QUANTIFYING ORGANIC AND PYROGENIC CARBON STOCKS IN TROPICAL SAVANNA SOIL: INSIGHTS FROM THE CTO-375 METHOD IN THE CERRADO

Understanding the carbon cycle is necessary to understand climate change and the environmental processes involved around the world, considering that the main Greenhouse Gases (GHGs) today are CO2 and CH4. PyC is a residual carbon that, once deposited, can persist in soils as a stable carbon pool. Its quantification is important because this carbon may originate from very ancient fires. In this study, different soil uses were analyzed using elemental carbon analysis (CTO-375 Thermal Oxidation Methodology), determining TOC and PyC stocks, thus verifying whether ICLF systems have higher refractory carbon stocks, compared to other systems. In the results, ICLF and pasture systems showed higher TOC values than the reference area for most of the layers analyzed. As well the PyC values were consistently higher in UP, MP and ILF systems, particularly at deeper layers. Also, the managed systems showed more variability in PyC, in special the UP and ICL systems.The agricultural scenario is crucial for Brazilian economy: according to the Confederação da Agricultura e Pecuária do Brasil (CNA), agribusiness accounted for 23.2% of the country's GDP in 2024. However, when we focus our attention on land use change and its consequences, we see that agriculture and cattle raising were responsible for the removal of 1,969,095 hectares of native forest, or 95.7% of the total hectares deforested in Brazil in 2022 (MAPBIOMAS, 2023). As follows, agribusiness must position itself in such a way as to increase productivity in the same area, reducing deforestation and seeking sustainable technologies aimed at reducing GHG emissions.In this perspective, Integrated Crop-Livestock-Forestry (ICLF) are systems characterized by having livestock, forestry and agricultural activities combined in the same area and are known as a more sustainable production strategy compared to systems based on conventional agriculture and have been widely adopted in Cerrado area (CORDEIRO et al., 2015; OLIVEIRA et al., 2024).According to the literature, soils managed under conservation practices absorb and retain more carbon than those under conventional systems (JOHNSON et al., 1991; ORTIZ et al., 2023; RIVERA et al., 2023) emerging on the current scene as potential sinks. Therefore, comparing different land uses with respect to their refractory and labile carbon fractions makes it possible to identify which conservation systems simultaneously enhance carbon incorporation and stability in the soil.Pyrogenic carbon (PyC) is a highly refractory form of carbon produced by the incomplete combustion of biomass at the land surface; once incorporated into the soil, it can persist for thousands of years, and its abundance within an ecosystem reflects its fire history and its fire characteristics (MASIELLO, 2004; WEI, 2018). Investigating and quantifying PyC stocks in Cerrado agrosystems is crucial for distinguishing the refractory and labile fractions of soil organic matter, thereby clarifying the factors that control its residence time in the soil.This study raises questions about the extent to which the management technology used in agrosystems influences and interferes with the PyC and labile carbon content in this environment. Moreover, adopting ICLF systems aligns with SDG 13 (climate Action) by supporting soil carbon retention and SDG 15 (life on Land) by supporting land restoration and biodiversity conservation.The area where the sampling was carried out covered 1 ha (10,000 m²), a grid was planned with 5 points at a distance of 50 m between them, for each land use. A trench was opened at the central point, measuring 100 x 100 x 100 cm, with 7 depths (three replicates): 0-5 cm, 5-10 cm, 10-20 cm, 20-30 cm, 30-40 cm, 40-70 cm and 70- 100 cm, for each of the land uses. Undeformed samples were collected using Kopeck rings of known volumes (each 5 cm high x 5 cm in diameter) at 3 points on the grid (LEITE, 2023).In this study, different land uses in the Cerrado biome were analyzed at Mato Grosso do Sul: Crop-Livestock Integration (ICL), Integration Livestock-Forestry (ILF), Managed Pasture (MP) and Unmanaged Pasture (UP), were compared to a reference area, a Native Forest (NF). The areas covered in this study have similar soil and climate conditions and 4 depths were analyzed: 0-10, 10-20, 20-30 and 30-40 cm (the average between the depths 0-5 and 5-10 cm was calculated so that a 10 cm analysis could be carried out between the depths).The average bulk density was calculated, TOC and PyC stocks were determined by the CTO-375 Thermal combustion method (GUSTAFSSON et al., 2001). Oxidation of the labile fraction at 370°C for 24 hours with oxygen fluxes in a muffle furnace, followed by loss-on-ignition analysis at 550°C for 1 hour to determine labile and refractory carbon fractions. The average between the three replicates was taken and graphs could then be generated in the excel program with their standard deviation values. The normality of the data was checked by Shapiro-Wilk test, as well the Kruskal-Wallis and Dunn’s (Bonferroni method) tests in R Studio.It was observed that MP peaked, followed by UP and the ICLF systems (3.86 ± 0.21; 3.83 ± 0.58; 3.44 ± 0.94 Mg C/ha), suggesting that the grasses roots favored carbon stock at depth, respectively as figure 1 shows. In the deeper layer (30-40 cm), ICL and NF showed the lowest values (1.99 ± 0.25; 2.03 ± 0.11 Mg C/ha). Kruskal-Wallis (p-value=0.003195) showed a statistically significant difference in TOC stock between at least one pair of land uses.It was noticed that UP and the integrated system ILF (0.87 ± 0.08; 0.85 ± 0.043 Mg C/ha) showed the highest PyC stock values in the surface layer. Figure 2 shows that UP had the highest PyC stock (2.13 ± 0.34 Mg C/ha), in the deeper layer, followed by MP and ILF (1.86 ± 0.03 ;1.47 ±0.07 Mg C/ha).These results may reflect the historical influence of the deposition of pyrogenic residues in production systems. Kruskal-Wallis’s test showed that there are significant differences in PyC stock between at least two land uses (p-value=0.007396) and Dunn’s test showed that only MP vs UP (p.adj = 0.0337) remained statistically significant at the 5% level.The reference area shows a narrow variability, indicating consistency across samples, however, shows the lowest PyC stock. On the other hand, figure 3 shows that pasture systems had the highest median and mean, with UP having the largest variability among samples, meanwhile MP shows moderate one. Moreover, ILF had less variability than ICL but still more than NF.This study highlights the importance of quantifying carbon storage soil due to its link with environmental processes and agricultural practices. The same pattern of increasing carbon stock according to deeper layers was observed for both TOC and PyC for all land uses, compared to the surface layer. Both pasture systems showed higher values for most of the depths analyzed, followed by ILF system. The native forest, the study's reference area, showed lowest TOC and PyC stock values for most of the layers analyzed, but the narrowest variability among the samples. In the future, the elemental Carbon Analyzer will be used for comparison purposes. Furthermore, the results open the possibility of questioning the influence of the management system on the carbon content in this environment and so combining radiocarbon dating to discover the residence time of PyC.