Non-negligible role of dead organic matter in a rainforest remnant in Northeast Brazil

Abstract

Dead organic matter represents an essential reservoir of carbon, especially that allocated in standing dead trees, coarse woody debris, and fine litter, playing a pivotal role in nutrient cycling and habitat provisioning. However, necromass is frequently disregarded in forest assessments. Here, we aimed to perform the first assessment of multiple necromass compartments in the Atlantic Forest of Northeast Brazil, providing a basis for future integrative studies related to necromass in this region. We registered 17 standing dead trees in 0.5 hectare and 239 logs of coarse woody debris. Necromass had 3.9 Mg.ha^-1 of standing dead trees, 54.24 Mg.ha^-1 of coarse woody debris and 7.2 Mg.ha^-1 of litter. We indicate that standing dead trees and coarse debris were mostly in the intermediate and final stages of decomposition. Leaves were the dominant component of litter, and drier months had more litterfall. Finally, we highlight that assessing standing dead trees and coarse woody debris adds 25.6% on top of aboveground tree mass, improving information about organic matter storage in rainforest ecosystems. Our findings emphasize that the necromass compartment must be considered in forest assessments, also including small pieces of coarse woody debris, which could inform better practices of forest management.

Key words
decay stage; DOM; line-intercept method; litter; urban forest

Resumo

A matéria orgânica morta representa uma reserva essencial de carbono, especialmente aquela alocada na necromassa arbórea (árvores mortas em pé, serapilheira grossa e serapilheira fina), desempenhando um papel fundamental na ciclagem de nutrientes e na provisão de habitat. Entretanto, a necromassa é frequentemente desconsiderada em inventários florestais. Aqui, objetivamos realizar o primeiro levantamento de vários componentes da necromassa na Floresta Atlântica do Nordeste do Brasil, fornecendo uma base para futuros estudos integrativos relacionados à necromassa nessa região. Registramos 17 árvores mortas em pé em 0,5 hectare e 239 tocos de serapilheira grossa. As estimativas de necromassa foram 3,9 Mg.ha^-1 para árvores mortas em pé, 54,24 Mg.ha^-1 para serapilheira grossa e 7,2 Mg.ha^-1 para serapilheira fina. Indicamos que a maior parte das árvores mortas em pé e serapilheira grossa estavam em estágios intermediários e finais de decomposição. A fração folhas foi o componente dominante da serapilheira fina e meses secos tiveram maior queda deste componente. Finalmente, destacamos que inventariar árvores mortas em pé e serapilheira grossa representa um incremento de 25,6% na biomassa arbórea acima do solo, aprimorando as informações sobre estoques de matéria orgânica em florestas úmidas. Destacamos que o compartimento de necromassa deve ser considerado em inventários florestais, também incluindo pedaços pequenos de serapilheira grossa, o que pode informar melhores práticas de manejo florestal.

Palavras-chave
estágio de decomposição; MOM; linha interceptadora; serapilheira; floresta urbana

Introduction

Dead organic matter (DOM) derived from trees in natural or anthropic ecosystems is stored in different compartments (e.g., standing or fallen dead trees, dead trunks, and branches, woody bark, and litter (Keller et al. 2004; Barbosa et al. 2009; Palace et al. 2012). These dead plant components represent a remarkable contribution to long-term ecosystem functioning and balance, especially in forest ecosystems (Sanchez et al. 2009; Gove et al. 2012; Bassett et al. 2015). In such systems, forest necromass helps to ensure energy and habitat for a wide range of living organisms, acting as an essential nutrient pool on the forest floor (Barbosa et al. 2017). Also, it stores impressive amounts of carbon in tropical forests (Palace et al. 2008; Sefidi & Mohadjer 2010), with further benefits for water absorption and storage (Zaninovich et al. 2016).

Forest plant necromass is commonly divided into two groups: fine litter/ litter (elements with diameter < 2 cm) and coarse litter (diameter > 2 cm). The coarse litter includes two important compartments, standing dead trees (SDT) and coarse woody debris (CWD) (Barbosa et al. 2009; Palace et al. 2012). The fine litter is composed of plant debris accumulated on the forest floor, corresponding to ~20% of the total aboveground necromass (Palace et al. 2007). Fine litter composition (hereafter, litter) consists of leaves, flowers, fruits, seeds, barks, and thin branches, as well as miscellaneous, non-identifiable material (Schessl et al. 2008; Espig et al. 2009).

In mature forests, total necromass (fine and coarse litter) contributes between 5 to 25% of total plant stock and can range from 5–40% (Brown 1997), 10–20% (Houghton et al. 2001), or 20–40% (Palace et al. 2012) of aboveground carbon. The forest necromass component has been frequently studied to understand its role in ecosystem functioning, such as rates of accumulation and decomposition involved in nutrient cycling and respiration. Furthermore, applied questions are also assessed, such as its role on forest management and conservation, carbon storage in restored areas (Longhi et al. 2011), and its changes along forest successional stages (Delaney et al. 1998; Chambers et al. 2000; Scheer 2008; Vendrami et al. 2012).

Previous studies in Brazil did not take into account all necromass compartments and were performed predominantly in the Amazon rainforest (Chambers et al. 2000; Keller et al. 2004; Rice et al. 2004; Palace et al. 2007; Baker et al. 2007; Palace et al. 2008; Chao et al. 2009; Cruz Filho & Silva 2009; Silva et al. 2016). In some regions of the Atlantic Forest, on the contrary, the study of necromass is in its early stages. Particularly concerning the coarse compartments of necromass (CWD and SDT), there are even fewer studies in the Atlantic forest, with more information in the southern regions of Brazil (Vieira et al. 2011; Ribeiro et al. 2012; Sanquetta et al. 2014; Deus et al. 2018; Fonsêca et al. 2019; Moreira et al. 2019b; Villanova et al. 2019).

The complete characterization of necromass compartments in forest inventories is critical to understand ecosystem dynamics and health, because it allows a better comprehension of ecosystem services provided by forest remnants, and contributes to forest planning, conservation, and management (Stinson et al. 2011; Fonsêca et al. 2019). Considering that few studies accurately assess necromass compartments, our question was related to how much necromass is stored in coarse and fine compartments. All forest compartments are dynamic, but the values ​​reach certain stability in mature systems. Data collected in a single moment represents this dynamic stability, which happens in various paces, from swift to seasonal or supra-annual rates. In this study, we performed a thorough inventory of plant dead organic matter (DOM) comprehensively including its two major components: coarse litter (fallen trees and standing dead trees) and aboveground fine litter, in a fragment of Atlantic Forest in the Northeast of Brazil. Since litter is the component with the fastest dynamics, we evaluated litter stock and seasonal production. Thus, we portrayed not only the relationship between compartments but also the intense dynamics necessary to maintain them, as exemplified by the litter.

Material and Methods

Study site

The study was conducted in a remnant of Atlantic rainforest located in the Dois Irmãos State Park (PEDI). The PEDI is a protected area in the state of Pernambuco, Northeast Brazil, comprising 1,157.72 ha and divided into two parts, with very distinct characteristics: mature forest (MF) with 384.7 ha (7°57’S; 34°56’W) and Regenerating forest (RF) with 773.02 ha (Fig. 1). The necromass inventory was conducted in the mature forest, which hosts the most conserved forest patch.

Figure 1
Representation of the study site, Dois Irmãos State Park, Northeast, Brazil. This study was carried out in the MF area, representing a mature and well-conserved forest surrounded by a peri-urban matrix.

The climate is classified as As’ according to Köppen classification, which represents a warm and humid climate, with an average annual precipitation of 2,417 mm. Rainfall is concentrated in autumn-winter, while monthly mean temperatures are always higher than 23 ºC (data available in the repository of the National Institute of Meteorology: <www.inmet.gov.br>). According to the Brazilian Soil Database, in the region where PEDI is located, ferralsols, acrisols, and arenosols are the predominant soil types. Soil acidity varies from medium to high, which is as expected for regions with high rainfall (Caldas 2007). The vegetation is composed of typical botanical families found in the Atlantic Forest, including Moraceae, Fabaceae and Melastomataceae, comprising 265 species of plants with average density of stems as 1025.75 ind.ha^-1 (PPBio Program/ Atlantic Forest Network/ PEDI site in the study area - Cunha et al. 2021). Diameter at the breast height varies from 5 to 114 cm (mean = 15.65 ± standard deviation = 13.82), and the height of living trees varies from 2 to 27 m (10.73 ± 5.87) (Fonsêca et al. 2020). Furthermore, the average tree biomass is 255.67 Mg.ha^-1, with average wood density registered of 0.6413 g.cm³ (Fonsêca et al. 2020).

Standing Dead Trees (SDT)

To assess the standing dead tree compartment (also known as ‘snags’), we used 50 plots (10 m × 10 m), which were 10 m apart from each other. We considered SDT as (i) dead individuals without leaves, (ii) with dead tissue under the bark (Veiga 2010), and (iii) DBH (diameter at breast height) ≥ 5 cm. We recorded DBH for each dead tree and additionally the diameter at soil level and height. Additionally, we classified all SDT according to their decomposition stage, using a tactile-visual categorization, following the PPBio protocol for necromass assessment (Barbosa et al. 2009) (Tab. 1). Decomposition stage is essential to estimate tree density since we did not use stem specific density as a parameter.

Table 1
Decomposition stages used to classify standing dead trees (SDT) (based on Barbosa et al. 2009) and stages of coarse woody debris (CWD) (based on Harmon et al. 1995).

Decomposition stage	Description of decomposition stage categories
Standing dead trees (SDT)
Stage 1	Trees recently dead with no sign of insect or fungal damage, showing no apparent decomposition (< 10% mass loss)
Stage 2	Trees with minor signs of insect and fungal damage, usually with small hollows but still showing high mechanical resistance to touch (10-30% mass loss)
Stage 3	Trees on advanced decomposition stage, with few signs of rotting and that often crumbled at touch (> 30% mass loss).
Coarse woody debris (CWD)
Stage 1	Logs with the most extensive bark cover, showing remaining leaves and twigs, and least decayed
Stage 2	Solid wood with intact bark (recently started to fall) but no leaves or branches
Stage 3	Logs with no bark, but the heartwood (inner part of the CWD) was relatively undecayed
Stage 4	Logs with no branches or bark cover, already rotten and fragile wood that can be broken if kicked
Stage 5	Rotten, fragile wood that could be disintegrated or easily break if clamped by hand

We classified SDT in four classes of diameter, based on DBH: class A (5 to 9.9 cm), class B (10 to 29.9 cm), class C (30 to 49.9 cm), and class D (≥ 50 cm). These classes were adopted to categorize SDT information of density (ind.ha^-1), necromass (Mg.ha^-1) and basal area (m².ha^-1). Then, we calculated volume by using the method indicated by Harmon & Sexton (1996) and adapted for this study, using the following equation:

$$V=\frac{H(A_b+4A_m+A_s)}{6}$$ (1)

where V = estimated volume (m³.ha^-1), H = height, A_b = basal area (using diameter at soil level), A_m = mean area (calculated using DBH), A_s = upper area (using the taper function indicated by Chambers et al. (2000) to determine D_h, the upper diameter).

Coarse woody debris (CWD)

We adopted the Line Intersect Sampling (LIS) method (Warren & Olsen 1964; van Wagner 1968) to sample logs on the forest floor. The LIS method comprises lines on the forest floor to sample logs that were intercepted by the line in its central axis (Deus et al. 2018; Fonsêca et al. 2019). In this study, we installed ten straight lines of 100 m, 20 m apart from each other, systematically positioned on the left side of the plots used to assess SDT, comprising 1 km of assessed length.

Log diameters were recorded at the point intersected by the line, considering pieces with ≥ 2 cm diameter, measured with a tape. In samples damaged by decomposition, we registered two perpendicular measures to avoid bias in the measure (Barbosa et al. 2009). We used diameters to calculate the estimated volume (V_Est) of coarse woody debris in the field, following van Wagner’s equation:

$$V_{Est}=\frac{{\pi }^2\times d^2}{8\times L}$$ (2)

where V_Est = estimated volume (m³.ha^-1), d = diameter of each piece included in the sampling line, L = length of the sampling line (100 m).

We adopted diameter classes indicated by Keller et al. (2004) to adjust the solid volume obtained for each of the pieces of wood assessed. Thus, we considered (i) small pieces (S) = between 2 and 4.9 cm, (ii) intermediate pieces (I) = between 5 and 9.9 cm, (iii) and large pieces (L) = diameter ≥ 10 cm. We adopted the same rationale used to classify decomposition stage of SDT, but with small adjustments. A tactile-visual categorization was performed regarding the decomposition stage (Tab. 1) of all pieces measured in the transects, following the recommendations of Harmon et al. (1995).

Necromass estimation for SDT and CWD

The necromass for both SDT and CWD was obtained based on the volume of each piece of wood and tree measured in the field, and the basic density (g.cm³) indicated by Veiga (2010), by diameter category and degree of decomposition of the pieces (Tab. S1, available on supplementary material <https://doi.org/10.6084/m9.figshare.19379882.v1>). We choose the values reported by Veiga (2010) because they were standardized in the same decomposition class that we adopted here. Moreover, Veiga (2010) reported wood density values from the Atlantic Forest that can also be considered in our study site (in the lack of an adequate assessment of wood density). The necromass of wood and tree pieces was obtained by multiplying the volume of each category by its basic density (g.cm³) (3):

$$Necr=V_{Est}\ast D_b$$ (3)

where Necr = necromass (Mg.ha^-1), V_Est = estimated volume (m³.ha^-1); D_b = basic density (g.cm³).

Production and storage of fine litter (Lt)

The fine litter was assessed from August 2012 to July 2014, comprising two years of monthly sampling. We placed squared litter traps with 0.25 m² (0.5 × 0.5 m), ~1 mm mesh size, in 1.0 m above the forest floor (Schessl et al. 2008). Litter traps were randomly placed within a subset of 10 plots from the 50 plots used to assess SDT. We used plastic bags to collect Lt every month and later weighed them on a 0.1 g precision scale to obtain the wet weight. Then, we separated samples into four categories (leaves, branches < 2 cm in diameter, reproductive parts and unidentifiable parts - miscellaneous) and finally dried at 60 °C for 72 h to measure the dry weight, our primary Lt variable (Schessl et al. 2008).

The litter-layer was estimated at the end of the study period, from the random arrangement of 10 squares with the same dimensions as the litter traps (0.5 × 0.5 m), in the same plots used to access the production. In those squares, we collected all litter above forest soil (stopping the sampling when we reached the fine root mat, which was not considered (Sampaio et al. 1988). Then, we stored in plastic bags, oven-dried and weighed using the same methods for litter production. We highlight that, even though the standard protocol suggests a time-replicated sampling of the litter-layer, our study site is a forest with evergreen vegetation which do not presents extreme values of leaf fall across the year. Therefore, we argue that potential gains of information from litter-layer would be greater than a possible bias from a single-time sampling.

Data analysis

We estimated the values of SDT, CWD and Lt to hectares. To describe the compartments, we calculated the mean, standard deviation, and confidence intervals. Correlation analysis was also performed with the historical record of rainfall averages (30 years) from the climatological series provided by INMET from 1961–1990. We further applied analysis of variance (ANOVA), considering blocks (two years of records) and replicates (10 litter traps), using the Assistat software (Silva & Azevedo 2002). We considered years as blocks in the ANOVA test because there is little evidence to expect significant differences between one year and the other. By doing this, we could understand if the same months from different years presented divergent patterns of litterfall. In both analyses, data were normalized by a logarithmic transformation to base 10.

Results

Standing dead trees (SDT)

The standing dead trees (SDT) compartment presented 17 trees, with an estimated density of 34 ind.ha^-1. The estimated volume was 10.31 m³.ha^-1, with a stored necromass of 3.9 Mg.ha^-1 (Fig. 2). The classification of SDT in three decomposition stages resulted in more individuals belonging to the intermediate stage of decomposition (stage 2 = 52.9%) than in initial and final stages (stages 1 = 11.7% and stage 3 = 35.2%). Regarding the diameters, we identified a predominance of individuals with intermediate diameters (class B = 52.9%,). Classes A and C presented the same proportion, with 17.6% of the individuals (Tab. 2). When the basal area values were analyzed, we found that despite classes C and D only adding up to 10 individuals, they had higher values than the other classes: 73.2% of the necromass and 72.5% of the total basal area.

Figure 2
Distribution of necromass components, sampled in a forest of the Dois Irmãos State Park, Northeast, Brazil.

Table 2
Distribution by diameter classes of standing dead tree density (ind.ha^-1), necromass (Mg.ha^-1) and basal area (m².ha^-1), with values between parenthesis indicating the range for each class at Dois Irmãos State Park, Pernambuco, Brazil.

	A (5–9.9 cm)	B (10–29.9 cm)	C (30–49.9 cm)	D (≥ 50 cm)	Total
Density (ind.ha-1)	6 (17.6%)	18 (52.9%)	6 (17.6%)	4 (11.8%)	34 (100%)
Necromass (Mg.ha-1)	0.04 (1.1%)	1.01 (25.7%)	1.52 (39%)	1.34 (34.2%)	3.9 (100%)
Basal area (m².ha-1)	0.02 (1.65%)	0.27 (25.83%)	0.30 (28.91%)	0.45 (43.61%)	1.03 (100%)

Coarse woody debris (CWD)

Our study registered 239 pieces of trunks and fallen trees on the ground (CWD), which comprised a volume of 284.74 m³.ha^-1 and a necromass of 54.24 Mg.ha^-1 (Fig. 2). Most logs were in intermediate to high decomposition stage, with 41.4% (n = 99) considered in DS4 and 28.4% in DS5 (n = 68). Other degrees of decomposition made up 30.13% of the log frequency.

However, when we considered not only the frequency but the necromass stored per each decomposition stage, we found that DS4 summed up to 50% (27 Mg.ha^-1) and DS3 28.48% (15.45 Mg.ha^-1) of total necromass. DS5, DS2 and DS1 had altogether 21.7% of the remaining necromass (Tab. 3). Regarding the adopted diameter classes, larger diameter CWD pieces (class G) predominated in the necromass stock, with 42 Mg.ha^-1 (79.5% of the fallen necromass, with 37 pieces). This value is higher than for M and P classes, which stored 15% and 8.1% of the necromass, respectively.

Table 3
Coarse woody debris (CWD) abundance and necromass per decomposition stage (DS), with percentage values between parenthesis, at the Dois Irmãos State Park, Pernambuco, Brazil.

	DS1	DS2	DS3	DS4	DS5	Total
Log abundance	1 (0.4%)	11 (4.6)	60 (25.1%)	99 (41.4%)	68 (28.45)	239 (100%)
Necromass (Mg.ha-1)	0.02 (0.05%)	3.08 (5.7%)	15.45 (28.48%)	27.0 (49.8%)	8.69 (16.0%)	100%

Fine litterfall and litter-layer

The monthly average of litterfall was 0.95 Mg.ha^-1 (Standard deviation ±0.33), with an annual fall of 11.36 Mg.ha^-1.year^-1 (SD ± 3.98). The wet mass of the fine litter was 1.55 Mg.ha^-1.month^-1 (SD ± 0.55) and 18.65 Mg.ha^-1.year^-1 (SD ± 6.56), with 37% of water loss after drying, which indicate a higher water accumulation in litter. Litterfall was mostly composed of leaves, which accounted for 58% (6.58 Mg.ha^-1.year^-1) of the total litterfall. Branches, reproductive parts and miscellaneous fractions accounted for 22.2% (2.52 Mg.ha^-1.year^-1), 13.7% (1.56 Mg.ha^-1.year^-1) and 6.2% (0.70 Mg.ha^-1.year^-1), respectively. Interestingly, the major fall of branches (0.4 Mg.ha^-1) occurred in July 2013, when the second-highest precipitation value (415.6 mm) was recorded. The total stock of fine litter on the forest soil summed 7.2 Mg.ha^-1 (SD ± 1.78) (Fig. 2), which is much lower than the annual fall of plant material.

Precipitation during the study period ranged from 8.2 mm (November 2012) to 481.5 mm (June 2013), with a monthly average of 183.5 mm (SD ± 128.4 mm), registering higher values between May and July and having an annual sum of 1,968 mm and 2,435 mm for the first and second year, respectively (average 2,201.5 mm). Data from the historical series ranged from 35.7 mm (November) to 388.1 mm (July), summing up to 2417.6 mm. The monthly temperature varied 4 °C during the study period, with a minimum of 23.8 °C (August/2012) and a maximum of 27.8 °C (February/2013).

The correlation between historical data of precipitation and present monthly litterfall revealed a significant negative relationship (r = -0.68, p < 0.05) (Fig. 3). We found a negative correlation between litterfall and mean current precipitation (r = -0.78, p < 0.01). Regarding temporal changes in litterfall during the study, as expected, there was a clear difference within each year (between months), with a sharp distinction between the low litterfall in wetter months (March, April and July) and high litterfall in drier months (October, December and January) (Fig. 3).

Figure 3
Values of fine litterfall (Mg.ha^-1) per each month studied, compared with historical and current precipitation (mm). The data were collected from August 2012 to July 2014, in the Dois Irmãos State Park, Northeast, Brazil.

Discussion

We found that necromass is a crucial compartment in mass storage, comprising ~58 Mg.ha^-1 of fallen and standing dead trees, plus ~7 Mg.ha^-1 of fine litter accumulated on the forest floor (Fig. 2). Furthermore, as a temporary component, fine litter summed ~11 Mg.ha^-1 per year. Our data suggest that necromass stocks add 25.6% on top of estimated aboveground tree biomass (previously measured for the same forest, Fonsêca et al. 2019). Necromass from late-stage succession forests are usually 5–25% aboveground biomass (FAO 2020), indicating that the study site still stores an impressive amount of necromass, even though such compartment is often neglected in forest ecosystem assessments.

Coarse woody debris represented the most significant part of necromass when standing dead trees and fine litter were compared. The pattern CWD > SDT > Lt is expected because coarse pieces of necromass are more frequent in the forest floor, usually have more volume and require more time to decompose due to woody compounds, such as lignin (Gessner et al. 2010). The literature indicates large values with a high variation of CWD in the Amazon rainforest, ranging from 31 to 62.3 Mg.ha^-1 (Palace et al. 2012; Keller et al. 2004; Nascimento & Laurance 2002; Rice et al. 2004). The few studies available in the Atlantic Forest were performed in very contrasting ecosystems, such as seasonal semi-deciduous forests, Araucaria angustifolia-dominated forests, and lowland rainforests. Their reports were lower than the Amazon and ranged from ~4 to 30 Mg.ha^-1 for CWD (Deus et al. 2018; Moreira et al. 2019a; Villanova et al. 2019; Fonsêca et al. 2019; Maas et al. 2020), which indicates an extensive variation even within Atlantic Forest systems. Accordingly, standing dead trees (SDT) follow the same pattern, but with less variation reported in the Amazon (from 4 to 10 Mg.ha^-1; Nascimento & Laurance 2006; Baker et al. 2007; Chao et al. 2008a) and with a remarkable divergence in the Atlantic Forest (0.6 Mg.ha^-1 to 19.6 Mg.ha^-1; Veiga 2010; Luccas 2011; Maas et al. 2020). Our sampling indicated a significant amount of necromass stored in CWD (54.24 Mg.ha^-1), which is even higher than previous estimates (the first in the North of Atlantic Forest) on the same study site using different methods (fixed-plots, Fonsêca et al. 2019).

In contrast to the CWD pool, SDT stored ~4 Mg.ha^-1, but most necromass was stored in few snags with high diameter, even though most dead trees had intermediate diameter. When putting our data in an integrated perspective, impressive high values of CWD and low values of SDT could suggest that dead trees do not remain standing, which increases CWD component. In the same forest, Fonsêca et al. (2020) found 37 recently fallen trees along 4 km of line intercepts, which might indicate that the forest floor is a major repository of dead trees and would explain the reason why SDT component had so few individuals and necromass. Moreover, standing dead trees are usually a result of physiological stress, such as light competition (Chao et al. 2008b). However, broken or uprooted dead trees are more likely an outcome of mortality events derived from the surrounding environment (Esquivel-Muelbert et al. 2020), such as windstorms and edge-effects (Magnago et al. 2015). Such predictions are in accordance with the fact that our study site is a small peri-urban forest, often under multiple disturbances derived from external factors (i.e., wind and edge-effects). Future investigations can elucidate whether mortality type (standing dead, uprooted, or broken trees) respond to wind, temperature, and light competition.

The decomposition stage of coarse necromass is a crucial parameter to characterize dead organic matter but also provides insights about ecosystem functioning because decomposition is the main process related to nutrient cycling in forest ecosystems. Coarse woody debris was often in intermediate to advanced stages of decomposition (Fonsêca et al. 2019; Moreira et al. 2019a), likely because of more time in contact with the forest floor, which contains most microbial activity (Gora & Lucas 2019). In contrast, late-decomposition CWD stores less carbon (Harmon et al. 2013) and is likely not as effective as live biomass to remove carbon from the atmosphere. Even though larger pieces of necromass are harder to decompose due to higher area to volume ratio (Harmon et al. 2020), we found that most CWD registered were in advanced stages of decomposition and had larger pieces of deadwood, storing more necromass (and potentially more carbon). The higher abundance of late-stage decomposition pieces of CWD is likely a function of accumulation through time. Such discrepancy was not found in the SDT compartment. Standing dead trees were in earlier stages of decomposition and mostly in intermediate sizes. This pattern likely exists because of two factors. First, decomposition is slower when the organic matter is far from the soil (Gora & Lucas 2019), due to reduced moisture which decreases microorganism activity (Gessner et al. 2010). Second, SDT in advanced stages of decomposition are highly susceptible to fall on the forest floor in episodes of forest disturbance (Anderegg et al. 2016), transferring more necromass to the CWD compartment.

The fine litter compartment embodied 11% of all necromass, with a litterfall of 11.36 Mg.ha^-1 per year and a stock of 7.2 Mg.ha^-1. Values of fine litter production that we reported in this study follow previous reports from forests located in similar areas of Atlantic Forest (10.07–14.74 Mg.ha^-1; Espig et al. (2009); Schessl et al. (2008)). As expected, most parts of litter were leaves (58%), which is similarly reported by several studies (Espig et al. 2009; Schessl et al. 2008; Longhi et al. 2011), with proportions varying from 48.5–74.7%. In contrast, litter stock seems to be lower than other sites. Maas et al. (2020) performed two samples of litter stock and indicated that such compartment represented 33% of all necromass in an Araucaria-dominated forest in the South of Brazil, while our values of litter stock represented 10% of total necromass. Sampaio et al. (1993) registered 13 Mg.ha^-1 of litter stock for the same forest studied here more than 30 years ago. However, we draw attention to caveats of comparing litter stocks due to non-standardized sampling. Still, Litter-layer stock values in our system must be interpreted carefully because this compartment typically needs seasonal sampling due to litterfall seasonality and correlation to abiotic conditions (Schessl et al. 2008). We emphasize that our study was performed in a region where the vegetation is evergreen and not seasonally deciduous, as shown by Lima et al. (2008). When compared to other Atlantic rainforest sites, whether this recorded low value is explained by a geographic pattern (response to climate) or is a consequence of the forest conservation status (size, age, disturbances) is still a question to be answered when more data are gathered. Furthermore, our aim was not to follow litter stocks through time but to account for as many necromass compartments as possible.

Our results from different compartments allow us to discuss the implications of coarse and fine necromass to the functioning of our study site. First, coarse necromass has a non-negligible role in carbon storage. Due to the correlation between biomass/necromass and carbon storage in tropical rain forests, we inferred that ~45% of the total stored mass is carbon (Veiga 2010; Moreira et al. 2019b). Thus, CWD and SDT are estimated to store 26 MgC.ha^-1, which is equivalent to ~20% of the total carbon stored in aboveground tree biomass for our study site (Fonsêca et al. 2019) and represents an important ecosystem service of regulation. This potential role of carbon storage is very relevant, especially to larger CWD pieces. Second, the litterfall deposition increased over time and is likely returning more nutrients to the forest floor. We found a massive increase in litterfall from the 1980s (Sampaio et al. 1988) to present values, with 3 Mg.ha^-1.year^-1 more litter at the present time. Considering the high decomposition rates (Sena et al. 2018) coupled with more litter falling aboveground, we could suggest that nutrients are cycling faster, and there is more primary production in the forest. That is especially true for leaves in the fine litter compartment. Strong evidence suggests that leaves decompose seven times faster than other fractions, adding inputs of particularly Fe, Mn, and N (Caldeira et al. 2008; Vital et al. 2004; Barbosa et al. 2017) that are absorbed by thin and superficial roots (Sampaio et al. 1988).

Information derived from coarse and fine compartments of necromass can be fundamental to forest management and monitoring, providing insights about ecosystem dynamics and stocks. For example, forest management initiatives should keep larger pieces of CWD in the forest because they store more carbon (Nascimento & Laurance 2002). However, our results indicate that if necromass with less than 10 cm is not considered, managers should be careful that necromass estimations can still be potentially 20% lower. We suggest that CWD from all sizes and decomposition stages should be considered in the line intercept method, mainly because this method is not as costly and time-consuming as the fixed-plots approach (Fonsêca et al. 2019). However, our results indicate that the SDT compartment stored a small portion of necromass compared to CWD. Therefore, we strongly suggest that forest inventories should prioritize CWD component over SDT when time and financial resources are limited.

After depicting necromass compartments, we found that long-term and transient compartments were differently distributed in the forest, with distinct characteristics. We firstly registered a high amount of necromass storage for the Northeast Atlantic forest, with higher values than other areas in the same ecosystem. Moreover, our results highlight that coarse woody debris were from intermediate to late-stage decomposition and were mostly composed of intermediate diameter, storing high amounts of necromass (and consequently carbon). Transient compartments, such as fine litterfall, are mostly comprised of leaves and present higher values in drier months. We indicate that necromass compartments must be included in forest inventories and ecological research, especially coarse woody debris compartment, which represented an impressive pool of necromass.

Acknowledgements

The authors would like to thank the Dois Irmãos State Park (PEDI) management team, for their authorization and support during this research. Prof. Enio Silva and Dr. Nadielan Lima provided part of the meteorological data used. We are also grateful to two anonymous reviewers who helped to improve our final text. Finally, we kindly thank the essential assistance provided by the Laboratório de Ecologia Vegetal (UFRPE) in the field and lab work. We also kindly thank the Programa de Educação Tutorial (PET/MEC/SESu) of Ecology (UFRPE), for granting scholarships to PHAS and ACBLS, with additional acknowledgments to the Institutional Program for the Scientific Initiation Scholarships (PIBIC/CNPq/UFRPE) granted to PHAS.

References

• Anderegg WRL, Martinez-Vilata J, Gailleret M, Camarero JJ, Ewers BE, Galbraith D, Gessler A, Grote R, Huang C, Levick SR, Powell TL, Rowland L, Sánchez-Salguero R & Trotsiuk (2016) When a tree dies in the forest: scaling climate-driven tree mortality to ecosystem water and carbon fluxes. Ecosystems 19: 1133-1147.
• Baker TR, Coronado ENH, Phillips OL, Martin J, van der Heijden GMF, Garcia M & Espejo JS (2007) Low stocks of coarse woody debris in a southwest Amazonian forest. Oecologia 152: 495-504.
• Barbosa RI, Silva LFSG & Cavalcante CO (2009) Protocolo de Necromassa: estoque e produção de liteira grossa. PPBIO - Programa de Pesquisa em Biodiversidade. Ministério da Ciência e Tecnologia. Instituto Nacional de Pesquisas da Amazônia, Boa Vista. 24p.
• Barbosa V, Barreto-Garcia P, Gama-Rodrigues E & Paula A (2017) Biomassa, carbono e nitrogênio na serapilheira acumulada de Florestas plantadas e nativa. Floram 24: e20150243.
• Bassett M, Chia EK, Leonard SWJ, Nimmo DG, Holland GJ, Ritchie EG, Clarke MF & Bennett AF (2015) The effects of topographic variation and the fire regime on coarse woody debris: insights from a large wildfire. Forest Ecology and Management 340: 126-134.
• Brown S (1997) Estimating biomass and biomass change of tropical forests: a Primer. FAO Forestry Paper 134. Food and Agriculture Organization of the United Nations, Rome. 55p.
• Caldas AM (2007) Solos, antropização e morfometria da microbacia do Prata, Recife, PE. Dissertação de Mestrado. Universidade Federal Rural de Pernambuco, Recife. 130p.
• Caldeira MVW, Vitorino MD, Schaadt SS, Moraes E & Balbinot R (2008) Quantificação de serapilheira e de nutrientes em uma Floresta Ombrófila Densa. Semina: Ciências Agrárias 29: 53-68.
• Chambers JQ, Higuchi N, Schimel JP, Ferreira LV & Melack JM (2000) Decomposition and carbon cycling of dead trees in tropical forests of the central Amazon. Oecologia 122: 380-388.
• Chao K, Phillips OL & Baker TR (2008a) Wood density and stocks of coarse woody debris in a northwestern Amazonian landscape. Canadian Journal of Forest Research 38: 795-825.
• Chao KJ, Phillips OL, Gloor E, Monteagudo A, Torres-Lezama A & Martínez RV (2008b) Growth and wood density predict tree mortality in Amazon forests. Journal of Ecology 96: 281-292.
• Chao KJ, Phillips OL, Baker TR, Peacock J, Lopez-Gonzalez G, Vasquez Martinez R, Monteagudo A & Torres-Lezama A (2009) After trees die: quantities and determinants of necromass across Amazonia. Biogeosciences 6: 1615-1626.
• Cruz DFO & Silva JNM (2009) Avaliação da quantidade de resíduos lenhosos em floresta não explorada e explorada com técnicas de redução de impactos, utilizando amostragem por linha interceptadora, no Médio Mojú, Amazônia. Acta Amazonica 39: 527-532.
• Cunha JAS, Fonsêca NC, Cunha JSA, Rodrigues LS, Gusmão RAF & Lins-e-Silva ACB (2021) Selective logging in a chronosequence of Atlantic Forest: drivers and impacts on biodiversity and ecosystem services. Perspectives in Ecology and Conservation 19: 286-292.
• Delaney M, Brown S, Lugo AE, Torres-Lezama A & Quintero NB (1998) The quantity and turnover of dead wood in permanent forest plots in six life zones of Venezuela. Biotropica 30: 2-11.
• Deus KHP, Bonete IP, Figueiredo AFO, Dias NA & Bonete IP (2018) Woody necromass stock in mixed ombrophilous forest using different sampling methods. Revista Caatinga 31: 674-680.
• Esquivel-Muelbert A, Phillips OL, Brienen RJW, Fauset S, Sullivan MJP, Baker TR, Galbraith D et al. (2020) Tree mode of death and mortality risk factors across Amazon forests. Nature Communications 11: 1-11.
• Espig SA, Freire FJ, Marangon LC, Ferreira RLC, Freire MBGS & Espig DB (2009) Sazonalidade, composição e aporte de nutrientes da serapilheira em fragmento de Mata Atlântica. Revista Árvore 33: 949-956.
• FAO - Food and Agriculture Organization of the United Nations (2020) Global forest resources assessment 2020 - Key findings. FAO Forestry Department, Rome. 16p.
• Fonsêca NC, Meunier IMJ & Lins-E-Silva ACB (2019) Evaluation of the plant necromass component: methodological approaches and estimates in Atlantic Forest, Northeast Brazil. Floram 26: 1-10.
• Fonsêca NC, Meunier IMJ & Lins-E-Silva ACB (2020) Can fallen trees enhance aboveground biomass estimation? A proposal for the Brazilian Atlantic Forest. Revista de Biologia Tropical 68: 1284-1297.
• Gessner MO, Swan CM, Dang CK, McKie BG, Bardgett RD, Wall DH & Hättenschwiler S (2010) Diversity meets decomposition. Trends in Ecology & Evolution 25: 372-380.
• Gora EM & Lucas JM (2019) Dispersal and nutrient limitations of decomposition above the forest floor: evidence from experimental manipulations of epiphytes and macronutrients. Functional Ecology 33: 2417-2429.
• Gove JH, Ducey MJ, Valentine HT & Williams MS (2012) A distance limited method for sampling downed coarse woody debris. Forest Ecology and Management 282: 53-62.
• Harmon ME & Sexton J (1996) Guidelines for measurements of woody detritus in forest ecosystems. Publication n° 20. U.S. Long-term Ecological Research. (LTER) Network office, University of Washington, Seattle. 73p.
• Harmon ME, Fasth B, Woodall CW & Sexton J (2013) Carbon concentration of standing and downed woody detritus: effects of tree taxa, decay class, position, and tissue type. Forest Ecology and Management 291: 259-267.
• Harmon ME, Whigham DF, Sexton J & Olmsted I (1995) Decomposition and mass of woody detritus in the dry tropical forests of the northeastern Yucatan Peninsula, Mexico. Biotropica 27: 305.
• Harmon ME, Fasth BG, Yatskov M, Kastendick D, Rock & Woodall CW (2020) Release of coarse woody detritus-related carbon: a synthesis across forest biomes. Carbon Balance Manage 15: 1.
• Houghton RA, Lawrence KT, Hackler J & Brown LS (2001) The spatial distribution of forest biomass in the Brazilian Amazon: a comparison of estimates. Global Change Biology 7: 731-746.
• INMET - Instituto Nacional de Meteorologia (2014) Normais climatológicas do Brasil (1961-1990). Available at <http://www.inmet.gov.br/portal/index.php?r=clima/normaisClimatologicas>. Access on 25 July 2019.
  » http://www.inmet.gov.br/portal/index.php?r=clima/normaisClimatologicas
• Keller M, Palace M, Asner GP, Pereira R & Silva JNM (2004) Coarse woody debris in undisturbed and logged forests in the eastern Brazilian Amazon. Global Change Biology 10: 784-795.
• Lima ALA, Rodal MJN & Lins-e-Silva ACB (2008) Phenology of tree species in a fragment of Atlantic Forest in Pernambuco-Brazil. Bioremediation, Biodiversity and Bioavailability 2: 68-75.
• Longhi RV, Longhi SJ, Chami LB, Watzlawick LF & Ebling AA (2011) Produção de serapilheira e retorno de macronutrientes em três grupos florísticos de uma Floresta Ombrófila Mista, RS. Ciência Florestal 21: 699-710.
• Luccas FS (2011) Estoques de necromassa em um cerrado sensu stricto e uma Floresta Ombrófila Densa Montana, no estado de São Paulo. Dissertação de Mestrado. Instituto de Botânica da Secretaria de Meio Ambiente, São Paulo. 100p.
• Maas GC, Sanquetta CR, Marques R, Machado SDA, Sanquetta MN, Corte APD & Schmidt LN (2020) Combining sample designs to account for the whole necromass carbon stock in Brazilian Atlantic Forest. Journal of Sustainable Forestry: 1-17.
• Magnago LFS, Rocha MF, Meyer L, Martins SV & Meira-Neto JAA (2015) Microclimatic conditions at forest edges have significant impacts on vegetation structure in large Atlantic forest fragments. Biodiversity and Conservation 24: 2305-2318.
• Moreira AB, Gregoire TG & Couto HTZ (2019a) Estimation of the volume, biomass and carbon content of coarse woody debris within two forest types in the state of São Paulo, Brazil. Forestry 92: 278-286.
• Moreira AB, Gregoire TG & Couto HTZ (2019b) Wood density and carbon concentration of coarse woody debris in native forests, Brazil. Forest Ecosystems 6: 18.
• Nascimento HEM & Laurance WF (2002) Total aboveground biomass in central Amazonian rainforests: a landscape-scale study. Forest Ecology and Management 168: 311-321.
• Nascimento HEM & Laurance WF (2006) Efeitos de área e de borda sobre a estrutura florestal em fragmentos de floresta de terra-firme após 13-17 anos de isolamento. Acta Amazonica 36: 183-192.
• Palace M, Keller M, Asner GP, Silva JNM & Passos C (2007) Necromass in undisturbed and logged forests in the Brazilian Amazon. Forest Ecology and Management 238: 309-318.
• Palace M, Keller M, Hurtt G & Frolking S (2012) A review of above ground necromass in tropical forests. In: Sudarshana P, Nageswara-Rao M & Soneji J (eds.) Tropical forests. Department of Geography of Maryland, IntechOpen, London. Pp. 215-252.
• Palace M, Keller M & Silva H (2008) Necromass production: studies in undisturbed and logged amazon forests. Ecological Applications 18: 873-884.
• Ribeiro A, Péllico-Netto S, Stall D, Leão RA & Nascimento FAF (2012) Proposta metodológica para realização de um inventário florestal de necromassa: um estudo de caso. Scientia Forestalis 40: 121-127.
• Rice AH, Pyle EH, Saleska SR, Hutyra L, Palace M, Keller M, Camargo PB, Portilho K, Marques DF & Wofsy SC (2004) Carbon balance and vegetation dynamics in an old-growth Amazonian forest. Ecological Applications 14: 55-71.
• Sampaio EVSB, Katia SN & Lemos EP (1988) Ciclagem de nutrientes na Mata de Dois Irmãos (Recife - PE) através da queda de material vegetal. Pesquisa Agropecuária Brasileira 23: 1055-1061.
• Sampaio EVSB, Dall’Olio A, Nunes KS & Lemos EP (1993) A model of litterfall, litter layer losses and mass transfer in a humid tropical forest at Pernambuco, Brazil. Journal of Tropical Ecology 9: 291-301.
• Sanchez E, Gallery R & Dalling JW (2009) Importance of nurse logs as a substrate for the regeneration of pioneer tree species on Barro Colorado Island, Panama. Journal of Tropical Ecology 25: 429-437.
• Sanquetta CR, Corte APD, Pinto C & Melo LAN (2014) Biomass and carbon in non-woody vegetation, dead wood and litter in Iguaçu National Park. Revista Floresta 44: 185-194.
• Scheer MB (2008) Decomposição e liberação de nutrientes da serapilheira foliar em um trecho de floresta ombrófila densa aluvial em regeneração, Guaraqueçaba (PR). Floresta 38: 253-266.
• Schessl M, Silva WLD & Gottsberger G (2008) Effects of fragmentation on forest structure and litter dynamics in Atlantic rainforest in Pernambuco, Brazil. Flora 203: 215-228.
• Sefidi K & Marvie Mohadjer MR (2010) Characteristics of coarse woody debris in successional stages of natural beech (Fagus orientalis) forests of Northern Iran. Journal of Forest Science 56: 7-17.
• Sena PHA, Lins-E-Silva ACB & Gonçalves-Souza T (2018) Integrating trait and evolutionary differences untangles how biodiversity affects ecosystem functioning. Oecologia 188: 1121-1132.
• Silva FAZ & Azevedo CAV (2002) Versão do programa computacional ASSISTAT para o sistema operacional Windows. Revista Brasileira de Produtos Agroindustriais 4: 71-78.
• Silva LFSG, Castilho CVC, Cavalcante CO, Pimentel TP, Fearnside PM & Barbosa RI (2016) Production and stock of coarse woody debris across a hydro-edaphic gradient of oligotrophic forests in the northern Brazilian Amazon. Forest Ecology and Management 364: 1-9.
• Stinson G, Kurz WA, Smyth CE, Neilson ET, Dymond CC, Metsaranta JM, Boisvenue C, Rampley GJ, Li Q, White TM & Blain D (2011) An inventory-based analysis of Canada’s managed forest carbon dynamics, 1990 to 2008. Global Change Biology 17: 2227-2244.
• Van Wagner CE (1968) The line intersect method in forest fuel sampling. Forest Science 14: 20-26.
• Veiga LG (2010) Estoque de madeira morta ao longo de um gradiente altitudinal de Mata Atlântica no nordeste do estado de São Paulo. Dissertação de Mestrado. Universidade Estadual de Campinas, São Paulo. 71p.
• Vendrami JL, Jurinitz CF, Castanho CT, Lorenzo L & Oliveira AA (2012) Produção de serrapilheira e decomposição foliar em fragmentos florestais de diferentes fases sucessionais no Planalto Atlântico do estado de São Paulo, Brasil. Biota Neotropica 12: 136-143.
• Vieira AS, Alves LF, Duarte-Neto PJ, Martins SC, Veiga LG, Scaranello MA, Picollo MC, Camargo PB, Carmo JB, Souza Neto E, Santos FAM, Joly CA & Martinelli LA (2011) Stocks of carbon and nitrogen and partitioning between above- and belowground pools in the Brazilian coastal Atlantic Forest elevation range. Ecology and Evolution 1: 421-434.
• Villanova PH, Torres CMME, Jacovine LAG, Soares CPB, Silva LF, Schettini BLS, Rocha SJSS & Zanuncio JC (2019) Necromass carbon stock in a secondary atlantic forest fragment in Brazil. Forests 10: 833.
• Vital, ART, Guerrini, IA, Franken WK & Fonseca RBF (2004) Produção de serapilheira e ciclagem de nutrientes de uma floresta estacional semidecidual em zona ripária. Revista Arvore 28: 793-800.
• Warren WG & Olsen PF (1964) A line intersect technique for assessing logging waste. Forest Science 10: 267-276.
• Zaninovich SC, Fontana JL & Gatti MG (2016) Atlantic Forest replacement by non-native tree plantations: Comparing aboveground necromass between native forest and pine plantation ecosystems. Forest Ecology and Management 363: 39-46.

Supplementary Material

See supplementary material at <https://doi.org/10.6084/m9.figshare.19379882.v1>

Edited by

Area Editor: Dr. Bruno Rosado

Publication Dates

• Publication in this collection
  11 Apr 2022
• Date of issue
  2022

History

• Received
  25 July 2020
• Accepted
  07 May 2021