This study investigated the effect of the silvi-agricultural and silvi-pastoral systems on silvicultural data, mean annual increment, CO2 sequestration, basic density, and wood features in Croton floribundus and Guazuma ulmifolia. Our results are significant and can provide valuable insights into the potential benefits of both types of systems. In addition to serving as examples for similar plantations around the world and, at the same time, showing that it is possible to align the production of native Brazilian wood, reducing the impact on native forests. The results show that species growth and volume increase presented different results depending on the system. The values for carbon fixed and carbon sequestered corroborated the MAI, with higher values in the silvi-agricultural system. Management systems did not interfere with wood quality in both species. Species differed significantly for basic density, fiber, and vessel length. There was no difference between species for fiber wall thickness. Basic density and anatomical dimensions varied in the pith-bark direction, where the lowest values are in the position close to the pith, and the highest near the bark. It was found that there were no significant interactions between management, species, and radial position for all properties under study, which demonstrated that there is no dependency between these three factors. The relationships between the properties of C. floribundus and G. ulmifolia wood varied depending on the management system.
Silvicultural data and mean annual increment
Our results indicated a better performance of the trees and consequently of the MAI in the silvi-agricultural system compared with silvi-pastoral system. In Pinus taeda it was observed that trees grown under uniform and fertilized conditions showed the highest growth (height, DBH), productivity (IMA and stem volume) and biomass (stem, trunk, branches, thick and total roots) compared to trees grown in heterogeneous and unfertilized conditions (Kulmann et al. 2023). Wound inflicted by livestock to trees may include branch breakage, trunk breakage, leaves browsing and bark stripping, causing severe damage to the trees, which can lead to plant death and depression in the wood quality (Nicodemo et al. 2010). In our study, both systems received fertilization, however, even though cattle are restricted to grazing in the paddocks, therefore, no barkstripping damage was inflicted by livestock, we wonder if just the grazing may explain the lower performance of the silvi-pastoral system compared to the silvi-agricultural system.
Another possible explanation would be soil variation, since the silvi-agricultural system is in Red Yellow Oxisol, which are soils associated with flat, gently undulating, or wavy relief. They occur in well-drained environments, being very deep and uniform in terms of color, texture and structure in depth; and the silvi-pastoral system is in dark Red Oxisol, which are soils with accentuated red colors, due to the higher levels and nature of iron oxides present in the material originating in well-drained environments, and characteristics of uniform color, texture and structure in depth, however, if the texture is clayey there will be a low amount of water available to plants and susceptibility to compaction (Santos et al. 2018). We suggest that soil type better explains the difference in performance between systems than livestock grazing.
Another suggestion for the differences found in the trees between the system could be the evaluation time of our experiment (4 and a half years) or the species used and their different development. In a study with Eucalyptus trees in agroforestry systems, it was reported that pasture renewal in the first years after system implementation promoted greater initial tree growth, which led to higher values of stem volume and biomass in the system with pasture renewal when compared to system without pasture renewal (Pezzopane et al. 2021). In our study, there was renewal of the pasture with Brachiaria decumbens cv. marandu, plus the addition of maintenance fertilizers, even so the performance of the silvi-pastoral system was inferior to that of the silvi-agricultural system.
We emphasize that C. floribundus and G. ulmifolia are shade-forming species well used in agroforestry systems in Brazil. However, it is important to clarify that C. floribundus is a classic pioneer species (Durigan and Nogueira 1990), while G. ulmifolia is cited as a pioneer (Rozza 1997) or non-pioneer (Ferretti et al. 1995). Which partly explains the volume and MAI results, since the results were different between species for each system (Table 2). It is suggested that in the system with cattle, even though the trunks are far from the animals, they could have access to the tree crown, the most susceptible or most preferred being the leaves of C. floribundus, which could reduce their photosynthetic area and consequently interfere in growth. In silvi-pastoral systems, special attention must be paid to the crown area, due to the shading caused by the tree component over the pasture. In this system, therefore, pruning must be carried out to reduce the shading caused by the trees, with the crown area being a variable to be observed with greater care (Almeida et al. 2022). In silvi-pastoral systems, it is recommended that animals access older trees, with larger dimensions in total height and DBH, as animals can cause damage to smaller trees and thus affect the quality of the wood (Gonçalves et al. 2022).
CO2 sequestration
Here we notice that values for CO2 sequestration corroborated the silvicultural data and mean annual increment with higher values in the silvi-agricultural system. Studies with agroforestry systems are essential to meet the needs of the growing world population, however, the environmental impact of these systems must be considered. In a study with four types of system, it was reported that over four years, the system with livestock and forestry was more efficient, with a more negative net balance C (i.e., C storage) when expressed as ha. Additionally, the association of crops and forestry in livestock systems increases environmental benefits, emphasizing the potential of integrated systems to offset greenhouse gas emissions (Monteiro et al. 2024). For a few decades, especially in the last few decades, the topic of carbon capture has gained importance around the world, due to its influence on climate change. Therefore, specific studies on carbon capture calculations by forests are essential (Arevalo et al. 2022). We understand that more studies must be registered so that wood producers are aware and can price the possibilities that ecosystem services can add to tree plantings.
However, regardless of the results of our study, there is an urgency to increase forest cover. Planting trees continues to be one of the most effective strategies for mitigating climate change, as trees could store many gigatons of carbon (Bastin et al. 2019).
Wood properties
Guazuma ulmifolia presented an average value of 0.36 g cm− 3 of basic density, a value lower than the 0.51 g cm− 3 that was obtained by Sousa et al. (2021) for 10-year-old wood. For Croton floribundus, an average value of 0.42 g cm− 3 of basic density was found, a lower value (0.48 g cm− 3) than that found by Lima et al. (2010). It is widely known that anatomical variations directly influence wood density (Wiedenhoeft and Eberhardt 2021), and that density, a key physical property in wood quality investigations, in turn influences and is used to estimate mechanical properties (Senalik and Farber 2021), wood chemicals and energetics (Rowell 2013).
Some studies evaluate wood quality from species managed by the agroforestry system, mainly using species native to Brazil. In a study on agroforestry system on wood quality of Schizolobium parahyba var. amazonicum, it was found that the timber produced had vessels with a smaller tangential diameter and fewer rays in the juvenile wood, and the wood also showed lower values of basic density and resistance to compression parallel to the fibers, when compared to the monocultural system (Silva et al. 2020).
In another study on agrosilvi-pastoral systems using the species Myracrodruon urundeuva and Peltophorum dubium, it was found that this system can be used successfully, as they produce wood with greater mechanical resistance and better quality for energy (Longui et al. 2021). In a study with Tectona grandis wood from an agrosilvi-pastoral system, it was found that when compared with literature results referring to homogeneous plantations, teak wood from an agrosilvi-pastoral system maintained its technological qualities in terms of basic density, the which is a favorable option for the use of this species in the format of associated plantations (Oliveira et al. 2020). Also, the biomass values found in Cordia goeldiana wood were consistent with the expected characteristics for species used in an agroforestry system (Mascarenhas et al. 2020).
The basic density and charcoal ash content of Mimosa scabrella introduced in an agroforestry system far from its natural occurrence site of the species, show few differences compared to those found for the species in natural occurring sites (Friederichs et al. 2015).
A similar result of radial variation was obtained for basic density, length of vessel elements, fiber length and fiber wall thickness, for wood from trees derived from natural population of Croton floribundus by (Lima et al. 2010). Croton floribundus showed mean values of 1085, 4.66 and 643 µm for fiber length, fiber wall thickness and vessel element length, respectively, with these values being lower than that was observed by Lima et al. (2010), which were 1360, 5.3 and 751 µm, for wood from Croton floribundus trees, originating from natural populations.
For fiber length, it was found that in the silvi-agricultural system Guazuma ulmifolia presented the highest values, differing significantly from the lowest values of Croton floribundus. However, in the silvi-pastoral system, the species did not differ significantly for this variable (Table 4).
In the case of fiber wall thickness, in the silvi-agricultural system, it was found that there is no significant difference between species (Table 4). But in the Silvi-agricultural system, Guazuma ulmifolia presented the highest values and Croton floribundus, the lowest, and these values differed significantly from each other (Table 4). When compared with other works with native species from Brazil managed by the agroforestry system, we must consider the fact that the wood analyzed in this research is considered still young (4 and a half years), and this is a factor that has a lot of influence on the quality of the wood produced for certain uses.
A similar result was verified for the fiber length of Ocotea porosa wood, where variation occurred in the direction of the bark pith, where the fiber length was smaller close to the pith, tending to increase towards the bark region (Vivian et al. 2021).
For the species Caryocar brasiliense, there is a positive correlation between basic density and fiber length (Abrahão et al. 2020). Therefore, we can consider that the relationships between the wood properties of the two species varied depending on the management system.
In general, it was found that we have few studies that assess the quality of wood produced by agrossilvipastoril systems. The results are still incipient, and many factors still must be considered to define the ideal management that produces wood in quantity and quality managed. Incorporating trees into agricultural landscapes represents a promising approach to sustainably supply goods for society, simultaneously enhancing biodiversity, ensuring animal welfare, and generating profits for stakeholders (Kruchelsk et al. 2021).
In summary, agrossilvipastoril systems have shown to be promising in several aspects, being possible to successfully combine fast-growing shrub species, resistant grasses consumed by animals and tree species that can produce different products with a cycle of more than 50 years, for example, ornamental plants, hearts of palm, landscaping, fruits, seeds, bee pasture and wood with different properties (Longui et al. 2021).