Agroforestry: A Sustainable Land-Use Practice for Enhancing Productivity and Carbon Sequestration in Madhupur Sal Forest, Bangladesh

Abstract

This paper explores the role of agroforestry in sequestering atmospheric carbon in the tropics and subtropics, specifically in the Madhupur Sal forest of Bangladesh. Agroforestry, combining trees with crops on agricultural lands, is recognized for its potential to act as a carbon sink and enhance productivity. The study assesses various agroforestry practices, including acacia–pineapple–turmeric–papaya, acacia–pineapple–ginger–banana, and sal–pineapple–aroid combinations. This study innovatively assessed both the carbon sequestration and economic viability of agroforestry in the Madhupur Sal forest, presenting a sustainable land-use model that balances environmental benefits and farm profitability. The research reveals improved farm productivity in these agroforestry systems, with different tree species sequestering varying amounts of carbon. Acacia species, ranging from 12 to 25 ft in height, sequestered an average of 23.35 lbs/year, while sal species (Shorea robusta), with trees 45 to 61 ft tall, sequestered 49.80 lbs/year on average. Factors such as tree height, diameter at breast height (DBH), number of leaves, and branches influence carbon sequestration. The paper suggests that the carbon sequestration (CS) potential of agroforestry results in greenhouse gas emission reduction in Bangladesh. By emphasizing the profitability of these practices alongside carbon sequestration, the study encourages the adoption of agroforestry as a sustainable and economically viable strategy.

1. Introduction

Agriculture is a vital driver of Bangladesh’s economic growth, contributing significantly to GDP, employment, and export revenues [1]. Despite its historical importance, the industry faces challenges such as diminishing arable land, population growth, and declining productivity. Agroforestry is the strategic cultivation of trees with crops and/or livestock to produce multiple sustainable products or benefits from the same management land [2]. In response to ecological challenges, agroforestry has been recognized as a viable strategy [3]. Trees play a crucial role in mitigating ecosystem effects by capturing atmospheric carbon through photosynthesis [4]. The area of Madhupur Sal forest in Bangladesh, with prevalent acacia–pineapple-based agroforestry practices, serves as a promising site for studying the productivity and carbon sequestration potential of different agroforestry systems [5].

Carbon sequestration is the process of capture and long-term storage of atmospheric carbon dioxide to mitigate global warming. In other words, it also refers to the process of removing carbon from the atmosphere and depositing it in a reservoir. These carbon storages or reservoirs are also known as carbon pools [6]. Tree species composition has a significant impact on carbon sequestration, with mixed stands showing the highest overall capacity. Converting pure forests to mixed forests is recommended as an effective strategy to enhance carbon sequestration, both under present and future climate scenarios [7]. Dalbergia sissoo exhibits the highest above-ground carbon sequestration at 916.98 kg t⁻¹ and a total biomass carbon of 254.72 kg t⁻¹ [8]. Eucalyptus saligna sequesters 38.74 t/ha of above-ground and 10.07 t/ha of below-ground carbon. B. ceiba is noted for its high carbon stock accumulation potential, reaching 181 kg [9]. Pure oak forests, consisting of 57% carbon by mass, stand out as significant carbon reservoirs [10]. Comparatively, mixed forests show a carbon storage potential of about 53%, benefiting from species diversity [11]. Additionally, Eucalyptus urophylla has demonstrated a total carbon stock of 236.89 MgC/ha in biomass [12].

The global rise in carbon dioxide and methane concentrations, mainly due to industrialization, underscores the importance of forests as natural climate change breaks [13]. Madhupur Sal forest, covering a substantial portion of Bangladesh’s forested territory, plays a crucial role in sequestering carbon [14]. This forest is a secondary tropical deciduous area located within the Tangail Forest Division of Bangladesh. The country’s forest policies and management practices have roots in the British colonial era, followed by bureaucratic policies from the East Pakistan period. These policies were primarily focused on exploiting forest resources for revenue generation. During the 1980s, as part of management strategies, a significant portion of the Madhupur Sal Forest was allocated by the Forest Department for rubber plantations and social forestry initiatives [15,16,17,18]. Simultaneously, the government designated 20,837.2 acres (8436.1 ha) of the forest as a national park to support conservation efforts [15]. Between 2010 and 2014, the natural forest area expanded by 202.4 ha, thanks to a revegetation program led by the national forest department in collaboration with local community groups [19].

The estimation of the total carbon stock in Madhupur Sal forest is essential for understanding its pivotal role in mitigating climate change at both the national and global levels. Bangladesh, being highly vulnerable to climate change, has experienced an increase in CO₂ levels, primarily from fossil fuel burning [20]. Efforts to reduce CO₂ emissions include managing carbon fluxes in agricultural systems and utilizing mechanisms like the Kyoto Protocol’s clean development mechanism. Forestry, particularly agroforestry, has gained global recognition as a solution to reduce CO₂ emissions, with tropical deforestation accounting for a significant portion of yearly CO₂ emissions. In Bangladesh, the government recognizes the importance of restoring degraded sal forest. The government implemented the project of restoring 137,800 ha of deforested hill and plain land sal forest and 20,000 ha of degraded hill and plain land sal forest [21].

While agroforestry’s benefits were initially emphasized at the local or regional scales, there is a growing awareness of its global environmental services, including sustainable production, carbon sequestration, and others. Assessing carbon sequestration potential is crucial for understanding agroforestry’s role in social and ecological perspectives. This research aimed to investigate various agroforestry systems in the Madhupur Sal forest, evaluating tree and crop productivity and the potential of carbon sequestration simultaneously. This research uniquely combines ecological and economic analyses of agroforestry practices, highlighting their role in carbon sequestration while simultaneously improving farm productivity. The study provides a practical framework for policymakers and stakeholders to integrate agroforestry into national climate strategies, promoting both environmental sustainability and rural economic growth. By identifying high-carbon-sequestering species and evaluating cost–benefit ratios, our findings offer actionable insights for scaling up agroforestry practices in Bangladesh and similar tropical regions.

2. Materials and Methods

2.1. Study Area

The study was carried out in the Madhupur tract (45,565.2 acres), commonly known as the Madhupur Sal Forest or Madhupur Garh. In Bangladesh, moist deciduous plain land sal forests are distributed over the relatively drier central and northwestern parts of the country [17,22]. This area is mainly made up of the districts of Tangail, Mymensingh, Gazipur, and Dhaka; of the above, it is in the Madhupur sal forest (under Tangail and Mymensingh districts) particularly that the majority of the sal forests exist. The Madhupur Sal forest is located in the northeastern part of the Tangail Forest Division, with a small portion running along the boundary with the Mymensingh Forest Division [23,24].

The climate of the Madhupur Garh varies slightly from north to south, the northern reaches being much cooler in winter. Average temperatures vary from 29.3 °C to 21.1 °C in summer, falling to 20 °C in winter, with extreme lows of 10 °C. Rainfall ranges between 1500 mm to 2100 mm annually and the average is 2011.6 mm. Mean annual relative humidity (RH) and total evaporation are 84.8% and 1050 mm, respectively [25]. The Madhupur Sal tract belongs to the Bio-ecological Zone No. 3 and the 28th AEZ (Agro-Ecological Zone) of Bangladesh [26].

The study was conducted in six randomly selected important villages (Dokhola, Auronkhola, Madhupur National Park, Makontinagar, Gaira, and Beribaith) where the ethnic Garo people mostly practice various types of agroforestry (Figure 1). Three different agroforestry practices (acacia–pineapple–turmeric–papaya, acacia–pineapple–zinger–banana, and sal–pineapple–aroid-based) were selected to quantify the productivity and sequestration status of carbon as a whole in Madhupur Sal forest of Bangladesh.

2.2. Soil Type

We conducted mechanical soil testing in a soil laboratory, analyzing three samples from each village. Soil texture was determined using the hydrometer method, following the guidelines of Bouyoucos (1927) and Piper (1950) [27,28]. The textural classes of the soil samples were identified by plotting the results on a triangular diagram, as developed by Marshal (1947), according to the USDA system [29]. The percentages of sand, silt, and clay particles, along with the soil textures, are provided in

Table 1. The type of soil in the study area.

Location	Sand (%)	Silt (%)	Clay (%)	Textural Class
Dhokola (1)	23.05	42.54	34.65	Clay loam
Auronkhola (2)	22.66	41.76	33.73	Clay loam
Madhupur National Park (3)	18.48	53.28	29.42	Silty clay loam
Makontinagar (4)	21.32	44.06	34.65	Clay loam
Gaira (5)	45.76	45.65	34.84	Silty clay loam
Beribaith (6)	22.78	43.21	32.47	Clay loam

.

2.3. Data Collection

The study used questionnaire survey techniques to collect data on the economic aspects of agroforestry practices in the Madhupur Sal forest area. The questionnaire included both open-ended and closed-ended questions, and its development was guided by the recommendations of Kumar (2018) [30]. To ensure scientific representation of the selected farmers, the study employed a stratified random sampling technique. The sample size was determined using the formula proposed by Krejcie and Morgan (1970) [31]

$$\mathrm{n}={\displaystyle \frac{N \times {\mathrm{Z}}^2 \times p(1-p)}{E^2\left(N-1\right)+Z^2 \times p(1-p)}}$$

(1)

where:

• n = sample size;
• N = population size;
• Z = Z-score corresponding to the desired confidence level (1.96 for 95% confidence);
• p = estimated proportion of the population (0.5 used to ensure maximum variability);
• E = margin of error (set at 0.05 for 5% precision).

On the basis of this formula, the study selected 100 farmers for data collection, ensuring proportional representation of different agroforestry practices and village locations. This method minimized selection bias and ensured coverage of the major agroforestry systems in the Madhupur Sal forest area.

On the contrary, for the carbon sequestration of agroforestry practices, the study used quadrant (20 m × 20 m) techniques for collecting ecological data from 60 sample plots (20 m × 20 m) (20 sample plots from each agroforestry practice) that were chosen at random from six different villages in the agroforestry field. A similar experimental design was employed by Farukh et al. (2023) [10]. In that study, the entire study area was divided into four segments based on specific location characteristics. Following a similar approach, our study divided the area according to prevailing agroforestry practices. Subsequently, plots were selected randomly within these segments for data collection. For calculating carbon sequestration, tree height, DBH, and the age of trees were collected from the sample plots. The height of the tree was measured using a Nikon Forestry Pro-II Hypsometer (Manufactured by Nikon, Singapore), and the DBH was measured with a diameter tape. We gathered information about the tree’s age from the agroforestry owners, who, in most cases, were local farmers.

During data collection, we accounted for various factors that could have impacted our results, including soil quality and fertility, climatic conditions, management practices, human activities, pest and disease pressure, farmers’ knowledge and participation, and policy and economic incentives [32,33].

Factors such as measurement inaccuracies, fluctuations in wood density, and variations in soil carbon storage can lead to uncertainties in carbon sequestration estimates. For example, discrepancies in wood density can substantially influence biomass calculations, potentially causing errors in carbon stock evaluations [34]. Furthermore, differences in soil carbon storage, influenced by soil type and land management practices, can contribute to additional uncertainty in estimating carbon sequestration [35]. To minimize these errors, we employed precise measurement technologies and conducted extensive sampling across time and locations to capture variability and enhance accuracy. Additionally, we implemented quality control measures, such as blind and hot checks, and performed soil testing to ensure consistent environmental conditions for data collection, particularly for soil quality and climatic factors. For the other factors, we categorized the population into groups on the basis of their influence and selected samples where these factors were consistent. To strengthen the estimation of Net Present Value (NPV) and benefit–cost ratio (BCR), we supplemented the questionnaire data with actual market prices and recent yield records.

2.4. Calculation of Net Present Value (NPV)

The difference between the present value of all future projected cash inflows and the present value of all future expected cash outflows is the Net Present Value. For calculating Net Present Value, first, we identified the future benefits and then worked out the present and future costs. Then the present value of future costs and benefits were calculated. The present value factor is

1/(1 + r)ⁿ

(2)

where:

• r = Discounting rate;
• n = Number of years

The formula for calculating the present value is:

Present Value of Future Benefits = Future Benefits × Present Value Factor

Present Value of Future Costs = Future Costs × Present Value Factor

(3)

We calculated Net Present Value using this formula:

NPV = ∑ Present Value of Future Benefits − ∑ Present Value of Future Costs

(4)

If the Net Present Value (NPV) is positive, the research should be undertaken. If the NPV is negative, the research should not be undertaken [36].

2.5. Calculation of Benefit–Cost Ratio (BCR)

The benefit–cost ratio (BCR) is a monetary or qualitative statistic that shows the link between the relative costs and benefits of a proposed project. The cost–benefit analysis formula helps firms compare and calculate which research or task would offer maximum profits with minimum costs involved.

$$\mathrm{Benefit}-\mathrm{Cost} \mathrm{Ratio} (\mathrm{BCR})={\displaystyle \frac{{\sum }_{t=0}^n\frac{Bt}{\left(1+r\right)}}{{\sum }_{t=0}^n\frac{Ct}{\left(1+r\right)}}}$$

(5)

where

• Bt = Gross benefit in the n th year;
• Ct = Total cost in the nth year;
• t = Number of years (1, 2, 3………n);
• r = Interest (discount) rate.

If the benefit–cost ratio is greater than 1, go ahead with the research project. If the benefit–cost ratio is less than 1, one should not go ahead with the research project [37].

2.6. Calculation of Carbon Sequestration

The growth features of the tree species, the local growth circumstances where the tree is placed, and the density of the tree’s wood all affect how quickly carbon is sequestered. Even so, we can roughly estimate the quantity of CO₂ a specific tree sequesters and obtain the annual sequestration rate by dividing it by the age of the tree. This process is as follows.

• Determination of the total (green) weight of the tree.
• Determination of the dry weight of the tree.
• Determination of the weight of carbon in the tree.
• Determination of the weight of carbon dioxide sequestered in the tree.
• Determination of the weight of CO₂ sequestered in the tree per year [33].

A similar approach was employed by Parida et al. (2022) and Yousefi et al. (2017) [38,39].

2.7. Determination of the Total (Green) Weight of the Tree

The following process was used to determine the weight of a tree, depending on its species.

• W = Above-ground weight of the tree in pounds;
• D = Diameter of the trunk in inches;
• H = Height of the tree in feet.

For trees with D < 11: W = 0.25 D²H

(6)

For trees with D ≥ 11: W = 0.15 D²H

(7)

The variables D² and H were extended to exponents above or below 1 depending on the species of tree and the coefficient (for example, 0.25) which was altered. The weight of a tree’s root system typically equals 20% of its above-ground weight. To calculate the tree’s total green weight, the above-ground weight of the tree was multiplied by 120% [40].

2.8. Determination of the Dry Weight of the Tree

The average tree has 72.5% dry matter and 27.5% moisture. To obtain the tree’s dry weight, the green weight of the tree was multiplied by 72.5% [40].

2.9. Determination of the Weight of Carbon in the Tree

The average carbon content is typically 50% of the tree’s total volume. Therefore, we can multiply the tree’s dry weight by 50% to obtain the amount of carbon it contains [40].

2.10. Determination of the Weight of CO₂ Sequestered in the Tree

CO₂ is composed of one molecule of C and two molecules of oxygen.

• The atomic weight of carbon is 12.001115.
• The atomic weight of oxygen is 15.9994.
• The weight of CO₂ is C + 2×O = 43.999915.
• The ratio of CO₂ to C is 43.999915/12.001115 = 3.6663.

Therefore, to determine the weight of CO₂ sequestered in the tree, we multiply the weight of carbon in the tree by 3.6663 [40].

2.11. Determination of the Weight of CO₂ Sequestered in the Tree per Year

Finally, the weight of CO₂ sequestered in the tree was divided by the age of the tree to determine the weight of CO₂ sequestered per year per tree. The simple equation is [40]:

Weight of CO₂ sequestered per year = weight of CO₂ sequestered per tree/tree age

(8)

2.12. Data Analysis

After compilation and tabulation, the data were analyzed statistically according to the study objectives using Microsoft Excel and R Statistical Software (v4.3.1. The obtained data were analyzed using descriptive statistics (mean, standard deviation, standard error, and percentages). For qualitative data (e.g., education, gender), a Chi-square goodness-of-fit test was performed. For quantitative data (e.g., tree weight, carbon sequestration), an analysis of variance (ANOVA) was conducted to determine significant differences between groups. Post-hoc tests, specifically the Tukey test, were applied to quantify the differences and identify specific groupings within the quantitative data.

3. Results

3.1. Demographic Features of the Respondents

Farmers from various age classes ranging from 20 to 85 years were involved in practicing agroforestry in Madhupur Sal forest and were again gender-categorized into male (74.36%) and female (25.64%). Mostly, middle-aged farmers were growing trees with crops as they could easily indulge in labor-intensive activities for diversified cultivation. Very few farmers from the elderly group, those aged more than 50 years, and farmers aged below 35 years were reluctant to practice agroforestry. The willingness of most farmers (88.48%) in the area to practice agroforestry was influenced by their level of literacy, as it enabled them to better understand the benefits and management practices associated with agroforestry systems. The majority of the farmers have medium-sized families (71.79%), followed by large and small families. The size of the families of the farmers does not influence their livelihood choices because most people rely on agroforestry methods for their living, regardless of how large or small their family is. Depending on the demand of the families, the area of the farms was classified, which indicated that most of the farmers (64.1%) hold small-sized farms up to 1 hectare, whereas 30.77% of farmers practice agroforestry (tree planting, sowing crops, maintenance activities, fruit-based agroforestry, homestead agroforestry, etc.) in their medium-sized farmlands. We found that the majority of respondents (47.44%) were highly concerned about the importance of applying sustainable agricultural practices, while 43.59% had intermediate/moderate knowledge, and only 8.97% of farmers were unaware of the fact of sustainability (

Table 2. Demographic features of the respondents (n = 100).

Characteristics	Categories	Frequency	p-Value
Gender	Male	74.36	<0.01 **
	Female	25.64
Age (years)	Below 35	17.95	<0.01 **
	36–50	58.97
	Above 50	23.08
Level of education	No formal	11.52	<0.01 **
	Primary	21.79
	Secondary	41.03
	Higher secondary	15.38
	Tertiary	10.26
Family size	Below 5	12.82	<0.01 **
	5–8	71.79
	Above 8	15.38
Farm size	Below 1 ha	64.1	<0.01 **
	1–3 ha	30.77
	Above 3 ha	5.13
Knowledge of sustainable agriculture	Poor	8.97	<0.01 **
	Moderate	43.59
	High	47.44

(**) = Highly significant

). The concept of different agroforestry practices in the area was prominently accepted by the farmers, as most of the farmers (62.82%) classified the practice as profitable and another 32.05% of farmers found agroforestry as highly profitable, 5.13% of farmers defined it as not profitable, and no participant denied any impact on the economy from practicing agroforestry.

3.2. Economic Perspective of Agroforestry Practices

Different economic features of the rural farmers of Bangladesh shape the growth and maintenance of tree crop-based agroforestry practices. According to the analysis of agroforestry experts on different popular agroforestry practices in the study area, this study focused on the three most prominent agroforestry practices in Madhupur Sal forest of Bangladesh.

3.3. Productivity Analysis of Selected Agroforestry Practices

3.3.1. Acacia–Pineapple–Turmeric–Papaya-Based Agroforestry

According to the economic perspective, this agroforestry system required intense agricultural labor, but the primary input cost of the agroforestry systems was comparatively lower for establishment (

Table 3. Financial cash flow of selected agroforestry practices (* USD 1 = BDT 109.65).

Components of Agroforestry Practices	Total Cost (USD/ha)	Total Income (USD/ha)	NPV	BCR
Acacia–pineapple–turmeric–papaya	2141.45	6212.95	1833.55	1.90
Sal–pineapple–aroid	2150.30	5745.10	1372.28	1.67
Acacia–pineapple–zinger–banana	2150.02	6584.95	2170.66	2.06
p-value	0.08 (NS)	<0.01 **	<0.01 **	<0.01 **

(**) = Highly significant; NS = not significant.

). However, the acacia tree requires high-quality seedlings to build agroforestry systems, so the initial sapling and tree installation expenses were higher. The total input cost of production was 2141.45 USD/hectare, and the total profit was 6212.95 USD/hectare. This agroforestry system’s Net Present Value (NPV) was 1833.55 USD/hectare, and the benefit–cost ratio was 1.90 (Table 3). This agroforestry system is more profitable than the general agricultural system. It also demonstrates the long-term profitability of the agroforestry system for every dollar invested: the system returns nearly double the investment, further reinforcing its economic viability.

3.3.2. Sal–Pineapple–Aroid-Based Agroforestry

As sal (Shorea robusta) is the most prominent species of the forest and Pineapple is widely spread in the area, sal–pineapple–aroid-based agroforestry is one of the major agroforestry practices. Sal can tolerate high temperatures and usually loses its leaves from February to March, with new leaves appearing in April and May. These leaves from sal are used for mulching so that soil can conserve moisture, and the leaves decompose to add soil nutrients. In this agroforestry model, the total production costs and profits were 2150.30 USD/hectare and 5745.10 USD/hectare worth for a Net Present Value (NPV) of 1372.28 USD/hectare, and the benefit–cost ratio was 1.67 (Table 3). In this agroforestry model, the Net Present Value illustrates the system’s profitability over time, while the benefit–cost ratio shows that the returns significantly outweigh the costs, making it a financially sustainable option.

3.3.3. Acacia–Pineapple–Zinger–Banana-Based Agroforestry

The study analyzed nine acacia–pineapple–zinger–banana-based agroforestry plots for a detailed economic analysis and extrapolated the results to hectares. The total input cost of this agroforestry model was 2150.02 USD/hectare, and the net profit was 4434.93 USD/hectare. However, the Net Present Value (NPV) was 2170.66 USD/hectare, and the benefit–cost ratio (BCR) was 2.06 (Table 3). In this agroforestry model, the Net Present Value highlights its strong long-term profitability, while the benefit–cost ratio demonstrates that the system more than doubles the return on investment, confirming its economic efficiency.

3.4. Ecological Perspective of Selected Agroforestry Practices

Species Composition of Agroforestry

A total of 9 tree species that were planted with diverse crops in the agroforestry model were observed, totaling 173 trees in the Madhupur Sal forest. The forest is enriched with sal (Shorea robusta), where acacia (Acacia auriculiformis) species were mostly planted in association followed the availability of teak (Tectona grandis), litchi (Litchi chinensis), and jackfruit (Artocarpus heterophyllus). The average height of the tree was 19.7 ft, DBH (diameter breast height) 4.1 inches, and an age of 7 years for acacia (Acacia auriculiformis). The average height of the tree ranged from 6.5 ft to 54 ft for mango (Mangifera indica), jackfruit (Artocarpus heterophyllus), teak (Tectona grandis), mahogany (Swietenia macrophylla), litchi (Litchi chinensis), pomelo (Citrus maxima), betel nut (Areca catechu), and sal (Shorea robusta) (p-value < 0.01 **). The average age of the tree varied from 4.5 years to 36 years for pomelo (Citrus maxima), betel nut (Areca catechu), teak (Tectona grandis), mango (Mangifera indica), jackfruit (Artocarpus heterophyllus), and sal (Shorea robusta), whereas the average diameter was 1.5 to 13 inches (p-value < 0.01 **) (Figure 2). Different trees showed various levels of carbon sequestration due to tree’s height, diameter, and age. Furthermore, particular tree management techniques influence the rate of carbon sequestration (CS) by urban trees.

3.5. Weight of the Trees

The average green weight of the trees varied from 4.5 lbs for litchi (Litchi chinensis) to 1342 lbs for sal (Shorea robusta) (p-value < 0.01 **), while the average dry weight varied from 3.27 lbs for litchi (Litchi chinensis) to 972.71 lbs for sal (Shorea robusta) (p-value < 0.01 **). The average carbon weighed 1.62 lbs for litchi (Litchi chinensis) to 486.36 lbs for sal (Shorea robusta) (p-value < 0.01 **). The average green weight, dry weight, and carbon weight of acacia (Acacia auriculiformis) were 151.18 lbs, 109.60 lbs, and 54.80 lbs, respectively, mostly observed in agroforestry practices (Figure 3). The green weights of mango (Mangifera indica), teak (Tectona grandis), and mahogany (Swietenia macrophylla) were 120.42 lbs, 230.48 lbs, and 262.69 lbs, respectively (p-value < 0.01 **). Concerning green weight, dry weight and carbon weight, litchi (Litchi chinensis), pomelo (Citrus maxima), and betel nut (Areca catechu) had minimum carbon sequestration and sal (Shorea robusta), mahogany (Swietenia macrophylla), jackfruit (Artocarpus heterophyllus), and acacia (Acacia auriculiformis) showed maximum carbon sequestration.

3.6. Carbon Sequestration of Various Tree Species

The result showed that sal (Shorea robusta) is the maximum carbon-sequestrating tree, and betel nut (Areca catechu), litchi (Litchi chinensis), and pomelo (Citrus maxima) are the minimum CO₂-sequestrating trees (Figure 4). Shorea robusta sequesters an average of 1783.83 lbs of carbon dioxide. However, Acacia auriculiformis sequesters 200.92 lbs; Litchi chinensis, 6 lbs; Swietenia macrophylla, 349.12 lbs; Tectona grandis, 306.31 lbs; Artocarpus heterophyllus, 316.16 lbs; and Mangifera indica, 160.04 lbs of carbon in Madhupur Sal forest (p-value < 0.01 **) (Figure 4). The maximum yearly CO₂ sequestration was 49.80 lbs/year for Shorea robusta and 31.84 lbs/year for Tectona grandis, and the minimum CO₂ sequestration was 4.43 lbs/year and 1.15 lbs/year for Citus maxima and Litchi chinensis, respectively (Figure 4). Acacia auriculiformis sequesters 23.35 lbs of CO₂ yearly. This variation in sequestration rates underscores the potential of selecting high-performing species like Shorea robusta and Tectona grandis for maximizing carbon capture in agroforestry systems. The ANOVA results indicated a highly significant difference in CO₂ sequestration rates across species, with a p-value < 0.01.

4. Discussion

Carbon sequestration plays a crucial role in mitigating climate change by removing CO₂ from the atmosphere and storing it in reservoirs such as forests, soils, and geological formations, thereby reducing greenhouse gas concentrations and enhancing ecosystem resilience. Human activities such as reforestation, soil management, and carbon capture and storage (CCS) further contribute to these efforts, stabilizing atmospheric CO₂ levels and limiting global warming. In addition to mitigating climate change, carbon sequestration provides co-benefits like improved soil health, enhanced biodiversity, and increased agricultural productivity, making it a key strategy for achieving global climate goals [41].

The results showed that Shorea robusta sequesters the most CO₂, averaging 1783.83 lbs, while species like Litchi chinensis and Pomelo sequester much less. Shorea robusta also has the highest annual sequestration at 49.80 lbs, while Citrus maxima and Litchi chinensis sequester as little as 1.15 lbs. The significant variation in sequestration rates, with a p-value < 0.01, indicates that the differences observed between species are highly unlikely to be due to random chance, confirming that tree species selection plays a crucial role in maximizing carbon capture in agroforestry systems. Global efforts can capture approximately 250 to 500 million tons of carbon dioxide annually, with projections suggesting this figure could reach upwards of 2000 million tons per year for several decades [42]. Agroforestry practices significantly boost carbon sequestration and mitigate the impacts of carbon dioxide emissions. For instance, the northeastern maple–beech–birch forests exhibit carbon sequestration rates of 1760 lbs of CO₂ per acre per year for a 25-year-old forest and 3909 lbs per acre per year for a 120-year-old forest [43]. The diameter at breast height (DBH), tree height, and age of trees are key factors influencing carbon sequestration, with larger DBH and tree height generally leading to higher carbon storage, while younger trees sequester carbon rapidly and older trees store more cumulatively [43,44]. Oak (Quercus leucotrichophora) forests demonstrated the highest carbon sequestration potential, ranging from 448.98 to 1123.16 Mg CO₂ per hectare, with soil organic carbon (SOC) stock varying between 64% and 77%, and carbon credit values estimated at EUR 3379.49 per hectare [45].

In Bangladesh, social forestry programs such as roadside plantations have the highest above-ground carbon sequestration rate (165.81 Mg C ha⁻¹), surpassing institutional plantations at 150.00 Mg C ha⁻¹. Other natural forests, such as protected areas, accumulate the highest above-ground carbon at 195.8 Mg C ha⁻¹, while bamboo stands have a carbon stock of approximately 52 Mg C ha⁻¹ [46,47,48,49,50,51,52]. The carbon sequestration potential in agroforestry systems also varies depending on the vegetation type. This diversity highlights the importance of selecting appropriate species for maximizing carbon storage.

Two key challenges could significantly affect the effectiveness of carbon sequestration in mitigating climate change: the secondary benefits of converting agricultural land to forests, which may outweigh the costs, and the potential issue of leakage at national and international levels. These factors could undermine carbon sequestration programs, requiring careful planning and implementation to ensure a net positive impact [43].

The demographic profile of agroforestry farmers in the Madhupur Sal forest area reflects broader trends across Bangladesh, with a high literacy rate (88.48%) and medium-sized families (71.79%). Most farmers (94.87%) find agroforestry profitable, and 47.44% prioritize sustainability, suggesting that agroforestry can improve both income and living standards. Furthermore, studies from Denmark show that agroforestry systems have higher land equivalent ratios (LER) than monocultures, with agroforestry’s gross margins differing significantly across regions [53,54].

In this study, agroforestry practices yielded an income of 6212.95, 5745.10, and 6584.95 USD/ha, with a Net Present Value (NPV) of 1833.55 USD/hectare and a benefit–cost ratio (BCR) of 1.90, indicating the financial viability of these practices [55,56]. The economic returns from agroforestry, combined with its environmental benefits, suggest that it is an attractive investment for sustainable land-use strategies [57,58,59,60].

Despite the positive findings, several limitations should be considered. The sample size, while scientifically determined, may not fully capture the heterogeneity of agroforestry practices across the Madhupur Sal forest region. Future studies should focus on expanding the sample size and incorporating more diverse agroforestry systems across various biogeographical regions to assess the generalizability of the results. Additionally, the use of more precise biometric tools for measuring tree growth and carbon sequestration would improve the accuracy of the data. It is also important to explore the interaction between the environmental and social impacts of agroforestry, particularly for marginalized groups such as women and smallholder farmers, to ensure more equitable outcomes. Furthermore, investigating the role of specific high-carbon sequestration species, like Shorea robusta and Tectona grandis, in enhancing the climate change mitigation potential of agroforestry systems would contribute to identifying optimal species for future implementation.

5. Conclusions

Many agroforestry systems, whether practiced in the tropics or temperate regions, are firmly based on strong ecological principles and contribute to the achievement of many local development goals through the provision of many basic needs and ecosystem services. Our study observed economic upliftment of production from agroforestry practices that improved carbon sequestration. Due to its favorable economic and ecological effects, agroforestry should garner more attention at a national and worldwide level in terms of sustainable land-use techniques. Different agroforestry practices can be introduced according to the ecological and social perspectives in tropical countries like Bangladesh.