Soil inorganic carbon in mangroves of tropical China: patterns and implications

Abstract

Soil inorganic carbon (IC) is neglected in most blue carbon studies despite the globally significant role of the calcium carbonate cycle in ocean C balance and climate change. We sampled soils to 1 m depth from seven mangrove reserves in Hainan Island, China. Only 45 out of 509 samples were rich in IC (greater than 10 mg cm⁻³). Most of the IC-rich samples were found at the outer part of Qinglan Bay, which is adjacent to the largest coral reef zone of Hainan Island. Soil IC concentration ranged from 0 to 66 g kg⁻¹ (or 0–67 mg cm⁻³), accounting for 0–92% of total C. IC concentration increased with soil depth where it was abundant. Soil pH was low (2.36–6.59) in IC-depleted soils, but increased to 5.67–7.99 in IC-rich soils. Soil total C stock and IC stock in mangroves of Hainan amounted to 0.76×10⁶ and 0.12×10⁶ Mg, respectively, with IC accounting for 16% of total C. Our study finds that carbonate concentrations can be high in mangrove soils but their spatial distribution indicates they are largely allochthonous in origin. Evidence of carbonate dissolution in mangroves suggests mangroves may increase total alkalinity to buffer acidification in seawater.

1. Introduction

In recent years, there has been increasing study of the carbon (C) stored in vegetated coastal wetlands (mangroves, saltmarshes, seagrasses) owing to their importance in sequestering carbon dioxide (CO₂) [1]. Soil (sediment) is the most important C pool in coastal wetlands [1]. In contrast to the well-recognized role of soil organic C in C cycling, soil inorganic carbon (IC) has been often neglected in blue carbon studies, although this may be an important factor when constructing C budgets of ecosystems. More importantly, calcium carbonate production and dissolution in the ocean and coastal ecosystems play a globally significant role in the ocean C balance and climate change [2,3]. In the oceans calcification facilitates the return of CO₂ to the atmosphere because the decrease of total alkalinity outweighs the uptake of dissolved inorganic carbon [3]. Specifically, for every mole of CaCO₃ precipitated, approximately 0.6 mol of CO₂ are released [2]. By contrast, for every mole of CaCO₃ dissolved, 0.6 mol of CO₂ are absorbed in seawater [3]. Owing to the opposite effects of calcium carbonate production and organic C sequestration in mitigating climate change, it has been proposed that calcium carbonate cycling needs to be accounted for in blue carbon studies to more accurately estimate C offsets in these coastal ecosystems [4].

Seagrass meadows have been found to be a large reservoir of calcium carbonates as they support diverse calcifying organisms [5], and may also receive carbonates from coral reefs [6]. IC stocks in the top 1 m of seagrass sediments average 654 Mg ha⁻¹ globally, exceeding organic C stocks by about a factor of five [5]. Although mangroves are often adjacent to seagrasses and coral reefs, there have been few studies focusing on assessing soil carbonates in mangroves [6].

In this study, we conducted a survey across seven mangrove reserves along the coasts of Hainan Island, which represent the tropical mangroves of China. We aimed to answer the questions: How much IC is stored in mangrove soils in Hainan? How does soil IC vary among sites and soil depths?

2. Material and methods

We sampled soil from 107 plots (10 × 10 m) in seven mangrove reserves in Hainan Island (18°10′–20°10′ N; 108°37′–111°03′ E), tropical China (figure 1). The area of the seven reserves (3303 ha) accounted for 92.4% of total mangrove area (3576 ha) in Hainan. The number of plots sampled in each reserve was determined roughly by the relative size of mangrove area: 47 in Dongzhai Bay, 29 in Qinglan Bay, 11 in Sanya, 5 in Dongfang, 6 in Danzhou, 5 in Dongchang-Caiqiao and 4 in Chengmai. Different plant communities and geomorphologic settings were considered in setting up plots in each reserve. Dominant mangrove species in sampled plots included Avicennia marina, Rhizophora stylosa, Rhizophora apiculata, Ceriops tagal, Bruguiera sexangula, Bruguiera gymnoihiza, Sonneratia alba, Sonneratia caseolares, Lumnitzera racemosa, Aegiceras corniculatum and Xylocarpus granatum. The study region is characterized by a tropical monsoon climate. The mean annual air temperature is 23–25°C, with a maximum in July and a minimum in January. The mean annual rainfall is around 1600 mm.

Figure 1.

Locations of the seven mangrove reserves sampled in the study. The blue strips and circles outside Qinglan Bay and Sanya indicate the two largest coral reef zones around Hainan Island [7].

Two soil cores were randomly taken from each plot with a semi-opened steel corer of 5 cm in diameter and 1 m in depth during low tide. Each soil core was separated into five segments by 20 cm intervals. Two core segments of the same soil depth from each plot were pooled as a composite sample for subsequent analyses.

Fresh weights were measured after sample collection, and subsamples were dried in an oven at 105°C to measure soil moisture content. Soil bulk density was measured as dry weight divided by volume (calculated from the corer). Air-dried soil samples were ground and passed through 0.15 mm mesh. Total C concentration of all 509 soil samples were measured by dry combustion with an elemental analyser (Elementar vario MAX CNS, Germany). Soil organic C concentration was measured by wet oxidation [8] for 291 samples. Soil IC concentration was calculated as the difference between total C and organic C concentrations (method 1). Ninety-two soil samples were again measured for IC concentration by the elemental analyser without or with removing inorganic C using HCl (method 2). Method 2 yielded slightly higher values of soil IC concentration (g kg⁻¹) than method 1 (Method 2 = 1.133 × Method 1 + 1.69; r² = 0.968; electronic supplementary material, figure S1) and method 1 failed to detect IC in samples with very low IC (typically less than 3 g kg⁻¹). We used this relation to calibrate the IC concentrations of samples measured by method 1. Carbon stable isotope composition, δ¹³C, was measured with an isotope ratio mass spectrometer (Thermo Scientific Delta V, Germany). The strong relationship between δ¹³C and the percentage of IC in total C (IC measured by method 2) (δ¹³C = 0.243 × (IC percentage) − 25.76; r² = 0.965; electronic supplementary material, figure S2) was used to calculate the IC concentrations of the other 218 samples not measured for IC but measured for δ¹³C. This relation was due to the fact that carbonates are much more enriched in ¹³C (around 0‰) than soil organic C. Volume-based IC concentration (mg cm⁻³) was calculated by multiplying weight-based concentration (g kg⁻¹) by soil bulk density (g cm⁻³). Soil pH (NBS scale) was measured with a pH meter at a soil/water ratio of 1/2.5.

In calculating soil total C and IC stocks of each reserve, we roughly distinguished the areas of different mangrove communities at different geomorphologic settings. Since the mangrove areas of the seven sampled reserves accounted for 92.4% of the total mangrove area in Hainan, soil total C and IC stocks in mangroves of Hainan were calculated as 1.08 times (1/0.924) the summed stocks of the seven reserves.

Differences of soil total C and IC concentrations among soil depths and reserves were analysed by two-way ANOVA with SPSS software. In nine plots of Qinglan Bay and two plots of Sanya where IC was rich, difference of soil IC concentration among soil depths was separately analysed by one-way ANOVA with the 11 plots as replicates.

3. Results

Soil total C concentration ranged from 1.8 to 120 g kg⁻¹ (or 2.7–82 mg cm⁻³), varying among soil depths and reserves (p < 0.001), and the pattern with soil depth varied among reserves as indicated by the significant interactions (p < 0.001; table 1). Soil inorganic C (IC) concentration ranged from 0 to 66 g kg⁻¹ (or 0 – 67 mg cm⁻³), accounting for 0–92% of total C. Only 45 out of 509 samples were rich in IC (greater than 10 mg cm⁻³), with 40 of them found at the outer part of Qinglan Bay and four found at Sanya (figure 2). Although two-way ANOVA showed no significant differences of IC among soil depths owing to the majority of samples having low IC (table 1), soil IC concentration increased with soil depth in 11 plots where IC was rich (p < 0.001; figure 2h).

Table 1.

Soil total C and inorganic C (IC) concentrations among soil depths in seven mangrove reserves of Hainan (mean ± s.e.).

reserve name (location)	soil depth (cm)	soil total C (g kg−1)	soil total C (mg cm−3)	soil IC (g kg−1)	soil IC (mg cm−3)
Qinglan Bay	0–20	34.2(3.2)	31.3(2.7)	9.8(2.6)	10.4(2.8)
(19°33′–19°37′ N 110°46′–110°50′ E)	20–40	39.1(4.5)	35.6(3.2)	11.6(2.9)	13.3(3.5)
	40–60	38.3(4.6)	39.7(4.1)	13.9(3.5)	16.9(4.4)
	60–80	36.7(3.7)	41.6(4.3)	16.0(4.1)	19.9(5.1)
	80–100	30.6(3.8)	36.6(4.8)	12.3(4.2)	16.1(5.6)
Chengmai	0–20	10.7(3.1)	13.2(3.3)	1.3(0.4)	1.8(0.6)
(19°54′ N 109°59′ E)	20–40	6.7(1.4)	10.6(2.0)	1.3(0.4)	2.1(0.7)
	40–60	6.5(1.1)	10.7(2.0)	1.3(0.4)	2.1(0.7)
	60–80	7.1(0.9)	11.0(1.7)	1.3(0.4)	2.0(0.7)
	80–100	7.9(1.0)	11.6(1.8)	1.4(0.3)	2.0(0.4)
Dongchang-Caiqiao	0–20	33.1(5.6)	31.3(2.5)	2.7(1.0)	2.6(0.8)
(19°50′–19°51′ N 109°31′–109°34′ E)	20–40	22.2(5.2)	23.2(3.6)	2.6(0.9)	2.8(0.7)
	40–60	18.9(4.2)	21.4(2.7)	1.8(0.1)	2.2(0.2)
	60–80	15.5(5.4)	19.5(4.9)	1.4(0.3)	2.1(0.7)
	80–100	18.3(0.1)	23.2(0.3)	1.0(0.7)	1.3(0.9)
Danzhou	0–20	34.3(3.8)	32.2(2.0)	3.1(1.6)	2.9(1.5)
(19°43′–19°46′ N 109°15′–109°18′ E)	20–40	27.6(5.3)	26.4(3.4)	1.4(0.3)	1.4(0.3)
	40–60	16.2(2.2)	19.7(2.5)	1.6(0.1)	1.9(0.1)
	60–80	13.1(2.6)	17.1(2.9)	1.4(0.3)	1.8(0.4)
	80–100	11.1(2.8)	15.0(3.8)	1.4(0.3)	1.9(0.4)
Dongfang	0–20	23.0(2.1)	14.2(1.2)	2.1(0.4)	1.3(0.3)
(19°12′ N 108°38′ E)	20–40	18.9(3.5)	13.7(1.8)	1.9(0.2)	1.5(0.3)
	40–60	24.3(3.6)	22.9(2.6)	1.9(0.2)	1.9(0.2)
	60–80	31.6(3.1)	30.9(1.4)	2.5(0.8)	2.5(0.8)
	80–100	32.3(0.8)	25.1(4.7)	4.1(2.4)	3.5(2.4)
Sanya	0–20	22.6(4.0)	20.6(2.5)	1.7(0.4)	1.8(0.6)
(18°13′–18°16′ N 108°34′–108°42′ E)	20–40	19.0(4.3)	19.1(3.0)	2.5(0.7)	3.1(1.0)
	40–60	20.0(4.6)	21.9(4.1)	5.2(2.8)	6.5(3.4)
	60–80	19.7(6.2)	22.2(6.6)	6.9(5.2)	8.0(5.9)
	80–100	11.3(4.0)	13.3(2.5)	1.7(0)	2.6(0.5)
Dongzhai Bay	0–20	34.6(3.8)	22.1(1.2)	0.1(0.1)	0.1(0.0)
(19°53′– 20°0′ N 110°32′–110°37′ E)	20–40	21.0(2.2)	17.8(1.3)	0.1(0.0)	0.2(0.0)
	40–60	11.3(1.2)	12.0(1.0)	0.2(0.0)	0.3(0.0)
	60–80	9.3(1.0)	10.3(0.9)	0.3(0.0)	0.3(0.0)
	80–100	8.3(0.9)	10.4(1.1)	0.4(0.0)	0.5(0.1)

Figure 2.

Frequency distributions of soil inorganic C (IC) concentrations in seven reserves (a–g) and the pattern of soil IC with soil depth in nine plots at Qinglan Bay and two plots at Sanya where soil IC was abundant (h). Different letters in (h) indicate significant differences of soil IC among soil depths (p < 0.05).

Soil pH was 2.36 – 6.59 in IC-depleted soils, but increased to 5.67 − 7.99 in IC-rich soils (electronic supplementary material, figure S3).

Soil total C stock and IC stock in mangroves of Hainan amounted to 0.76 × 10⁶ and 0.12 × 10⁶ Mg, respectively, with IC accounting for 16% of total C (table 2).

Table 2.

Area, soil total C and inorganic carbon (IC) densities and stocks in seven mangrove reserves of Hainan.

reserve name	area (ha)	soil total C density (Mg ha−1)	soil IC density (Mg ha−1)	soil total C stock (Mg)	soil IC stock (Mg)	percentage of IC in total C (%)
Qinglan Bay	984	62–739	3–577	316 982	91 920	29.0
Chengmai	154	69–166	2–27	17 579	3063	14.7
Dongchang-Caiqiao	157	129–281	16–29	35 152	3627	10.3
Danzhou	115	125–321	5–36	25 166	2299	9.1
Dongfang	250	183–255	11–37	54 631	4886	8.9
Sanya	84	72–478	13–325	15 532	4676	30.1
Dongzhai Bay	1559	44–294	0–7	240 875	3401	1.4
all reserves	3303			705 917	114 202	16.2
non-protected	273			58 062	9393	16.2
Hainan in total	3576			763 979	123 595	16.2

4. Discussion

We found high concentrations of carbonates in mangrove soils at some sites, but these carbonates may be largely allochthonous in origin. IC-rich soil samples were found at the outer part of Qinglan Bay and Sanya (figure 2). Correspondingly, the largest and second largest coral reef zones around Hainan were adjacent to Qinglan Bay (105.5 km²) and Sanya (14.4 km²), respectively (figure 1) [7]. Coral reefs may thus be the major source of soil carbonates in these mangroves given their geographical proximity, as also suggested in a previous study [6]. It is unlikely that the abundant carbonates were produced in situ by calcifying organisms because mangroves are not a favourable habitat for calcification compared with seagrasses owing to relatively lower pH [9]. The carbonates are also unlikely to be of lithogenic origin. Hainan Island has few karst areas and none of the seven reserves sampled in our study is located at a karst zone [10].

The low pH in carbonate-depleted soils and the increased IC concentration with soil depth suggest that mangroves may play an important role in carbonate dissolution. In carbonate-depleted soils, conditions were acid (pH 2.36–6.59) (electronic supplementary material, figure S3), which favours carbonate dissolution [11]. The higher pH (5.67–7.99) in carbonate-rich soils might be due to the reaction between acids and carbonates. Similarly in Gazi Bay mangroves in Kenya, pH ranged from 3.5 to 8 in soils depleted in carbonates, but was buffered at around 8 in soils rich in carbonates [11]. Acids in sediments are generated via aerobic decomposition of organic matter, root respiration, oxygen exudation from roots and oxidation of reduced iron and sulfur species [11,12]. Given that aerobic decomposition rates, oxidation–reduction reactions and mangrove root biomass all decrease with soil depth, carbonate dissolution may also decrease with soil depth, leading to increased IC concentration at depth (figure 2h). The increasing IC with depth may be also due to the fact that mangroves colonize soils after soil elevation reaches mean sea level. If soil accretion after mangrove colonization was not thick enough, carbonate dissolution by mangroves may be confined to shallow soil depths.

We showed that IC concentration can be high in some mangrove sites where allochthonous carbonates may have contributed to surface elevation rise. Mangroves may partially counteract seawater acidification through carbonate dissolution and the export of total alkalinity to adjacent oceans [13].

Ethics

This study was conducted under verbal approval of each reserve involved (reserve full names listed in electronic supplementary material, table S1).

Data accessibility

Data are available as the electronic supplementary material.

Authors' contributions

Y.X. designed the study; W.G., Y.X. and B.L. collected the data; Y.X. wrote the manuscript and W.G. and B.L. provided critical revision; all authors approved the final version of the manuscript and agree to be held accountable for the content.

Competing interests

We have no competing interests.

Funding

The study was funded by the Chinese Academy of Forestry (CAFYBB2017MA005), Guangdong Forestry Department (2014KJCX021-02) and the National Natural Science Foundation of China (41776103).