions modeled. Model fits returned R2 values between 0.64 and 0.82, with the overall deviance explained being greater than 67%. All the litterfall components displayed strong trends and seasonality, as indicated by statistically significant yearly and monthly variations.
The lagged GAM showed that leaf litterfall was influenced by SM, ET, and Tmax (Figure 6 and Table 1). Figure 6 demonstrates how the effect of these significant environmental variables changed at lags of 0–5 months. The strongest influence was exerted by SM, which had a statistically significant positive effect on litterfall one and two months prior to collection (with a u-shaped functional relationship) for values between 0.25 and 0.35. However, SM measured four months before litter collection was associated with reduced leaf shedding. Other lags failed to exhibit statistically significant relationships with litter fractions, as indicated by odds ratios < 1. Elevated levels of ET at lags of 3–5 months were associated with greater litterfall. However, high ET values at lags of 1–2 months were associated with reduced leaf litterfall.
The effects of Tmax at different lags were divided around temperatures above and below 35 °C. Lags over 1–4 months appeared to display a more linear relationship with litterfall. Tmax > 35 °C at lag zero did not cause significant litterfall, while the same temperatures at lags of five months may have increased litterfall. Values of Tmax at a lag of two months had a positive linear effect on leaf litterfall. Overall, Tmax > 35 °C appears to promote litterfall and high ET (>100 mm), which could be balanced by SM values between 0.25 and 0.35, which could ensure adequate water uptake.
Flower litterfall was influenced by ET, VPD, and light (Figure 7 and Table 1). Evapotranspiration exerted the strongest effect on litterfall and appeared to promote leaf litterfall at lags of 1–2 months. However, at ET values of 75–140 mm, lags of two months were associated with double flower fall. The effects of VPD were oscillatory, with lags of 3–4 months being different from lags of 1 month. Normal air dryness is characterized by VPD values of 0–2 kPa, with higher values indicating greater relative drought. We observed that VPD values of 0.8–1.1 kPa and >2 kPa were associated with increased flower litterfall, while VPD values of 1.3–1.6 kPa were associated with reduced flower litterfall. A VPD > 2 kPa would result in increased litterfall at all lags tested in the model. Available light between 17–21 MJ m−2 day−1 was associated with increased flower litterfall at lags of 1–3 months. Overall, flower litterfall in the study of the forest was associated with ET between 75 and 140 mm, VPD of 0.8–1.1 kPa, and light levels of 17–21 MJ m−2 day−1. Such conditions are very specific relative to the leaf fall, given that peak leaf and flower falls occur during the cool, dry season.
The GAM analysis showed that fruit litterfall was influenced by SOI, Tmean, and light (Figure 8 and Table 1), with the effects of SOI having a highly cyclic pattern. Values of SOI > +8 (indicating wet years or La Niña) and −8 (indicative of dry years or El Niño) increased fruit litterfall at most lags, with most of the fruit litterfall promoted by La Niña (wet) events. The effect of Tmean was comparatively stable and monotonically increased fruit litterfall at temperatures of 20–35 °C at lags of 0–1 month. However, increasing Tmean at lags of 3–5 months tended to reduce fruit litterfall. The influence of light was also oscillatory in nature, with values between 16 and 21 MJ m−2 day−1 at lags of 0, 2, and 5 months tending to increase the fruit litterfall. A shift in the peak light intensity affecting litterfall was also observed at 19 MJ m−2 day−1 at a zero lag and 17.5 MJ m−2 day−1 at lags of 4–5 months. In summary, fruit litterfall was influenced by oscillatory SOI and light, accompanied by a monotonically increasing Tmean.
Yearly plots of SOI and fruit litterfall for individual species (Figure 9a) indicated that the peaks in fruit litterfall were significantly correlated (p = 0.009) with La Niña (>+8 or wet) and El Niño (<−8 or dry) events. Neither leaf nor flower litterfall responded in this way. The greatest values of peak fruit litterfall were found during La Niña, particularly the 1999–2000 and 1995–1996 events. Fruit production in the MDF was therefore promoted under wet conditions, while fruit formation may have been inhibited by dry conditions during La Niña events, particularly the dry years of 1997–1998. SOI apparently had no significant effects on fruit litterfall for selected species, however, some variational trends were found (Figure 9b). A high necromass of fruit litterfall in D. alatus was mostly found during the dry El Niño years of 2003 and 2007, which contrasted with the peak fruit litterfall in P. siamensis, which occurred during the La Niña years of 1996, 2000, and 2008. Contrasting with both of these species, P. macrocarpus exhibited a constant production of fruit litterfall.

4. Discussion

In this study, we studied the variability of litterfall components using 28 years of litter trap data in a TDF in western Thailand. We estimated litterfall necromass and the intra- and interannual distribution of necromass in leaf, flower, and fruit components, as well as seasonal patterns, together with a set of meteorological variables that had the potential to influence litter production. The use of such long-term datasets can help identify possible patterns of climatic influence over litterfall production [33,49].

4.1. Annual Litterfall Production and Seasonality

Tropical forests usually produce more litter than other terrestrial biomes [50]. In this study, we found that annual litterfall production ranged from 6.79 to 12.22 Mg ha−1 yr−1, with an average value of 9.46 ± 1.56 Mg ha−1 yr−1. These values are generally higher than litterfall production measured in other TDFs, which ranged from 3.8 to 7.7 Mg ha−1 yr−1 in previous studies [27,38,51,52,53,54]. However, our measurements were slightly lower than those reported for humid tropical forests, such as lower montane forest in Thailand (Marod et al. [8]; 9.43–12.22 Mg ha−1 yr−1), primary humid forest in Amazonia (Barlow et al. [9]; 9.4–12.4 Mg ha−1 yr−1), and old-growth upper montane forest in Costa Rica (Köhler et al. [55]; 12.27–13.49 Mg ha−1 yr−1).
Relative to tropical moist forests, there is a distinct seasonality of litterfall production in TDFs, and most litterfall takes place during the dry season [9,22,55,56,57]. Overall litter production is similar in TDFs and tropical moist forests [58]. As in previous studies, our results demonstrated clear seasonal litterfall patterns with major peaks during the dry season and significant yearly and monthly variations (Table 1). The leaf litterfall fraction was the major component of litterfall production and comprised approximately 67% of the total (Figure 5). These values were broadly comparable to the 60%–76% reported for forests worldwide [50]. The maximum leaf contribution was found during the cool, dry season (November–February) and reached its minimum during the rainy season (May–August). A similar pattern was found in the flower litterfall, while fruit litterfall followed flower litterfall.
It is notable that deciduous tree species in a given habitat can differ dramatically in their timing of leaf emergence and abscission compared to evergreen species [59]. In our study, the temporal pattern of litterfall components varied among three selected species (Figure 5). Although the peak of leaf litterfall mostly occurred during the cool dry season (November–January), evergreen D. alatus, which was found along the banks of a creek running through the permanent plot, shed its leaves 1–2 months earlier than deciduous P. macrocarpus and P. siamensis.

4.2. Relationship between Litterfall Seasonality and Climate Change

Environmental factors can produce either delayed or instantaneous responses in actual litterfall, being influenced by a complex interaction of biotic and abiotic variables [60,61]. In addition, climate change may exert an important influence over litterfall dynamics, even in evergreen species [51,62]. Rainfall, temperature, wind speed, relative humidity, and light availability can all affect litterfall production [20,63]. Leaf fall occurs after the buildup of seasonal stresses, particularly water stress related to soil moisture and temperature during the dry season [64,65]. In this study, we found that the main driving variables differed among litterfall components (Table 1). Leaf litterfall was influenced by SM, ET, and Tmax (Figure 7 and Table 1), flower litterfall was influenced by ET, VPD, and light (Figure 8 and Table 1), and fruit litterfall was influenced by SOI, Tmean, and light (Figure 9 and Table 1).
Peak leaf and flower litterfall had lagged responses of 1–4 months to environmental forcing. Peak leaf litterfall occurred 1–2 months after the strongest influence of SM, and a similar trend was found for Tmax. The effects of ET were felt at lags of 3–5 months, with higher ET values resulting in higher litterfall. Peak litterfall, therefore, happened during the cool, dry period, just after the rainy season. The strongest association of ET with flower litterfall was observed at lags of 1–2 months. Peak flower litterfall occurred during the dry season in response to high evaporative demand and transpiration under elevated VPD and light levels (Figure 8). Such conditions are very specific relative to leaf litterfall, given that peak leaf and flower litterfall occur during the cool, dry season. A lag in litterfall response was observed by Zalamea and González [13], with peak total litterfall production occurring two weeks after rainfall. Detto et al. [32] reported lags that were related to seasonal and ENSO cycles. de Queiroz et al. [66] reported a four-month lag following the onset of the dry season in peak leaf deposition and observed flower fall, which occurred 2–3 months following the onset of rains.
In contrast to leaf and flower litterfall, fruit litterfall responds instantaneously to environmental conditions, such as lightning storms accompanied by winds [67,68,69]. Higher fruit litter production during the wet season could constitute a propagation strategy, as the increased soil water availability benefits seed germination and subsequent seedling growth [70]. Our results showed a response to fruit litterfall for individual species during the wet months (Figure 9a). In addition, fruit litterfall was influenced by cyclic patterns in the SOI. Fruit litterfall during ENSO events tended to increase at most of the lags and was enhanced by wet La Niña events, particularly in 1999–2000 and 1995–1996. The La Niña influence contrasted with that of El Niño, particularly with regard to the severe drought in 1997–1998. Fruit litterfall patterns also varied among selected species. In particular, fruit litterfall peaked in evergreen D. alatus under moderate El Niño conditions (Figure 9b).
Drought could therefore emerge as an important driver of fruiting phenomena by triggering mast flowering, particularly in species from the Dipterocarpaceae [21,71,72]. A contrasting pattern was found for P. siamensis (family Dipterocarpaceae, deciduous), for which fruit litterfall peaked under La Niña conditions. Comparable results have been reported from the wet tropics. Working at Barro Colorado Island, Panama, Wright et al. [73] found that high forest-wide fruit production occurred during El Niño events, with low fruit production following a mild dry season one year later. In peninsular tropical Malaysia, Ashton et al. [74] also found that El Niño drought years produced mass flowerings, particularly among Dipterocarpaceae, and induced mast fruiting on windward slopes. In the wet tropics, El Niño droughts could enhance light availability by reducing cloud cover [14,75]. This condition may stimulate mass flowering and fruiting production [73]. This knowledge can improve our understanding of the responses of tropical forest ecosystems to climate change [34,36], which may lead to the degradation of ecosystem services provided by such forests.

5. Conclusions

We investigated long-term (28-year) seasonal and annual patterns of litterfall production in a TDF in western Thailand. Our results showed that total annual litterfall varied from year to year across a range of 6.79–12.22 Mg ha−1 yr−1, and that leaf litterfall was the main component of litter production. Strong seasonality was observed in all litter fractions, and the greatest leaf and flower litterfall accumulated principally during the cool, dry season, while fruit litterfall occurred mostly during the rainy season. Environmental variables associated with litterfall varied among the litterfall components. For leaf litter, significant deviations in maximum temperature (Tmax), volumetric soil moisture content (SM), and evapotranspiration (ET) during the months prior to litterfall collection were the strongest correlates of litterfall. The most significant factors affecting flower litterfall were VPD, light, and ET. Interestingly, light, Tmean, and the SOI were the most significant factors affecting fruit litterfall, and a high correlation was observed between the occurrence of wet La Niña events and elevated fruit production. Significant lags of up to five months between climatic variables and peak litterfall production were recorded. Variability in litterfall production could increase during climate change since interannual variation in climatic patterns affects both the triggering of litterfall and its quantity. Such changing patterns could potentially alter forest function and nutrient cycling. Our results can be used to monitor the future alleviation of climate change-related effects on litterfall.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/f14102107/s1, Figure S1: The location of permanent plot in the mixed deciduous forest of the Mae Klong Watershed Re-search Station (MKWRS), Kanchanaburi province, western Thailand; Figure S2: (a) Monthly rainfall variation (dots) and (b) total monthly rainfall with mean monthly maximum and minimum temperature for 1992–2021 at MKWRS, western Thailand.

Author Contributions

Conceptualization, D.M. and T.N.; methodology, T.S., K.H. and S.T.; investigation, validation, data curation, D.M., S.T., W.P., L.A. and N.D.; formal analysis, N.D., L.A. and S.P.; writing—original draft preparation, S.P. and N.D.; writing—review and editing, D.M., T.N. and K.H.; project administration, D.M. and T.S.; funding acquisition, D.M., T.N. and K.H. All authors have read and agreed to the published version of the manuscript.

Funding

This work was financially supported by the Office of the Ministry of Higher Education, Science, Research, and Innovation; and the Thailand Science Research and Innovation through the Kasetsart University Reinventing University Program 2021. The Japan Society for the Promotion of Science, JSPS, Grant-in-Aid for Scientific Research (B) (KAKENHI Grant Number JP22H02395) also supported.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to all data generated during this study are included in this article.

Acknowledgments

We thank previous professors who initiated the long-term ecological research at MKWRS, such as Utis Kutintara, Chanchai Yarwudhi, Shigeo Kobayashi, and Masamichi Takahashi. We also thank the editors and reviewers for their detailed and constructive comments.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Berg, B. Litter decomposition and organic matter turnover in northern forest soils. For. Ecol. Manag. 2000, 133, 13–22. [Google Scholar] [CrossRef]
  2. Attiwill, P.M.; Adams, M.A. Nutrient cycling in forests. New Phytol. 1993, 124, 561–582. [Google Scholar] [CrossRef]
  3. Bigelow, S.W.; Canham, C.D. Litterfall as a niche construction process in a northern hardwood forest. Ecosphere 2015, 6, 1–14. [Google Scholar] [CrossRef]
  4. Sayer, E.J.; Joseph Wright, S.; Tanner, E.V.J.; Yavitt, J.B.; Harms, K.E.; Powers, J.S.; Kaspari, M.; Garcia, M.N.; Turner, B.L. Variable responses of lowland tropical forest nutrient status to fertilization and litter manipulation. Ecosystems 2012, 15, 387–400. [Google Scholar] [CrossRef]
  5. Chave, J.; Navarrete, D.; Almeida, S.; Álvarez, E.; Aragão, L.E.O.C.; Bonal, D.; Châtelet, P.; Silva-Espejo, J.E.; Goret, J.Y.; von Hildebrand, P. Regional and seasonal patterns of litterfall in tropical South America. Biogeosciences 2010, 7, 43–55. [Google Scholar] [CrossRef]
  6. Kho, L.K.; Malhi, Y.; Tan, S.K.S. Annual budget and seasonal variation of aboveground and belowground net primary productivity in a lowland dipterocarp forest in Borneo. J. Geophys. Res. Biogeosci. 2013, 118, 1282–1296. [Google Scholar] [CrossRef]
  7. Sayer, E.J.; Powers, J.S.; Tanner, E.V.J. Increased litterfall in tropical forests boosts the transfer of soil CO2 to the atmosphere. PLoS ONE 2007, 2, e1299. [Google Scholar] [CrossRef]
  8. Marod, D.; Andriyas, T.; Leksungnoen, N.; Kjelgren, R.; Thinkamphaeng, S.; Chansri, P.; Asanok, L.; Hermhuk, S.; Kachina, P.; Thongsawi, J.; et al. Potential variables forcing litterfall in a lower montane evergreen forest using Granger and superposed epoch analyses. Ecosphere 2023, 14, e4572. [Google Scholar] [CrossRef]
  9. Barlow, J.; Gardner, T.A.; Ferreira, L.V.; Peres, C.A. Litter fall and decomposition in primary, secondary and plantation forests in the Brazilian Amazon. For. Ecol. Manag. 2007, 247, 91–97. [Google Scholar] [CrossRef]
  10. Lopes, M.C.A.; Araújo, V.F.P.; Vasconcellos, A. The effects of rainfall and vegetation on litterfall production in the semiarid region of northeastern Brazil. Braz. J. Biol. 2015, 75, 703–708. [Google Scholar] [CrossRef]
  11. Campanella, M.V.; Bertiller, M.B. Leaf litterfall patterns of perennial plant species in the arid Patagonian Monte, Argentina. Plant Ecol. 2010, 210, 43–52. [Google Scholar] [CrossRef]
  12. Saarsalmi, A.; Starr, M.; Hokkanen, T.; Ukonmaanaho, L.; Kukkola, M.; Nöjd, P.; Sievänen, R. Predicting annual canopy litterfall production for Norway spruce (Picea abies (L.) Karst.) stands. For. Ecol. Manag. 2007, 242, 578–586. [Google Scholar] [CrossRef]
  13. Zalamea, M.; González, G. Leaffall phenology in a subtropical wet forest in Puerto Rico: From species to community patterns. Biotropica 2008, 40, 295–304. [Google Scholar] [CrossRef]
  14. Wright, S.J.; van Schaik, C.P. Light and the phenology of tropical trees. Am. Nat. 1994, 143, 192–199. [Google Scholar] [CrossRef]
  15. Méndez-Alonzo, R.; Pineda-García, F.; Paz, H.; Rosell, J.A.; Olson, M.E. Leaf phenology is associated with soil water availability and xylem traits in a tropical dry forest. Trees 2013, 27, 745–754. [Google Scholar] [CrossRef]
  16. Smith, V.C.; Ennos, A.R. The effects of air flow and stem flexure on the mechanical and hydraulic properties of the stems of sunflowers Helianthus annuus L. J. Exp. Bot. 2003, 54, 845–849. [Google Scholar] [CrossRef]
  17. Aryal, D.R.; De Jong, B.H.J.; Ochoa-Gaona, S.; Mendoza-Vega, J.; Esparza-Olguin, L. Successional and seasonal variation in litterfall and associated nutrient transfer in semi-evergreen tropical forests of SE Mexico. Nutr. Cycl. Agroecosystems 2015, 103, 45–60. [Google Scholar] [CrossRef]
  18. Harrison, R.D. Adaptive significance of phenological variation among monoecious hemi-epiphytic figs in Borneo. Symbiosis 2008, 45, 83–90. [Google Scholar]
  19. Reich, P.B. Phenology of tropical forests: Patterns, causes, and consequences. Canad. J. Bot. 1995, 73, 164–174. [Google Scholar] [CrossRef]
  20. Zhang, H.; Yuan, W.; Dong, W.; Liu, S. Seasonal patterns of litterfall in forest ecosystem worldwide. Ecol. Complex. 2014, 20, 240–247. [Google Scholar] [CrossRef]
  21. Sakai, S.; Harrison, R.D.; Momose, K.; Kuraji, K.; Nagamasu, H.; Yasunari, T.; Chong, L.; Nakashizuka, T. Irregular droughts trigger mass flowering in aseasonal tropical forests in Asia. Am. J. Bot. 2006, 93, 1134–1139. [Google Scholar] [CrossRef] [PubMed]
  22. Liu, W.; Fox, J.E.D.; Xu, Z. Litterfall and nutrient dynamics in a montane moist evergreen broad-leaved forest in Ailao Mountains, SW China. Plant Ecol. 2002, 164, 157–170. [Google Scholar] [CrossRef]
  23. Miles, L.; Newton, A.C.; DeFries, R.S.; Ravilious, C.; May, I.; Blyth, S.; Kapos, V.; Gordon, J.E. A global overview of the conservation status of tropical dry forests. J. Biogeogr. 2006, 33, 491–505. [Google Scholar] [CrossRef]
  24. Murphy, P.G.; Lugo, A.E. Dry forests of Central America and the Caribbean. In Seasonally Dry Tropical Forests; Bullock, S.H., Mooney, H.A., Medina, E., Eds.; Cambridge University Press: Cambridge, UK, 1995; Volume 85, pp. 9–34. [Google Scholar]
  25. Sánchez-Azofeifa, G.A.; Kalacska, M.; Quesada, M.; Calvo-Alvarado, J.C.; Nassar, J.M.; Rodríguez, J.P. Need for Integrated Research for a Sustainable Future in Tropical Dry Forests. Conserv. Biol. 2005, 19, 285–286. [Google Scholar] [CrossRef]
  26. Morellato, L.P.C.; Talora, D.C.; Takahasi, A.; Bencke, C.C.; Romera, E.C.; Zipparro, V.B. Phenology of Atlantic rain forest trees: A comparative study. Biotropica 2000, 32, 811–823. [Google Scholar] [CrossRef]
  27. Lawrence, D. Regional-Scale Variation in Litter Production and Seasonality in Tropical Dry Forests of Southern Mexico. Biotropica 2005, 37, 561–570. [Google Scholar] [CrossRef]
  28. Eamus, D.; Prior, L. Ecophysiology of trees of seasonally dry tropics: Comparisons among phenologies. Adv. Ecol. Res. 2001, 32, 113–197. [Google Scholar]
  29. Mooney, H.A.; Bullock, S.H.; Medina, E. Introduction. In Seasonally Dry Tropical Forests; Bullock, S.H., Mooney, H.A., Medina, E., Eds.; Cambridge University Press: Cambridge, UK, 1995; Volume 85, pp. 1–8. [Google Scholar]
  30. Pezzini, F.F.; Ranieri, B.D.; Brandão, D.O.; Fernandes, G.W.; Quesada, M.; Espírito-Santo, M.M.; Jacobi, C.M. Changes in tree phenology along natural regeneration in a seasonally dry tropical forest. Plant Biosyst. 2014, 148, 965–974. [Google Scholar] [CrossRef]
  31. Meentemeyer, V.; Box, E.O.; Thompson, R. World patterns and amounts of terrestrial plant litter production. BioScience 1982, 32, 125–128. [Google Scholar] [CrossRef]
  32. Detto, M.; Wright, S.J.; Calderón, O.; Muller-Landau, H.C. Resource acquisition and reproductive strategies of tropical forest in response to the El Niño–Southern Oscillation. Nat. Commun. 2018, 9, 913. [Google Scholar] [CrossRef]
  33. Nakagawa, M.; Ushio, M.; Kume, T.; Nakashizuka, T. Seasonal and long-term patterns in litterfall in a Bornean tropical rainforest. Ecol. Res. 2019, 34, 31–39. [Google Scholar] [CrossRef]
  34. Wagner, F.H.; Hérault, B.; Bonal, D.; Stahl, C.; Anderson, L.O.; Baker, T.R.; Becker, G.S.; Beeckman, H.; Boanerges Souza, D.; Botosso, P.C. Climate seasonality limits leaf carbon assimilation and wood productivity in tropical forests. Biogeosciences 2016, 13, 2537–2562. [Google Scholar] [CrossRef]
  35. Thuille, A.; Schulze, E.D. Carbon dynamics in successional and afforested spruce stands in Thuringia and the Alps. Glob. Chang. Biol. 2006, 12, 325–342. [Google Scholar] [CrossRef]
  36. Scheer, M.B.; Gatti, G.; Wisniewski, C. Nutrient fluxes in litterfall of a secondary successional alluvial rain forest in Southern Brazil. Rev. Biol. Trop. 2011, 59, 1869–1882. [Google Scholar] [CrossRef]
  37. Parsons, S.A.; Congdon, R.A.; Shoo, L.P.; Valdez-Ramirez, V.; Williams, S.E. Spatial variability in litterfall, litter standing crop and litter quality in a tropical rain forest region. Biotropica 2014, 46, 378–386. [Google Scholar] [CrossRef]
  38. Sundarapandian, S.M.; Swamy, P.S. Litter production and leaf-litter decomposition of selected tree species in tropical forests at Kodayar in the Western Ghats, India. For. Ecol. Manag. 1999, 123, 231–244. [Google Scholar] [CrossRef]
  39. Bhatti, J.S.; Jassal, R.S. Long term aboveground litterfall production in boreal jack pine (Pinus banksiana) and black spruce (Picea mariana) stands along the Boreal Forest Transect Case Study in western central Canada. Écoscience 2014, 21, 301–314. [Google Scholar] [CrossRef]
  40. Takahashi, M.; Hirai, K.; Limtong, P.; Leaungvutivirog, C.; Panuthai, S.; Suksawang, S.; Anusontpornperm, S.; Marod, D. Topographic variation in heterotrophic and autotrophic soil respiration in a tropical seasonal forest in Thailand. J. Soil Sci. Plant Nutr. 2011, 57, 452–465. [Google Scholar] [CrossRef]
  41. Marod, D.; Kutintara, U.; Yarwudhi, C.; Tanaka, H.; Nakashisuka, T. Structural dynamics of a natural mixed deciduous forest in western Thailand. J. Veg. Sci. 1999, 10, 777–786. [Google Scholar] [CrossRef]
  42. Marod, D.; Kutintara, U.; Tanaka, H.; Nakashizuka, T. The effects of drought and fire on seed and seedling dynamics in a tropical seasonal forest in Thailand. Plant Ecol. 2002, 161, 41–57. [Google Scholar] [CrossRef]
  43. Proctor, J.; Anderson, J.M.; Chai, P.; Vallack, H.W. Ecological studies in four contrasting lowland rain forests in Gunung Mulu National Park, Sarawak: I. Forest environment, structure and floristics. J. Ecol. 1983, 71, 237–260. [Google Scholar] [CrossRef]
  44. Metcalfe, D.; Meir, P.; Aragao, L.E.O.C.; da Costa, A.; Almeida, S.; Braga, A.; Gonçalves, P.; Athaydes, J.; Malhi, Y.; Williams, M. Sample sizes for estimating key ecosystem characteristics in a tropical terra firme rainforest. For. Ecol. Manag. 2008, 255, 558–566. [Google Scholar] [CrossRef]
  45. Burnham, K.P.; Anderson, D.R. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach; Springer: Berlin/Heidelberg, Germany, 2002; pp. 149–205. [Google Scholar]
  46. Guisan, A.; Edwards Jr, T.C.; Hastie, T. Generalized linear and generalized additive models in studies of species distributions: Setting the scene. Ecol. Model. 2002, 157, 89–100. [Google Scholar] [CrossRef]
  47. R Core Team. R: A Language and Environment for Statistical Computing. 2020. Available online: http://www.R-project.org/ (accessed on 7 March 2022).
  48. Wood, S.; Wood, M.S. Package ‘mgcv’. In R Package Version 1.29; R Foundation for Statistical Computing: Vienna, Austria, 2015. [Google Scholar]
  49. Wang, C.G.; Zheng, X.B.; Wang, A.Z.; Dai, G.H.; Zhu, B.K.; Zhao, Y.M.; Dong, S.J.; Zu, W.Z.; Wang, W.; Zheng, Y.G. Temperature and Precipitation Diversely Control Seasonal and Annual Dynamics of Litterfall in a Temperate Mixed Mature Forest, Revealed by Long-Term Data Analysis. J. Geophys. Res. Biogeosci. 2021, 126, e2020JG006204. [Google Scholar] [CrossRef]
  50. Bray, J.R.; Gorham, E. Litter production in forests of the world. In Advances in Ecological Research; Elsevier: Amsterdam, The Netherlands, 1964; Volume 2, pp. 101–157. [Google Scholar]
  51. Ahirwal, J.; Saha, P.; Nath, A.; Nath, A.J.; Deb, S.; Sahoo, U.K. Forests litter dynamics and environmental patterns in the Indian Himalayan region. For. Ecol. Manag. 2021, 499, 119612. [Google Scholar] [CrossRef]
  52. Hanpattanakit, P.; Chidthaisong, A. Litter production and decomposition in dry dipterocarp forest and their responses to climatic factors. GMSARN Int. J. 2012, 6, 169–174. [Google Scholar]
  53. Martínez-Yrízar, A.; Sarukhán, J. Litterfall patterns in a tropical deciduous forest in Mexico over a five-year period. J. Trop. Ecol. 1990, 6, 433–444. [Google Scholar] [CrossRef]
  54. Whigham, D.F.; Zugasty Towle, P.; Cabrera Cano, E.F.; O’Neill, J.P.; Ley, E. Effect of annual variation in precipitation on growth and litter production in a tropical dry forest in the Yucatan of Mexico. Trop. Ecol. 1990, 31, 23–34. [Google Scholar]
  55. Köhler, L.; Hölscher, D.; Leuschner, C. High litterfall in old-growth and secondary upper montane forest of Costa Rica. Plant Ecol. 2008, 199, 163–173. [Google Scholar] [CrossRef]
  56. Swamy, S.L.; Dutt, C.B.S.; Murthy, M.S.R.; Mishra, A.; Bargali, S.S. Floristics and dry matter dynamics of tropical wet evergreen forests of Western Ghats, India. Curr. Sci. 2010, 99, 353–364. [Google Scholar]
  57. Tang, J.-W.; Cao, M.; Zhang, J.-H.; Li, M.-H. Litterfall production, decomposition and nutrient use efficiency varies with tropical forest types in Xishuangbanna, SW China: A 10-year study. Plant Soil 2010, 335, 271–288. [Google Scholar] [CrossRef]
  58. Park, B.B.; Rahman, A.; Han, S.H.; Youn, W.B.; Hyun, H.J.; Hernandez, J.; An, J.Y. Carbon and nutrient inputs by litterfall in evergreen and deciduous forests in Korea. Forests 2020, 11, 143. [Google Scholar] [CrossRef]
  59. Givnish, T.J. Adaptive significance of evergreen vs. deciduous leaves: Solving the triple paradox. Silva Fenn. 2002, 36, 703–743. [Google Scholar] [CrossRef]
  60. Borchert, R.; Meyer, S.A.; Felger, R.S.; Porter-Bolland, L. Environmental control of flowering periodicity in Costa Rican and Mexican tropical dry forests. Glob. Ecol. Biogeogr. 2004, 13, 409–425. [Google Scholar] [CrossRef]
  61. Nanda, A.; Suresh, H.S.; Krishnamurthy, Y.L. Phenology of a tropical dry deciduous forest of Bhadra wildlife sanctuary, southern India. Ecol. Process. 2014, 3, 1. [Google Scholar] [CrossRef]
  62. Yang, X.; Wu, J.; Chen, X.; Ciais, P.; Maignan, F.; Yuan, W.; Piao, S.; Yang, S.; Gong, F.; Su, Y. A comprehensive framework for seasonal controls of leaf abscission and productivity in evergreen broadleaved tropical and subtropical forests. Innovation 2021, 2, 1–7. [Google Scholar] [CrossRef]
  63. Liu, C.; Westman, C.J.; Berg, B.; Kutsch, W.; Wang, G.Z.; Man, R.; Ilvesniemi, H. Variation in litterfall-climate relationships between coniferous and broadleaf forests in Eurasia. Glob. Ecol. Biogeogr. 2004, 13, 105–114. [Google Scholar] [CrossRef]
  64. Cleverly, J.; Eamus, D.; Coupe, N.R.; Chen, C.; Maes, W.; Li, L.; Faux, R.; Santini, N.S.; Rumman, R.; Yu, Q. Soil moisture controls on phenology and productivity in a semi-arid critical zone. Sci. Total Environ. 2016, 568, 1227–1237. [Google Scholar] [CrossRef]
  65. Yang, Y.S.; Guo, J.F.; Chen, G.S.; Xie, J.S.; Gao, R.; Li, Z.; Jin, Z. Litter production, seasonal pattern and nutrient return in seven natural forests compared with a plantation in southern China. Forestry 2005, 78, 403–415. [Google Scholar] [CrossRef]
  66. De Queiroz, M.G.; da Silva, T.G.F.; Zolnier, S.; de Souza, C.A.A.; de Souza, L.S.B.; Neto, A.J.S.; de Araújo, G.G.L.; Ferreira, W.P.M. Seasonal patterns of deposition litterfall in a seasonal dry tropical forest. Agric. For. Meteorol. 2019, 279, 107712. [Google Scholar] [CrossRef]
  67. Lin, K.-C.; Hamburg, S.P.; Tang, S.-l.; Hsia, Y.-J.; Lin, T.-C. Typhoon effects on litterfall in a subtropical forest. Can. J. For. Res. 2003, 33, 2184–2192. [Google Scholar] [CrossRef]
  68. An, J.Y.; Han, S.H.; Youn, W.B.; Lee, S.I.; Rahman, A.; Dao, H.T.T.; Seo, J.M.; Aung, A.; Choi, H.-S.; Hyun, H.J. Comparison of litterfall production in three forest types in Jeju Island, South Korea. J. For. Res. 2020, 31, 945–952. [Google Scholar] [CrossRef]
  69. Song, Y.-J.; Tian, W.-B.; Liu, X.-Y.; Yin, F.; Cheng, J.-Y.; Zhu, D.-N.; Ali, A.; Yan, E.-R. Associations between litterfall dynamics and micro-climate in forests of Putuoshan Island, Zhejiang, China. Chin. J. Plant Ecol. 2016, 40, 1154. [Google Scholar]
  70. Bou, J.; Caritat, A.; Vilar, L. Litterfall and growth dynamics relationship with the meteorological variability in three forests in the Montseny natural park. Folia For. Pol. Ser. A For. 2015, 57, 145–159. [Google Scholar] [CrossRef]
  71. Brearley, F.Q.; Proctor, J.; Suriantata; Nagy, L.; Dalrymple, G.; Voysey, B.C. Reproductive phenology over a 10-year period in a lowland evergreen rain forest of central Borneo. J. Ecol. 2007, 95, 828–839. [Google Scholar] [CrossRef]
  72. Curran, L.M.; Caniago, I.; Paoli, G.D.; Astianti, D.; Kusneti, M.; Leighton, M.; Nirarita, C.E.; Haeruman, H. Impact of El Nino and logging on canopy tree recruitment in Borneo. Science 1999, 286, 2184–2188. [Google Scholar] [CrossRef]
  73. Wright, S.J.; Carrasco, C.; Calderon, O.; Paton, S. The El Niño Southern Oscillation, variable fruit production, and famine in a tropical forest. Ecology 1999, 80, 1632–1647. [Google Scholar] [CrossRef]
  74. Ashton, P.S.; Givnish, T.J.; Appanah, S. Staggered flowering in the Dipterocarpaceae: New insights into floral induction and the evolution of mast fruiting in the aseasonal tropics. Am. Nat. 1988, 132, 44–66. [Google Scholar] [CrossRef]
  75. Van Schaik, C.P.; Terborgh, J.W.; Wright, S.J. The phenology of tropical forests: Adaptive significance and the consequences for primary consumers. Annu. Rev. Ecol. Evol. Syst. 1993, 24, 353–377. [Google Scholar] [CrossRef]
Forests 14 02107 g001
Figure 1. Trends in maximum and mean temperature (a,b) and evaporation (c) during 1992–2021 at MKWRS. Dots represent the monthly value of each variable.
Figure 1. Trends in maximum and mean temperature (a,b) and evaporation (c) during 1992–2021 at MKWRS. Dots represent the monthly value of each variable.
Forests 14 02107 g001
Forests 14 02107 g002
Figure 2. Annual litterfall production (bars) and rainfall (broken line) during 1993–2021 at MKWRS, western Thailand.
Figure 2. Annual litterfall production (bars) and rainfall (broken line) during 1993–2021 at MKWRS, western Thailand.
Forests 14 02107 g002
Forests 14 02107 g003
Figure 3. Variations in the four litter components (leaf, flower, fruit, and others measured in Mg ha−1) during 1993–2021 in MKWRS, western Thailand.
Figure 3. Variations in the four litter components (leaf, flower, fruit, and others measured in Mg ha−1) during 1993–2021 in MKWRS, western Thailand.
Forests 14 02107 g003
Forests 14 02107 g004
Figure 4. (a) Total annual and (b) mean monthly variations in the relative contribution to the litterfall pool by leaf, flower, fruit, and other/miscellaneous categories during 1993–2021 in TDF at MKWRS.
Figure 4. (a) Total annual and (b) mean monthly variations in the relative contribution to the litterfall pool by leaf, flower, fruit, and other/miscellaneous categories during 1993–2021 in TDF at MKWRS.
Forests 14 02107 g004
Forests 14 02107 g005
Figure 5. Mean monthly litterfall components: (a) leaf litter, (b) flower litter, and (c) fruit litter, for three dominant species (evergreen D. alatus, Dipala, and two deciduous species, P. macrocarpus, Ptema, and P. siamensis, Pansia).
Figure 5. Mean monthly litterfall components: (a) leaf litter, (b) flower litter, and (c) fruit litter, for three dominant species (evergreen D. alatus, Dipala, and two deciduous species, P. macrocarpus, Ptema, and P. siamensis, Pansia).
Forests 14 02107 g005
Forests 14 02107 g006
Figure 6. Estimated lag effect variations in the most significant environmental variables on leaf litterfall: (a) SM, (b) ET, and (c) Tmax. The inset indicates the effect of an incremental change in the respective environmental variable on either increasing or decreasing the chances of leaf litterfall. Incremental changes in 0.1, 10 mm, and 1 °C were used for SM, ET, and Tmax, respectively. The lagged variables plotted in gray on the larger graphs indicate nonsignificant changes in litterfall as determined by odds ratios in the inset graph.
Figure 6. Estimated lag effect variations in the most significant environmental variables on leaf litterfall: (a) SM, (b) ET, and (c) Tmax. The inset indicates the effect of an incremental change in the respective environmental variable on either increasing or decreasing the chances of leaf litterfall. Incremental changes in 0.1, 10 mm, and 1 °C were used for SM, ET, and Tmax, respectively. The lagged variables plotted in gray on the larger graphs indicate nonsignificant changes in litterfall as determined by odds ratios in the inset graph.
Forests 14 02107 g006
Forests 14 02107 g007
Figure 7. Estimated lag effect variations among statistically significant environmental effects on flower litterfall: (a) ET, (b) VPD, and (c) light. Insets indicate the effects of an incremental change in an environmental variable on the probability of flower litterfall. Incremental changes of 10 mm, 0.1 kPa, and 1 MJ m−2 day−1 mm were used for ET, VPD, and light, respectively. The lagged variables plotted in gray (in the main graph) indicate a nonsignificant change in litterfall, as determined through the odds ratios in the inset graph.
Figure 7. Estimated lag effect variations among statistically significant environmental effects on flower litterfall: (a) ET, (b) VPD, and (c) light. Insets indicate the effects of an incremental change in an environmental variable on the probability of flower litterfall. Incremental changes of 10 mm, 0.1 kPa, and 1 MJ m−2 day−1 mm were used for ET, VPD, and light, respectively. The lagged variables plotted in gray (in the main graph) indicate a nonsignificant change in litterfall, as determined through the odds ratios in the inset graph.
Forests 14 02107 g007
Forests 14 02107 g008
Figure 8. Estimated lag effects of the most significant environmental influences on fruit litterfall: (a) SOI, (b) Tmean, and (c) light. Insets indicate the effects of an incremental change in a given environmental variable on either increasing or decreasing the chances in fruit litterfall. Incremental changes of one unit, 1 °C, and 1 MJ m−2 day−1 mm were used for SOI, Tmean, and light, respectively. Lagged variables plotted in gray (in the main graph) indicate nonsignificant changes in litterfall, as determined through the odds ratios in the inset graph.
Figure 8. Estimated lag effects of the most significant environmental influences on fruit litterfall: (a) SOI, (b) Tmean, and (c) light. Insets indicate the effects of an incremental change in a given environmental variable on either increasing or decreasing the chances in fruit litterfall. Incremental changes of one unit, 1 °C, and 1 MJ m−2 day−1 mm were used for SOI, Tmean, and light, respectively. Lagged variables plotted in gray (in the main graph) indicate nonsignificant changes in litterfall, as determined through the odds ratios in the inset graph.
Forests 14 02107 g008
Forests 14 02107 g009
Figure 9. (a) SOI (black) and fruit litterfall (green) during the study period, and (b) fruit litterfall of D. alatus (red), P. siamensis (blue), and P. macrocarpus (green). The La Niña (>+8 or wet) and El Niño (<−8 or dry) events are shaded in blue and red, respectively.
Figure 9. (a) SOI (black) and fruit litterfall (green) during the study period, and (b) fruit litterfall of D. alatus (red), P. siamensis (blue), and P. macrocarpus (green). The La Niña (>+8 or wet) and El Niño (<−8 or dry) events are shaded in blue and red, respectively.
Forests 14 02107 g009
Table 1. Results of GAM fits to litter components. The model structure includes the most significant variables indicated by the model, deviance, and R2 values.
Table 1. Results of GAM fits to litter components. The model structure includes the most significant variables indicated by the model, deviance, and R2 values.
Litter FractionModel StructureDeviance Explained [%]R2 [%]Variable Significance
LeafLeaf.litter~f(Year) + f(Month) +f(Tmax) + f(SM) + f(ET)83.682.1Year [p < 0.0001 ***]
Month [p < 0.0001 ***]
Tmax [p = 0.0293 *]
SM [p < 0.0001 ***]
ET [p = 0.0013 **]
FlowerFlower.litter ~ f(Year) + f(Month) +f(VPD) + f(Light) + f(ET)67.265Year [p < 0.0001 ***]
Month [p = 0.03486 *]
VPD [p = 0.00284 **]
Light [p = 0.02393 *]
ET [p < 0.0001 ***]
FruitFruit.litter ~ f(Year) + f(Month) +f(Light) + f(Tmean) + f(SOI)67.264.9Year [p < 0.0001 ***]
Month [p < 0.0001 ***]
Light [p = 0.001668 **]
Tmean [p < 0.0001 ***]
SOI [p < 0.0001 ***]
Significant codes: *** < 0.001; ** < 0.01; * < 0.05.
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

Marod, D.; Nakashizuka, T.; Saitoh, T.; Hirai, K.; Thinkampheang, S.; Asanok, L.; Phumphuang, W.; Danrad, N.; Pattanakiat, S. Long Term Seasonal Variability on Litterfall in Tropical Dry Forests, Western Thailand. Forests 2023, 14, 2107. https://doi.org/10.3390/f14102107

AMA Style

Marod D, Nakashizuka T, Saitoh T, Hirai K, Thinkampheang S, Asanok L, Phumphuang W, Danrad N, Pattanakiat S. Long Term Seasonal Variability on Litterfall in Tropical Dry Forests, Western Thailand. Forests. 2023; 14(10):2107. https://doi.org/10.3390/f14102107

Chicago/Turabian Style

Marod, Dokrak, Tohru Nakashizuka, Tomoyuki Saitoh, Keizo Hirai, Sathid Thinkampheang, Lamthai Asanok, Wongsatorn Phumphuang, Noppakun Danrad, and Sura Pattanakiat. 2023. "Long Term Seasonal Variability on Litterfall in Tropical Dry Forests, Western Thailand" Forests 14, no. 10: 2107. https://doi.org/10.3390/f14102107

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