Figure 1.Study area location.

Study was carried out in two plant formations, namely: a planted forest located opposite the Dénis Sassou N’GUESSO University, more precisely in the urban commune of Kintélé, on the northern outskirts of Brazzaville, in the Pool Department, Republic of Congo (Figure 1). It is linked to Talangaï by the Talangaï-Kintélé viaduct, bordered to the west by the Djiri river and to the south by the Congo river; a natural forest, the Lisanga forest, located at the very heart of Mbamou Island, which is a sub-prefecture of the Brazzaville Prefecture. It is closer to Brazzaville than to Kinshasa, and a little upstream from both cities [9]. Almost 30 km long and 13 km wide, it occupies a large part of the Stanley Pool, where, after flowing for over 200 km through the narrow valley of the corridor, the Congo River spreads out in a sheet 25 km wide and over 35 km long, bounded to the west by the Kintambo rapids [10].

The forests of Mbamou island and Kintélé are the tropical type, with two seasons: a dry season (June to August) and a rainy season (September to May), with a slowdown in rainfall from December to February. Irregular rainfall during the transitional months of May and September contributes, depending on the year, to extending the duration of the dry or rainy seasons. Average annual rainfall is around 1400 mm, and average annual temperature is 25˚C (Figure 2).

Figure 2.Climograph of the main meteorological station around the study area (average from 2013-2023).

Annual rainfall varies between 1200 and 1500 mm, with rains starting lightly in late September, establishing themselves in October and ending in May. The wettest months are generally April and November, with an average of 200 mm. June to August and part of September are dry [12]. Humidity is always high and the annual hygrometric amplitude is low. The wettest months are December (95.3%) and January (94.4%), while the driest are August (87%) and September (86.6%). Evaporation varies inversely with atmospheric humidity. It has a relative maximum in March and an absolute maximum in September [12].

Study plots were set up in each forest ecosystem. In the planted forest, as in the natural forest, three square plots measuring 50 m × 50 m, i.e. 2500 m², were set up using a double decameter, following the compass directions (North-South and East-West). We then marked out our plots with stakes. Tree species were then marked by number, using chalk sticks in planted forest and a machete to make notches in natural forest to avoid double counting in the study plots. A total of six plots (15,000 m²) were recorded for this study. Diameter at breast height was measured using a forest compass at 1.30 m above ground level for each tree. The scale was to count all trees contained in the study plots with a DBH ≥ 10 cm. The Haglof Vertex Laser GEO 2 was used to measure tree height in both forest ecosystems. All trees have been identified, checked and confirmed by The Missouri botanical garden’s herbarium database, which is the one of world’s outstanding research resources for specimens and information on plants (see http://www.missouribotanicalgarden.org), and The Xycol database (The list of scientific and vernacular woods names: Accessed December 12, 2024 at http://www.xycol.net/index.php?categorie=0&sess_langue=430).

The use of allometric models is a crucial step in the estimation of aboveground biomass [13]. Aboveground biomass was estimated using the standard allometric equation of [14]. This equation was chosen for three reasons: This model takes into account three structural parameters (Diameter, Total Tree Height and Wood Density); trees with DBH ≥ 10 cm are identified and measured for the tropical area; this equation has been recommended by the REDD+ process, used successfully and repeated in several works [15]-[17]. The allometric Equation (1) developed by [14] is presented as follows:

ρ =wood density (g∙cm⁻³);

D = diameter at breast height (cm);

H = height of tree (m);

AGB = aboveground biomass (t∙ha⁻¹).

However, the specific wood density for each tree [18] was provided by DRYAD’s Global wood density database (accessed on September 16, 2024, at https://doi.org/10.5061/dryad.234). Regarding the species for which specific densities could not be found, the default value for specific wood density of 0.5 g∙cm⁻³ was used [18].

Basal Area and Expansion Factor

The basal area of a tree is the cross-sectional area of a tree at 1.30 m (2). Basal area gives an indication of the degree to which the forest has been filled [19]. It is expressed in square meter per hectare (m²/ha). Basal area of a tree were calculated as follow:

where G is basal area (in m²·ha⁻¹), D i is diameter at breast height of individual i at 1.3 m above the ground (in cm), π is 3.14, ha is hectare and n is the number of stems per plot.

Basal area is the area of a given section of land that is occupied by the cross-section of tree trunk and stem at the base. However, the diametric structure was assessed to understand the level of degradation of Kintélé forest ecosystem. To extrapolate the values from the study plots to values per hectare, an expansion factor is required. These expansion factors indicate the area represented by each study plot or sample (3). This standardization is necessary in order to easily interpret the results and also to make comparisons with other studies.

However, the PAST program used includes standard statistical tests. The data of this study were compiled with SigmaPlot v. 10.0 and PAST ver 3.05 statistical softwares. Study area’s location map has been performed using the QGIS ver 2.0 software.

The basal area values for a stand shown in Table 1 indicate that the sum of the parameters encountered in plot 2 appears to be much greater than that of the other two plots (1 and 2). The values in Table 1 show the basal area of the stand in each plot (1, 2, and 3), and plot 2 appears to be larger than plots 1 and 3. The values in Table 1 show that the natural forest has a large basal area. This increase in basal area may be due to the high floristic diversity found in the forest ecosystem of Mbamou Island.

Table 1.Distribution of Basal area of planted forest (Kintele forest) and natural forest (Mbamou Island forest). G is basal area (in m²∙ha⁻¹) calculated for each plot according to ForestPlots (http://www.forestplots.net) and AfriTRON (http://www.afritron.org) protocols; n is the number of sampled trees by plot.

Table 2 shows that the greater variation in biomass among individuals encountered in each plot, the greater carbon stock despite the biomass. Plot 2 has a high biomass (221.7 t∙ha⁻¹) and carbon stock (104.2 t∙ha⁻¹) compared to plots 1 and 3, which have 242.3 t∙ha⁻¹ of biomass and 113.9 t∙ha⁻¹ of carbon, and 206 t∙ha⁻¹ of biomass and 96.8 t∙ha⁻¹ of carbon. These results are certainly due to the reasonable functioning of this ecosystem, which leads to variations in aboveground biomass. Table 2 also shows the amount of aboveground biomass (AGB) and carbon for each study plot. Plot 2 has a significant amount of AGB and carbon, with 311.8 t∙ha⁻¹ and 146.5 t∙ha⁻¹, respectively, compared to plots 1 and 3 with 190.6 t∙ha⁻¹ of AGB; 89.6 t∙ha⁻¹ of carbon and 160.7 t∙ha⁻¹ of AGB; 75.5 t∙ha⁻¹ of carbon.

However, statistical tests show that: the probability of these two forests is greater than P-value; i.e., P-values of 0.96. As P-value is greater than 0.05, there is difference between the two forests in terms of carbon stock. However, it should be noted that from a technical point of view, there is a small but significant difference between the variables. The planted forest has a higher biomass and carbon stock than the natural forest.

Table 2.Evaluation of aboveground biomass in two forest types. AGB is aboveground biomass (in t∙ha⁻¹); AGBC is the carbon stock of aboveground biomass (in t∙ha⁻¹); n is the number of individual tree by plot.

The study shows that aboveground biomass and carbon stock has influenced by trees heigh and dimater. Figure 4(a) shows two phases of change: aboveground biomass and carbon stocks. An increase in biomass quantity and carbon stocks in the small diameter classes (with 0.62 t∙ha⁻¹ of biomass and 0.29 t∙ha⁻¹ of carbon stocks for the 10 cm to 19.9 cm class, i.e., class I) towards the medium classes reaching its peak in the 30 cm to 39.9 cm class, class III, with 290.9 t∙ha⁻¹ of AGB and 136.7 t∙ha⁻¹ of carbon, followed by a decrease towards the upper diameter classes, ranging from class IV with 231.6 t∙ha⁻¹ of AGB and 108.8 t∙ha⁻¹ of carbon stock to class VII with 11.7 t∙ha⁻¹ of AGB and 5.5 t∙ha⁻¹ of carbon stocks. This is linked to a small number of standing trees. Figure 4(b) also shows two phases of evolution: the aboveground biomass and carbon stock of each class. There is an increase in biomass and carbon stock from the small classes (91.9 t∙ha⁻¹ of AGB and 43.1 t∙ha⁻¹ of carbon for class I) to the medium classes, reaching a peak in class III (185 t∙ha⁻¹ of AGB and 86.9 t∙ha⁻¹ of carbon). There is then a decrease towards the higher diameter classes, ranging from class IV (140.7 t∙ha⁻¹ of AGB and 66.1 t∙ha⁻¹ of carbon) to class VII (24.8 t∙ha⁻¹ of AGB and 11.6 t∙ha⁻¹ of carbon). In general, it should be noted that diameter is a key factor. This study shows that diameter has a significant influence on biomass and carbon stock in the two studied forests.

Figure 4.Distribution of aboveground carbon (AGB) and carbon stock by diameter and height class: (a) Variation in AGB and carbon stock by diameter class in the planted forest; (b) Variation in AGB and carbon stock by diameter class in the natural forest; (c) Variation in AGB and carbon stock by height class in the planted forest; (d) Variation in AGB and carbon stock by height class in the natural forest.

Figure 4(c) shows that height class II is the most represented, with 582.8 t∙ha⁻¹ of AGB and 273.9 of carbon stock, followed by height class I with 47.48 of AGB and 22.3 of carbon stock. Height class III is poorly represented in the planted forest. However, Figure 4(d) shows that height class II still contains the most biomass (304.4 t∙ha⁻¹) and carbon stock (143 t∙ha⁻¹), followed by height class III with 241 t ha⁻¹ of AGB and 113.2 t∙ha⁻¹ of carbon stocks, and height class I is less represented (117.8 t∙ha⁻¹ for AGB and 55.3 t∙ha⁻¹ for carbon) in the natural forest. Figure 4(d) shows two phases of change in biomass and carbon stocks. There was an increase in the amount of biomass in the lower height classes (with 47.4 t∙ha⁻¹ of biomass and 22.2 t∙ha⁻¹ of carbon stocks for the 10 m to 19.9 m class) towards the middle class, reaching its peak in class II, from 20 to 29.9 m with 582.8 t∙ha⁻¹ of AGB and 273.9 t∙ha⁻¹ of carbon. The decrease in the amount of AGB and carbon in class III, with 39.8 t∙ha⁻¹ of AGB and 18.7 t∙ha⁻¹ of carbon is remarkable. Figure 4(d) also shows the same phases of change. An increase in the amount of AGB and carbon in the small diameter classes (with 117.8 t∙ha⁻¹ of AGB and 55.3 t∙ha⁻¹ of carbon corresponding to the 10 to 19.9 m class, i.e., class I), in class II, 20 to 29.9 m it reaches its peak (with 304.4 t∙ha⁻¹ of AGB and 143.07 t∙ha⁻¹ of carbon), in class III, 30 to 39.9 m, there is a decrease in the amount of AGB and carbon (with 241 t∙ha⁻¹ of AGB and 113.2 t∙ha⁻¹ of carbon). These results show that the height parameter has a strong influence on biomass and carbon stocks in these two forest ecosystems. This study leads us to understand that the height parameter also influences carbon stocks. The variations in biomass by diameter class and height class in the two forest types studied are remarkable.

This study shows an average density of 210.6 ha⁻¹, ranging from 176 to 236 trees per ha in planted forests, and an average density of 362.6 ha⁻¹, ranging from 220 to 452 trees per hectare in natural forests. The high density of natural forests shows that they have significant floristic diversity. Despite its higher density than the planted forest, the results for the natural forest do not corroborate to [20], which states that in dense tropical forests, densities vary from 450 to 750 trees per hectare. This low density can be explained by the DBH scale, which states that only trees with a DBH ≥ 10 cm should be inventoried. The forest on Mbamou Island is undergoing regeneration following various forms of environmental degradation. In contrast, the planted forest in Kintélé has a low density due to the absence of trees in the plots following uncontrolled logging.

These densities enabled us to determine the population structure of each forest (planted and natural). Figure 4 shows that the number of individuals in the planted forest varies from class I to class III, then decreases to class VII. In contrast, the number of trees decreases from the smallest class (class I) to the largest class (class VII) in natural forests. In fact, there are more small-diameter trees than large-diameter trees in the natural forest, which shows that it is a young forest in full regeneration, unlike the planted forest, which has more large-diameter trees (medium and large).

This study shows that the basal area of the stand varies from 22.7 to 20.4 m²∙ha⁻¹, with an average of 22.3 m²∙ha⁻¹ for planted forest, and from 24.9 to 51.2 m²∙ha⁻¹, with an average of 41.1 m²∙ha⁻¹ for natural forest, which corroborates to [21], which states that the higher the basal area values, the larger the diameters of the trees in the stand, and the smaller the diameters of the trees in the stand, the lower the basal area.

In order to preserve these forest ecosystems, the indirect method was used, employing an allometric equation that highlights three important parameters: density, height, and diameter. The equation developed by [14] enabled us to estimate the biomass and carbon stock contained in these two forests (planted and natural), with 223.41 ± 18.19 t∙ha⁻¹ of AGB and 105.00 ± 8.55 t∙ha⁻¹ of carbon stock in the planted forest. 221.10 ± 80.02 t∙ha⁻¹ of AGB and 103.92 ± 37.61 t∙ha⁻¹ of carbon stock in the natural forest. The probability obtained is 0.96 according to statistical texts, where P-Value > 0.05 means there is a significant difference and P-Value < 0.05 means there is no significant difference. In our case, as P-Value = 0.96 > 0.05, there is a significant difference in the carbon stock contained in these two forests (planted and natural). Furthermore, from a technical point of view, it should be noted that there is a small difference, although the natural forest has greater floristic diversity. However, it should be noted that the carbon stock (103.9 t∙ha⁻¹) of the natural forest is lower than the carbon stock (105 t∙ha⁻¹) of the planted forest due to its biomass. Our results are low compared to the results from [22] in the secondary forest of Pool Department, which obtained 167 t∙ha⁻¹ of stock, and [7] in the Likouala forest, with 190.7 t∙ha⁻¹ of carbon stock.

The different diameter classes observed in the planted forest show that the more biomass increases in relation to diameter, the greater the carbon stock becomes. In contrast, in the natural forest, it can be seen that it is not only the diameter parameter but also the number of individuals that influence the carbon stock. The more the population increases, the greater the stock becomes, while also taking diameter into account. This is why [23] state that biomass increase according to tree diameter.

The height classes observed in planted forests show that height is a parameter that influences biomass. In class II, there is a peak in biomass (582.8 t∙ha⁻¹ for aboveground biomass and 273.9 t∙ha⁻¹ for carbon stock), resulting in a significant variation in carbon stock. However, the drop in the peak towards class III leads us to understand that this amount of carbon stock is due to a low variation in biomass (with 39.8 t∙ha⁻¹), which is necessarily due to a low number of individuals. The height classes of the natural forest show a good variation in biomass and carbon stocks, which shows a perfect balance between the number of individuals and the height parameter. The taller the trees and the greater their number, the greater the carbon stock. The results of the present study therefore corroborate to [24], which states that trees of average height improve the prediction of wood biomass and total tree biomass.

However, uncertainties in forest carbon studies are remarkable by stem from measurement errors in field data (like tree dimensions), sampling errors in plot selection, model uncertainties in biomass estimation, and land-cover mapping discrepancies from remote sensing [25]-[27]. Future changes in forest carbon stocks are further complicated by uncertainty in biomass turnover rates, the persistence of carbon in forests, and the variability of ecological processes. Regarding sampling uncertainty, the inaccurate or unrepresentative field plots can lead to significant uncertainty in estimates of carbon stocks and changes, especially in large-scale assessments. About model uncertainty, the models used to estimate forest biomass and carbon fluxes can be inaccurate due to flawed assumptions, improper model forms, or parameter errors.