We hypothesize that (1) older and narrower fragments (with greater edge effect) will present higher C stocks than the younger and larger ones and that (2) even the very young reforested areas (10 years-old) will already be able to increase carbon stocks in the soil in relation to the levels of the surrounding sugarcane areas, despite of its size.
To answer these questions, we begin with the hypothesis 1 discussed below.
Despite the higher values of carbon concentration in the soil found in the older (Fig. 8) reforested areas (20 years) and the higher stocks of C in the litter in the narrower planting strips (Fig. 6), which are in agreement with our first hypothesis, the synergy of these two factors, as expected to be expressed by significantly higher overall C stocks in Noboro forest, was not observed. The large variation among intra-site plots reflects the functional heterogeneity of these fragments of new riparian forests undergoing the first two decades of restoration. In fact, the studied sites varied in a combination of factors (see Tables 1 and 2) capable of affecting carbon sequestration and storage in different ways in riparian forests, creating a mosaic of conditions and possibilities of compensatory or confounding effects with those of aging and width of the planted areas (Dhiedt et al. 2021, Nunes et al. 2018, Culot et al. 2017, Magnago et al. 2017, Chaplin-Kramer et al. 2015, Berenguer et al. 2014). Another consideration to be made is that if the different carbon stocks evaluated in this study have the same sensitivity to the time scale (10 and 20 years of planting) and to the small-scaled edge effect (30 and 100m planting width). It was also reported by William et al (2019) that older forests continue to accumulate carbon in the soils. As trees get older, they absorb more carbon every year, and because they are bigger, they store more carbon. He adds that preserving existing mature forests will have an even more profound effect on slowing global warming in the coming decades, since immature trees sequester far less CO2 than older ones. Functional groups of tree species distribution, as the proportion of pioneer trees, can also influence the C sequestration as shown in Table 2 (Forrester et al. 2017). Another explanation for the patterns found in the accumulation of C in the aerial biomass in the studied areas may be related to the edge effect, since in smaller fragments there is greater entry of light, facilitating the establishment of pioneer species that normally have higher densities but a reduced stock of carbon per hectare (Magnago et al. 2017). In fact, the highest percentages of pioneers were found in the reforestations Noboro and Santa Barbara, which were 30m.
According to McKinley et al (2011), forests of different ages play different roles in removing carbon from the atmosphere and storing it in the wood. Old forests accumulated more carbon than younger or pioneer forests; however, young forests grow more rapidly, sequestering carbon at a faster rate than older forests, but also showing faster turnover. Also, pioneers compete greatly, they are very hyper active and carryout photosynthesis much often than older forest thereby accumulating or increasing carbon stocks. By this; loss of leaves, reduction in root strength, thin leaves and roots, tissues become less resistant to sun and rain. Thus, trees-sites of different ages can stock almost equal amount of carbon overtime. Trees shed leaves when CO2 uptake is low to support photosynthesis. When photosynthesis is equal to respiration, plants maintain its position. At light intensities below this point, more respiration occurs and at higher light intensities, more photosynthesis occur and the trees becomes more active and productivity increases. The photosynthetic compensation point is where light intensity is at the point where the rate of photosynthesis is equal to the rate of respiration. For shade-adapted plants, the compensation point is lower - their rate of photosynthesis will exceed the rate of respiration at lower light intensities than the plants adapted to sun. Moreover, as reported by Ash & Helman (1990), similarities in carbon biomass stock between age and width of restored forest can be attributed, in part, to differences in past logging activity and time since last disturbance. Considering that all areas had the same past, they donated soil for the construction of the dam, their different levels of protection after planting and management may have increased the variation in carbon sequestration between areas, potentially explaining our results. The 20-year-old Noboro fragment was the one that received management after the first planting and was more protected against the entry of people, showing an average stock of aerial C that was twice as large compared to younger fragments. In contrast, Delta had the highest levels of disturbance, due to the continuous presence of people, in addition to being younger, a combination that led to the lowest C stock value.
Other factors must still be considered. According to Dybala et al (2018) and Melo & Durigan (2006), narrow strips planted around reservoirs in soil conditions with high nutrient availability may suffer less from competition for light and water. As a result, young planted forests can exceed C stocks in relation to natural riparian forests or upland forests.
The differences we found in the litter stock were that areas of narrow width produce more litter than those of higher width. This may be attributed to the higher amount of stand density found in narrow width. Higher stand densities (e.g. trees per hectare) are associated with smaller size due to increased competition for physical space or available resources such as light, nutrients and water (Long et al. 2004). Also, tree height may be another factor according to a similar report by Ruiz-Jaen & Potvin (2011) for natural tropical forests in Panama, and it is not surprising considering that height is a good predictor of total biomass of the plants, which directly influences the amount of C contained in both the above and below-ground portions of the standing vegetation and incorporated into the soil as litter at aging. We found that soil carbon concentration was higher in older ages of 20years than younger ages of 10years and no difference in the stock. This was as a result of higher litter and organic matter decomposition in the soil overtime. Soil organic matter (SOM) is formed from decomposing biomass and increases the water holding capacity of the surface. Also, root and leaf litter production and the accumulation of coarse woody debris might be highest in old-growth forests (Pregitzer and Euskirchen 2004).
The difference observed in litter stock at different widths can be explained, since the highest productions were verified in the areas of greater widths (Martins 2016), but we did find significant differences, this shows a greater decomposition rates in the larger-width fragments. Areas with higher width produce more litter decomposition comparing to those with lower width because of the residency and transition period. It allows the area to rejuvenate itself and also nutrients thereby allowing or helping in more new establishment and thus increase in productivity in wider areas.
The 0–10 cm depths differed from the 10–20 and 20–30 layers, with the highest stocks being verified at the lowest depths. This pattern was already expected, since the nutrient cycling process predominates in the surface layers (Schumacher et al. 2003). Moreover, plots located near the lake were also at a lower altitude (Sämuel et al. 2011). Greater biomass productivity and carbon inputs are expected to increase soil carbon stocks in older forest fragments. As the litter is decomposed, some of the C is emitted as CO2 into the atmosphere, and some is incorporated into the soil, increasing stocks. This increase in stocks has been observed in experiments in other countries, under various climate and soil conditions and this was similar to reports found in Atlantic forest and Cerrado areas. Coelho et al (2017) found that natural edges of gallery forests have invariable number of individuals and basal area between natural and anthropogenic edges, revealing that natural edges are also prone to resource limitation and stressful conditions (e.g. soil fertility, moisture and fire).
We expected soil carbon stock growth to be influenced by climate and previous work has identified differences in the effects of land use and land cover change on soil carbon stock by climate, soil depth and short-term decreases in soil carbon stocks have been attributed to soil disturbances caused by planting (Laganière et al. 2010). Further, biomass and soil carbon stocks are expected to be larger in riparian forests in wide, complex floodplains with frequent inundation (Sutfin et al. 2016).
The estimated values of C stock in the soil of the new riparian forests studied demonstrate that even in their first decade of growth, such reforestations are capable of significantly increasing the baseline of the surrounding agricultural areas, dominated by the cultivation of sugarcane. The corroboration of our second hypothesis indicates that active restoration can be an essential tool to accelerate carbon sequestration in tropical riparian environments. Although implantation and maintenance costs are in general higher when compared to those of passive restoration the relationship with the benefits may be more advantageous.
Similarly, Silva et al (2007) reports low carbon stocks in sugarcane areas. The cultivation of sugarcane for nearly 50years following deforestation at the São Martinho farm resulted in a decrease in carbon stocks, compared to the soil under native or restored vegetation. Similarly, the measured soil carbon stocks in areas with more than 50years of continuous sugarcane cropping in Hawaii were 12–26% smaller than in adjacent native forest areas. Silva et al (2007), describing a sugarcane chronosequence study, observed a sharp decrease in soil carbon soon after the deforestation and planting, followed by a slow recovery in the soil C concentration.