Study on Decomposition Characteristics of Early Spring Ephemeral Plant Litter in Various Forest Types

Abstract

In terrestrial ecosystems, the decomposition of early spring ephemeral plant litter (ESPL for short) is one of the important processes in the carbon and nutrient cycles during the early spring stage. The current study focused on four typical spring ephemeral plant species in three forest types of Northeast China and investigated the decomposition characteristics of herb litters, using litterbag decomposition experiments. The study results indicate that the mass loss rate of ESPL decomposition after 50 days can be as high as 73.15% to 80.44%. Throughout the entire decomposition period, there is a significant correlation between the decomposition of ESPL and time, with initial decomposition relatively fast and later decomposition slowing down. Overall, Hylomecon japonicum exhibits slightly faster decomposition, and Cardamine leucantha decomposes relatively slowly, while Cardamine leucantha shows the highest mass loss rate in the first 10 days, reaching 38.71%. The mass loss rates of the four types of ESPL are significantly correlated with the litter nutrient conditions, as are the stage-specific mass loss rates. Furthermore, there are distinct differences in the nutrient composition affecting the decomposition of different types of ESPL. Across different forest stands, influenced by different decomposition environments, such as soil conditions, the decomposition of ESPL is fastest in the deciduous broad-leaved forest, with decomposition reaching 50% and 95% in only 15–18 days and 63–88 days, respectively. In the broad-leaved forest, it takes 18–23 days and 78–110 days, while, in the birch forest, it takes 22–32 days and 99–136 days.

1. Introduction

The decomposition of litter is a critical process in the material cycling and energy flow of forest ecosystems. It is one of the important ways for effective nutrient return to the soil, directly impacting the level of forest primary productivity, soil fertility maintenance, and even the sustainable development of ecosystems. Litter decomposition is a significant ecological process involving a series of physical, chemical, and biological processes. In the early stages of decomposition, the high levels of nutrients such as nitrogen, phosphorus, and sulfur stimulate the leaching and breakdown of water-soluble substances, non-woody fibers, and hemicellulose. In the later stages of decomposition, the preservation of lignin material leads to a predominant decomposition of lignin mass loss [1,2,3]. This process effectively converts organic matter into stable forms, contributing to soil fertility, long-term carbon storage, and nutrient cycling [1].

For a long time, climate, litter quality, and decomposer communities have been considered the main driving factors of litter decomposition [4]. However, the impact of these primary drivers on litter decomposition varies across different ecosystems [5,6], and this variation is evident at different stages of decomposition [7]. Globally, the composition of litter types is a direct and crucial factor influencing litter decomposition [8,9,10]. Approximately 65% of the initial variability in litter decomposition can be explained by litter type, and 13% is related to the biological community. In wet environments, rainwater filters soluble compounds from the litter [3], supporting soil microbial and animal activity in further decomposing the litter [11]; in contrast, in dry or arid grasslands, water limits the transport of compounds and the activity of decomposers.

During the brief period between snowmelt and canopy leaf-out in early spring (from March to April), there is minimal above-ground biological activity. Nutrient uptake by early spring plants is similar to the loss within the ecosystem. When short-lived plants die and decompose after the canopy closes, they provide nutrients for tree growth. This plays a crucial role in bridging the material and energy flow between winter and spring in forest ecosystems, contributing to the vigorous activity of ecosystem recovery in spring. Therefore, early spring plants are described as “Vernal Dam” retaining nutrients that could potentially be lost from the ecosystem. This phenomenon is especially important in temperate ecosystems, where spring ephemerals act as nutrient reservoirs. Previous studies have documented how these plants sequester nutrients, such as nitrogen and phosphorus, in the spring before canopy closure and the onset of more active plant growth. For example, research in temperate forests has shown that spring ephemeral herbs, such as Claytonia virginica, can accumulate significant amounts of nutrients during their brief growing season, preventing nutrient loss through leaching or erosion [12]. This temporary sequestration of nutrients is later returned to the soil upon decomposition, providing a crucial resource for surrounding vegetation and enhancing nutrient cycling in the ecosystem [13,14]. This study aimed to analyze the impact of the soil environment, quality of litter substrate, and forest type on the decomposition of litter from various early spring plants. It is hypothesized that the decomposition rate of spring ephemeral plant detritus is closely related to plant species, forest type, and their nutrient composition, with the decomposition process exhibiting distinct stage-specific patterns. To investigate the decomposition patterns of typical spring ephemeral plants, the current study focused on four typical spring ephemeral plant species across three forest types in Northeast China. Through litterbag decomposition experiments, the decomposition characteristics of herbaceous litter were examined, along with their relationship to plant species, forest type, and nutrient composition. The study results will provide crucial scientific evidence to guide the regulation of carbon and nutrient storage and release during the early spring period in northern temperate forests.

2. Materials and Methods

2.1. Study Area

The study area is located in the State-owned Forest Protection Center of Jilin Forestry Experimental Area (127°44′~127°44′ E, 43°57′~43°58′ N) in Jilin Province. The region has a typical temperate continental climate, with an average annual temperature of 3.8 °C and an annual rainfall of 700 to 800 mm. The elevation ranges from 459 to 517 m, and the forest coverage rate is 88.4%. The dominant forest types include broad-leaved Korean pine forests, as well as mixed forests of Quercus mongolica, Populus davidiana, Betula platyphylla, Fraxinus mandshurica, and Juglans mandshurica. The soil types mainly consist of dark brown soil, cinnamon soil, meadow soil, and marsh soil, with dark brown soil as the most widely distributed [15,16]. To compare and analyze the impact of different forest types on the decomposition of ESPL, this study selected birch forests dominated by Betula platyphylla with a small amount of Quercus mongolica, deciduous broad-leaved forests dominated by Tilia amurensis, Fraxinus mandschurica, and Acer mandshuricum, and broad-leaved Korean pine forests dominated by Pinus koraiensis, Fraxinus mandschurica, and Acer mandshuricum (the straight-line distance between the three forest types is less than 2 km). The elevations of the three forest types are slightly higher for the broad-leaved Korean pine forest (550 m) and slightly lower for the mixed broad-leaved forest (480 m) and birch forest (410 m). The terrain is gently sloping, with slopes of less than 5 degrees, and the aspect is sunny for the birch forest, and semi-shady for the mixed broad-leaved forest and the broad-leaved Korean pine forest. The canopy closure of the birch forest and the mixed broad-leaved forest is 0.75, slightly higher at 0.80 for the broad-leaved Korean pine forest.

2.2. Study Design and Data Collection

Based on the life history characteristics of early spring plants, the aboveground litter of early spring plants was collected at the end of May 2020. This study selected four early spring plants widely distributed in the temperate forests of Northern China, including Anemone raddeana (A. raddeana), Corydalis repens (C. repens), Hylomecon japonicum (H. japonicum), and Cardamine leucantha (C. leucantha). The first three are typical short-lived early spring plants, while C. leucantha is an early spring flowering plant. Its aboveground parts did not wither during the sampling period, but we still synchronously collected its aboveground parts for comparative research. Decomposition experiments were conducted using the litterbag method in birch forests, deciduous broad-leaved forests, and broad-leaved Korean pine forests to investigate the decomposition patterns of litter from four early spring plants over time. These samples were brought back to the laboratory, cleaned of impurities, and naturally air-dried. Approximately 5 g of litter from each plant species was then placed into decomposition bags with a pore size of 0.5 mm. Additionally, some samples were retained and dried in an oven (Shanghai Xinmiao Medical Equipment Manufacturing Co., Ltd., Shanghai, China) at 65 °C until a constant mass was achieved. The moisture content of the naturally air-dried samples was determined, and a portion of the samples was also retained to measure their initial nutrient content. The determination of soil organic carbon or plant total carbon organic carbon (C) was conducted using the potassium dichromate-sulfuric acid heating method [17], while total contents of nitrogen (TN), phosphorus (TP), and potassium (TK) were obtained by digesting with sulfuric acid (Xinsheng Chemical Co., Ltd., Shanghai, China) to obtain the test solution. The determination of N was carried out using the Kjeldahl method, P was determined using a spectrophotometer (Shimadzu Corporation Management (China) Co., Ltd., Shanghai, China) colorimetric method, and K was determined using a flame photometer (Shimadzu Corporation Management (China) Co., Ltd., Shanghai, China) colorimetric method [18]. The determination of soil hydrolysable nitrogen (HN), available phosphorus (AP), and available potassium (AK) was conducted using the chemical reagent leaching method [19,20,21].

The litterbag decomposition experiment was completed on 1 June. Specifically, in representative areas of birch forest, mixed broad-leaved forest, and broad-leaved Korean pine forest, 3 plots of 10 m × 10 m were set up for each forest type, with a distance of over 10 m between each plot. In each experiment area, 4 types of early spring plants were randomly placed on the surface of the litter (simulating the natural wilting and attachment of aboveground parts of early spring plants to the litter). Five bags of the same type of litter were placed in each row, with a spacing of 2 m between rows and bags. To prevent the decomposition bags from curling up or being blown by the wind, small tree branches were used to secure the four corners of the decomposition bags. Self-recording thermometers were used to continuously monitor the temperature changes in the surface soil (10 cm) of each experiment area. At the same time, mixed soil samples from the surface of each experiment area of the same forest type were collected for chemical property analysis. Starting from the day of experiment, samples were taken every 10 days. One decomposition bag for each of the 4 types of early spring plants was collected from each experiment area of each forest type, totaling 5 sampling events. The retrieved decomposition bags were carefully cleaned of impurities and placed in a 65 °C drying oven (Shanghai Xinmiao Medical Equipment Manufacturing Co., Ltd., Shanghai, China) until a constant mass was achieved. The bags were then weighed to calculate their mass loss rate, residue rate, decomposition rate, and simultaneously measure the nutrient element content of each sample.

2.3. Statistical Analysis

Statistical analysis was conducted using SPSS (IBM^® SPSS Statistics 26.0.0), including one-way analysis of variance (ANOVA) to compare the means of different groups, followed by least significant difference (LSD) tests for pairwise comparisons. Pearson two-tailed tests were employed to analyze the correlation between the mass loss rate of early spring plants and decomposition time as well as plant internal constituents. Additionally, a linear regression model was used to fit the dynamic equation for the decomposition of early spring plants. The significance level (α) for all tests was set at 0.05, corresponding to a 95% confidence level. Prior to conducting the ANOVA and correlation analyses, we checked the assumptions of normality and homoscedasticity. Normality of the data was assessed visually using histograms and Q-Q plots and formally tested using the Shapiro–Wilk test. For homoscedasticity, we used the Breusch–Pagan test to assess whether the variance across groups was consistent. If any significant departures from normality or homoscedasticity were detected, we applied appropriate data transformations (e.g., log or square root transformations) as necessary. All graphs were created using SigmaPlot (Systat Software, Inc. 10.0.0.54, San Jose, CA, USA). Relevant indicators and calculation formulas are as follows:

$$\mathrm{ML}=\left[\right({\mathrm{M}}_0-{\mathrm{M}}_{\mathrm{t}})/{\mathrm{M}}_0]\times 100\%$$

(1)

$${\mathrm{ML}}_{10}=\left[\right({\mathrm{M}}_{\mathrm{t}-1}-{\mathrm{M}}_{\mathrm{t}})/{\mathrm{M}}_{\mathrm{t}-1}]\times 100\%$$

(2)

$$\mathrm{MR}=({\mathrm{M}}_{\mathrm{t}}/{\mathrm{M}}_0)\times 100\%$$

(3)

$${\mathrm{M}}_{\mathrm{t}}/{\mathrm{M}}_0=\mathrm{a}{\mathrm{e}}^{-\mathrm{kt}}$$

(4)

where ML is mass lost rate; ML₁₀ is 10-day mass loss rate; MR is mass remaining rate; M₀ is the initial dry mass of the litter; M_t is the dry mass of the litter at time t; a is coefficient; k is the decomposition coefficient; and t is the decomposition time. M_t/M₀ is an exponential decay model used to describe and predict the dynamics of litter decomposition [22], as well as the time required for 50% decomposition (t_0.05) and 95% decomposition (t_0.95) of the litter.

3. Results

3.1. Soil

In the three forest types selected for this study, the broad-leaved Korean pine forest represents a relatively undisturbed primary forest and is the top community in this region. The birch forest and mixed broad-leaved forest, on the other hand, are natural forests formed through secondary succession after selective logging and partial cutting disturbances. The soil base of these three forest types exhibits significant similarities, with the key difference lying in the impact of logging disturbances and vegetation recovery on the soil. Currently, the birch forest shows relatively lower pH, available phosphorus (AP), and total content of potassium (TK) (

Table 1. Topsoil chemistry of three stand types.

Forest Type	pH	C (g/kg)	HN (µg/kg)	AP (µg/kg)	AK (µg/kg)	TN (g/kg)	TP (g/kg)	TK (g/kg)
BF	4.47	459.42	1553.84	2.57	350.29	16.58	4.17	1.33
DBF	5.31	332.26	261.87	10.30	344.58	10.70	1.15	11.83
BKF	5.99	103.56	954.21	24.08	354.91	3.89	4.25	19.62

Note: BF is birch forest; DBF is deciduous broad-leaved forest; BKF is broad-leaved Korean pine forest; HN is hydrolysable nitrogen; AP is available phosphorus; AK is available potassium; TN is total nitrogen content; TP is total phosphorus content; and TK is total potassium content.

), while the mixed broad-leaved forest exhibits lower hydrolysable nitrogen (HN) and the total content of phosphorus (TP). The broad-leaved Korean pine forest, on the other hand, has relatively lower carbon (C) and TN content but higher HN, TP, and TK content.

Continuous measurements of soil surface temperatures over a 50-day decomposition period showed a gradual increase in soil temperatures across all forest types with time (

Table 2. Surface soil temperature of three stand types.

Time (d)	Mean Temperature ± SD °C
	Birch Forest	Deciduous Broad-Leaved Forest	Broad-Leaved Korean Pine Forest
0–10	11.68 ± 1.09	11.52 ± 1.03	10.79 ± 1.12
10–20	13.33 ± 0.38	13.24 ± 0.34	12.24 ± 0.42
20–30	14.37 ± 0.49	14.57 ± 0.61	13.98 ± 0.58
30–40	15.89 ± 0.47	16.01 ± 0.38	15.03 ± 0.39
40–50	17.51 ± 0.88	17.83 ± 1.13	17.11 ± 1.29
Mean	14.56 a	14.62 a	13.82 a

Note: The same lowercase letters indicate no significant differences in the substance content between species (p > 0.05).

). However, the average temperatures did not exhibit significant differences among the different forest types, except for a slightly lower temperature in the relatively closed canopy broad-leaved Korean pine forest compared to the slightly higher temperature in the mixed broad-leaved forest, with a difference of 0.8 °C.

3.2. Litter Nutrients

In the initial nutrient content of the four ESPL (

Table 3. Initial nutrient content and ratio of 4 ESPL.

Species	C (g/kg)	N (g/kg)	P (g/kg)	K (g/kg)	C/N	C/P
C. leucantha	456.39 a	15.20 a	0.72 c	190.38 b	30.02 b	635.13 a
C. repens	446.90 a	15.56 a	3.20 b	199.22 b	28.72 b	139.68 b
H. japonicum	436.30 a	16.34 a	3.05 b	347.12 a	26.70 b	143.26 b
A. raddeana	447.12 a	10.46 b	4.34 a	159.94 c	42.74 a	103.05 b

Note: Different lowercase letters indicate significant differences in the substance content between species (p < 0.05), while the same lowercase letters indicate no significant differences in the substance content between species (p > 0.05).

), the differences in carbon (C) content are not significant, ranging from 436.30 to 456.39 g/kg. The nitrogen (N) and potassium (K) content are highest in H. japonicum and lowest in A. raddeana. The phosphorus (P) content is notably lower in C. leucantha compared to the other three, being only 16.59% of the highest content found in A. raddeana. The C/N ratio is only higher than 40 in A. raddeana. Due to the lowest P content in A. raddeana, its C/P ratio is significantly higher, being four to six times that of the other three. Overall, H. japonicum has relatively low C content and relatively high nutrient element content, followed by C. repens and A. raddeana, while C. leucantha has relatively high C content and the lowest nutrient element content, especially in terms of P.

3.3. Decomposition of ESPL

3.3.1. Mass Loss

After 50 days of decomposition, the mass loss rates of the four early spring plants at each experiment site ranged from 73.15% to 80.44%, with H. japonicum showing the fastest rate and C. leucantha the slowest (Figure 1). Among them, C. leucantha exhibited the highest mass loss rate at 10 days of decomposition (38.71%), approaching 40%. The rate significantly slowed down by 20 days of decomposition, with only a relative 3% stage-wise mass loss. By 30 days of decomposition, the rate accelerated again (15.21%), resulting in an overall loss of nearly 60%. In the subsequent 20 days, the decomposition rate slowed down again, with stage-wise mass loss rates of around 8% at 10 days. A. raddeana also showed the highest mass loss rate in the initial 10 days of decomposition (34.07%), significantly slowing down by 20 days, accelerating at 30 days, and gradually slowing down again in the subsequent 20 days. H. japonicum and C. repens both exhibited relatively fast mass loss rates in the initial 10 and 20 days, with stage-wise loss rates of about 30%, gradually slowing down in the subsequent decomposition, reaching around 3% by 50 days.

Overall, the ML of the four early spring plants did not show a significant difference (F = 1.95, Sig. = 0.138), indicating that mass loss is not related to the type of litter. However, the impact of decomposition time was extremely significant (F = 58.97, Sig. < 0.001). Although the interaction between litter type and decomposition time did not significantly affect the ML, the cumulative ML at 40 d and 50 d of decomposition were significantly higher for H. japonicum compared to C. leucantha (Figure 1a). This difference was attributed to the significant variation in stage-wise decomposition between the two during the first 20 days (Figure 1b). While C. repens also exhibited significantly higher ML₁₀ compared to C. leucantha during this period, its relative mass loss rates were slightly lower at 40 d and 50 d of decomposition. This was also evident in the post hoc LSD comparison of mass loss rates and decomposition time differences, where the variation in litter decomposition with time was only significant overall within 30 days, with no significant differences in mass loss rates between 30 d, 40 d, and 50 d of decomposition.

To further investigate the relationship between the ML, ML₁₀, decomposition time, and the nutrient content of various early spring plants, we conducted a correlation analysis by aligning the mass loss rate and 10-day mass loss rate with their respective initial content before decomposition. This involved examining the correspondence of the 10-day decomposition mass loss rate, stage mass loss rate, and initial content, as well as the correspondence of the 20-day decomposition mass loss rate and stage mass loss rate with the content after 10 days of decomposition, and so on. The results are presented in

Table 4. Analysis of ML, ML₁₀, decomposition time, and their correlation with the content of internal substances.

Source	Species	Time	Mass Lost	C	N	P	K	C/N	C/P
Mass lost rate	C. leucantha	0.823 **	1	−0.623 *	0.091	0.481	−0.618 *	−0.589 *	−0.570 *
	C. repens	0.838 **	1	−0.270	0.723 **	0.788 **	−0.579 *	−0.661 **	−0.775 **
	H. japonicum	0.886 **	1	0.616 *	0.770 **	0.586 *	−0.717 **	−0.690 **	−0.321
	A. raddeana	0.933 **	1	−0.480	0.849 **	−0.331	−0.637 **	−0.838 **	0.118
10 d Mass lost rate	C. leucantha	−0.604 *	−0.324	0.716 **	0.457	−0.396	0.253	0.146	0.667 **
	C. repens	−0.876 **	−0.659 **	0.264	−0.454	−0.888 **	0.807 **	0.483	0.860 **
	H. japonicum	−0.823 **	−0.610 *	−0.452	−0.611 *	−0.328	0.815 **	0.544 *	0.007
	A. raddeana	−0.780 **	−0.656 **	0.387	−0.796 **	0.471	0.392	0.802 **	−0.150

Note: Pearson Correlation Sig. (2-tailed) N = 15, ** means p < 0.01, extremely significant; * means p < 0.05, significant.

.

According to Table 4, the ML of the four early spring plants are all significantly positively correlated with decomposition time. However, their ML₁₀ is strongly negatively correlated with decomposition time, indicating that, as decomposition time increases, ML decreases. The slower rate of ML₁₀ is related to the remaining amount of litter; except for C. leucantha, ML₁₀ is strongly negatively correlated with ML. Additionally, it is also related to the content of internal substances in the litter. Specifically, the ML of the four early spring plants is strongly negatively correlated with K content and the C/N ratio, and it is positively correlated with N content. Except for C. leucantha, these correlations are all highly significant. The correlations with the C, P, and C/P ratio vary in terms of significance and direction. However, ML₁₀ is positively correlated with K content and the C/N ratio. There are significant or highly significant correlations between C. repens and H. japonicum with K, and H. japonicum and A. raddeana with C/N. The correlations with the C, N, P, and C/P ratio vary in terms of significance and direction.

3.3.2. Litter Decomposition of Different Stands

Although there were no significant differences in surface soil temperature among the three forest types (Table 2), and their soil substrates were also relatively similar (Table 1), it is evident from Figure 1 that there is a considerable standard deviation in the ML and ML₁₀ of the four early spring plants. Consequently, we plotted the decomposition residual rate curves of the early spring plants in different forest types and obtained their decomposition dynamic equations through linear fitting. Using this, we estimated the time required for 50% decomposition (t_0.05) and the time required for 95% decomposition (t_0.95), as shown in Figure 2 and

Table 5. The model fitting of decomposition rate of four kinds of early spring plants in different forest types.

Species	Forest Type	Regression Equation	R2	P	K (g·g−1·d−1)	t0.05 (d)	t0.95 (d)
C. leucantha	BF	y = 100.6054e−0.0220t	0.9594	0.0006	0.0220 Bb	32	136
	DBF	y = 92.3789e−0.0332t	0.9588	0.0025	0.0332 Aa	18	88
	BKF	y = 90.8058e−0.0263t	0.8814	0.0055	0.0263 ABa	23	110
C. repens	BF	y = 98.9167−0.0301t	0.9671	0.0004	0.0301 Bab	23	99
	DBF	y = 98.9881e−0.0467t	0.9747	0.0002	0.0467 Aa	15	64
	BKF	y = 98.5882e−0.0292t	0.9662	0.0004	0.0292 Ba	23	102
H. japonicum	BF	y = 102.6333e−0.0306t	0.9876	<0.0001	0.0306 Ba	24	99
	DBF	y = 100.6109e−0.0475t	0.9726	0.0003	0.0475 Aa	15	63
	BKF	y = 102.4126e−0.0388t	0.9797	0.0002	0.0388 ABa	18	78
A. raddeana	BF	y = 97.7255e−0.0301t	0.9691	0.0002	0.0301 Ba	22	99
	DBF	y = 96.0367e−0.0433t	0.9451	0.0012	0.0433 Aa	15	68
	BKF	y = 97.8242e−0.0294t	0.9775	0.0002	0.0294 Ba	23	101

Note: ^A and ^B indicate the difference in forest type under the same early spring plant species, ^a and ^b indicate the same forest type under different early spring plant species, the same letter indicates no significant difference, while different letters indicate significant difference.

.

In the three forest types, the mass residual rates of the four ESPL all exhibit a common characteristic: their decomposition is generally faster in deciduous broad-leaved forests, while the differences between birch forests and mixed coniferous–broad-leaved forests vary. The results of the equation fitting show that all decomposition models reached highly significant levels (p < 0.01, Table 5), with R² values ranging from 0.8814 to 0.9876 and decomposition coefficients (k) ranging from 0.0220 to 0.0475 g·g⁻¹·d⁻¹. There are significant differences in the k values of early spring plants across different forest types. The DBF forest type generally shows higher decomposition rates, especially for plants like C. repens and H. japonicum, with K values significantly higher than those in the BF and BKF forest types. This indicates that the forest type plays a key role in plant decomposition, with the DBF forest type accelerating the decomposition of plant litter. Although the decomposition rates of C. leucantha and A. raddeana show smaller differences across forest types, they still exhibit faster decomposition rates in the DBF forest type. Overall, the interaction between plant species and forest type determines the decomposition rate, with the DBF forest type promoting the decomposition of most early spring plants. Among the four types of litter, the time required for 50% decomposition and 95% decomposition in deciduous broad-leaved forests is only 15–18 days and 63–88 days, while, in broad-leaved forests, it requires 18–23 days and 78–110 days, and, in birch forests, it takes 22–32 days and 99–136 days. The time required is longest for C. leucantha and shortest for H. japonicum, with C. repens and A. raddeana falling in between. The decomposition coefficient of the four early spring plants is significantly higher in deciduous broad-leaved forests compared to birch forests, and C. repens and A. raddeana are also significantly higher in deciduous broad-leaved forests compared to mixed coniferous–broad-leaved forests, with no significant differences between birch forests and mixed coniferous–broad-leaved forests. However, the differences in decomposition rates of different early spring plants in the same forest type only manifest in birch forests, where the decomposition rate of C. leucantha is significantly lower than that of H. japonicum and A. raddeana. This indicates that the decomposition of the four early spring plants is significantly influenced by forest type, far more than by differences in early spring plant species.

4. Discussion

Early spring plants are the first to appear in the ecosystem, marking the end of winter and the restart of biological activity, and they support the growth and reproduction of other organisms by providing food and shelter and contribute to soil stabilization, reducing the risk of soil erosion. They play a positive role in connecting the material and energy flow between winter and spring seasons in forest ecosystems and in the restoration of ecosystem activity in spring [13]. Research has shown that they play a “Vernal Dam” role in nutrient cycling during seasonal changes [23,24]. In the process of nutrient cycling, the decomposition of ESPL is a crucial ecological process. It involves a series of physical, chemical, and biological processes that transform organic matter into increasingly stable forms, contributing to soil fertility, long-term carbon storage, and nutrient renewal [25,26], and plays an important role in linking the aboveground and belowground carbon and nutrient cycles in terrestrial ecosystems [27,28]. Compared to other plants, early spring plants can decompose rapidly. In this study, three typical early spring plants and one early spring flowering plant, the white-flowered plant, were selected for investigation to explore the influence of litter type, soil conditions, and forest type on the decomposition process of early spring plants. The results showed that, after 50 days of decomposition, the mean mass loss of the four early spring plants ranged from 73.15% to 80.44%, significantly faster than woody plant litter. Research has indicated that, in a broad-leaved Korean pine forest, 35 woody plant species experienced a 77% leaf mass loss over 6 years [29], which was significantly slower than the decomposition of early spring plant litter in this study. Other studies have also shown that herb species biomass decomposes faster than that of tree leaves [30,31]. There is a high correlation between leaf economic spectrum and the decomposition ability of leaf litter, ranging from fast-strategy species with relatively low construction costs, higher physiological activity, and shorter lifespan, to slow-strategy species with relatively higher construction costs, lower physiological activity, and longer lifespan, as the decomposition rate of litter varies from fast to slow [27,32,33,34]. Additionally, based on the ML in this study, the decomposition of litter from the four early spring plants showed an initial rapid phase followed by a slower phase, possibly due to the increasing proportion of lignin as the litter decomposes [35]. Furthermore, the favorable light conditions in early spring may also promote the transformation and relative loss of lignin in the early stages of litter decomposition [36]. Research showed that the litter decomposition rate was higher in spring ephemerals compared to summer green herbaceous species, as plants with smaller total aboveground biomass and leaf area decomposed more rapidly [37]. In the current study, there were significant differences in the decomposition of litter among the four early spring plants, with an overall trend of H. japonicum decomposing slightly faster and C. leucantha decomposing relatively slower. In the later stages of decomposition, the cumulative ML of H. japonicum was significantly higher than that of C. leucantha. Additionally, in the early stages of decomposition, H. japonicum and C. repens showed relatively faster ML. Similar research results have been obtained by other scholars. For example, a study showed that, in the Reaumuria soongorica community of the Sangong River Basin, it takes 2.52 years and 3.69 years for the litter to decompose by 50% for two different plant communities and 14.64 years and 20.13 years to decompose by 95%, but the former is a herbaceous plant community composed of Chenopodium glaucum and Artemisia scoparia, while the latter is a shrub-grass plant community composed of Artemisia scoparia and Salsola collina [38].

The decomposition process of plant litter is mainly influenced by the types and quantity of decomposers, the quality of the substrate, environmental temperature and humidity, and soil nutrient supply [39,40,41]. In experiments involving the decomposition of single-species and mixed-species litter in temperate natural secondary forests and artificial larch forests, it was found that the substrate quality of litter is the primary factor determining its decomposition rate, while the litter decomposition environment also plays a significant role [42]. Although the current study primarily focused on substrate quality and the decomposition environment as key drivers of ephemeral plant litter decomposition in early spring, we recognize that soil biota may also play an important role in this process. Some studies have even suggested that soil biota is the second most important driver, after substrate quality [42]. Microbial activity and invertebrate abundance in the soil may regulate or accelerate the decomposition of litter. As this study did not conduct a quantitative analysis of these factors, we are unable to further explore their specific roles based on the current data. However, we suggest that future research should incorporate the monitoring of microbial activity and soil fauna communities to gain a more comprehensive understanding of their potential contributions to ephemeral plant litter decomposition in spring. The C, N, P, and other elements in litter can provide essential nutrients for microbial decomposers to synthesize vital substances for life. Therefore, during the decomposition process, litter with higher nutrient content is more likely to stimulate the activity of microbial decomposers, thereby accelerating litter decomposition [43]. In this study, the ML of litter from the four early spring plants showed a significant or extremely significant negative correlation with K content and C/N ratio, and a positive correlation with N content. Except for C. leucantha, the correlations were all extremely significant. However, ML₁₀ showed a positive correlation with K content and the C/N ratio, with significant or extremely significant correlations observed only between C. repens and H. japonicum with K and between H. japonicum and A. raddeana with C/N. A 15-month study on the decomposition of litter from Gliricidia sepium, Canarium indium, and Theobroma cacao revealed that the mass loss of Gliricidia litter (59%) was greater compared to that of Theobroma cacao (37%) and Canarium indium (10%), attributed to the higher average concentrations of total nitrogen, boron, iron, and phosphorus in Gliricidia [44]. There are differences in C:N stoichiometry between microbial decomposers and the decomposing substrate, which drives the physiological constraints on litter decomposition and nutrient release [6,45,46]. Generally, the C/N ratio of fresh litter is higher than that of microbial decomposers, and microbes need to obtain nutrients from the surrounding environment (soil or precipitation) to meet their own growth requirements [47,48]. Numerous studies have found a strong correlation between the initial N content of litter and N-related substrate quality indices with the litter decomposition rate in the early stages of decomposition, demonstrating the limiting role of N in litter decomposition [45,49]. Although this evidence is indirect, indicators such as the C/N ratio and lignin/N ratio are often used to predict litter decomposition rates [6,49,50,51]. Some studies have shown that litter is primarily limited by nitrogen in the early stages of decomposition and by lignin concentration or lignin/nitrogen limitation in the later stages [50]. Additionally, the substrate quality of litter significantly influences soil nutrients through decomposition rates [52]. The supply of soil nutrients affects litter decomposition not only because decomposers (soil organisms) require nutrients from litter or soil to sustain their life activities [53], but also because soil nutrients indirectly influence litter decomposition by transforming litter into characteristics that enter the forest ecosystem, thereby changing the microenvironment during litter decomposition and directly impacting the process [54].

In this study, the environmental conditions such as soil temperature and nutrients during the decomposition process of the four ESPL did not show significant differences. However, the measured ML and ML₁₀ still exhibited considerable standard deviation values for the four plant species. To further elucidate the impact of environmental conditions on the decomposition of ESPL, litterbag decomposition experiments were conducted in three types of forest stands: birch forest, mixed broad-leaved forest, and broad-leaved Korean pine forest. The results revealed that the decomposition of the four ESPL types was fastest in the deciduous broad-leaved forest. This may be because the canopy closure in deciduous broadleaf forests regulates temperature and humidity, promoting microbial activity and accelerating organic matter decomposition, while the moisture released from thawing leaves is absorbed by the soil, helping to maintain soil moisture and reduce dry periods, further enhancing the decomposition process [55,56]. Research indicates that photodegradation, influenced by litter quality, is a key driver of decomposition, not only in arid regions but also in mesic ecosystems, such as temperate deciduous forests, following gap formation [57]. In the deciduous broad-leaved forest, it took only 15–18 days to reach 50% decomposition and 63–88 days to reach 95% decomposition. In the mixed broad-leaved forest, the corresponding times were 18–23 days and 78–110 days, while, in the birch forest, it took 22–32 days and 99–136 days. Among the plant species, C. leucantha required the longest time for decomposition, while H. japonicum required the shortest time, with C. repens and A. raddeana falling in between. However, the differences among the early spring plant species within the same forest type were not significant, except for C. leucantha, which showed significantly slower decomposition rates compared to H. japonicum and A. raddeana in the birch forest, possibly due to phosphorus limitation [58,59,60]. Multiple studies have shown that litter composition is a crucial direct regulatory factor in litter decomposition [9,10]. For example, based on incubations performed in the soil and canopy in different forests along an elevation gradient, it was found that overall litter composition explained 83% of the variability of mass loss, while 11% was related to microsite and 4% to the overall site characteristics [8]. It is evident that, in this study, the composition of the four ESPL did not emerge as the primary limiting factor affecting their degradation. Therefore, under similar climatic and environmental conditions, the differences in the decomposition patterns of the four ESPL species are not significant. The impact of the decomposition environment formed under different forest stand conditions is significantly greater than the differences in the species of the four early spring plants studied in this research.

5. Conclusions

This study analyzed the decomposition characteristics of four typical early spring plant litters in temperate forests in Northern China and clearly identified the prominent decomposition rates of early spring plants, with the mass loss of the four ESPL reaching as high as 73.15% to 80.44% after 50 days. Over the entire decomposition period, there was a significant correlation between the decomposition of ESPL and the decomposition time, showing an overall pattern of rapid initial decomposition followed by slower decomposition later, with H. japonicum exhibiting slightly faster decomposition and C. leucantha decomposing relatively slowly. The decomposition of the four ESPL showed a clear correlation with litter substrate quality, but the overall differences in ML and ML₁₀ among different plant litter types were not significant, indicating that differences in substrate nutrient composition did not lead to significant variation in ESPL decomposition in this study. Additionally, the environmental conditions such as soil temperature and soil nutrients during the decomposition process of the four ESPL studied were not significantly different. However, in different forest stands, the decomposition of ESPL was faster in deciduous broad-leaved forests, with the time to reach 50% decomposition and 95% decomposition taking only 15–18 days and 63–88 days, compared to 18–23 days and 78–110 days in mixed broad-leaved forests, and 22–32 days and 99–136 days in birch forests. This indicates that the decomposition environment, including factors such as soil temperature and humidity, soil water retention capacity, and microbial activity, formed under different forest stand conditions is a crucial direct regulatory factor for ESPL decomposition, compared to the differences in the species of the four early spring plants and soil environments. The research results on the decomposition characteristics of ESPL not only provide a basis for a comprehensive understanding of the nutrient cycling processes in northern temperate forest ecosystems but also offer scientific references for regulating the material and energy flow in early spring forest ecosystems.