Harvesting Criteria Application as a Technical and Financial Alternative for Management of Degraded Tropical Forests: A Case Study from Brazilian Amazon

: This article addresses a case study on the application of criteria for harvesting, aiming at restoration and proﬁtability in a degraded tropical forest in the Amazon. The objective is to provide technical and economic information to promote a truly sustainable silvicultural management system in forests with this proﬁle and turn them into a desirable ﬁnancial asset for conservation and social development. In the forest census, 85.907 trees ha − 1 (100.8566 m 3 ha − 1 ) were inventoried with diameter at breast height ( dbh ) ≥ 25 cm, belonging to 106 commercial species. When applying the harvest criteria, 19.923 trees ha − 1 (29.99 m 3 ha − 1 ), referring to 53 species, were destined for harvest. Some trees were selected by more than one criterion, totalizing 17.985 trees ha − 1 by density, 1.831 trees ha − 1 by compromised health, 0.212 trees ha − 1 by maximum dbh , 18.933 trees ha − 1 by minimum dbh , 1.385 trees ha − 1 by tree stem (quality 3), and 0.080 trees ha − 1 by species conservation. In all scenarios, the application of criteria for harvesting proved to be proﬁtable with excellent cost–beneﬁt ratios. The selection of trees with a minimum cutting diameter of 25 cm in shorter cycles tends to allow the promotion of new commercial species. The set of actions presented has the potential to favor the maintenance of biodiversity and expansion of low-density populations, health and the potential increment of the forest productivity. In addition, it is more feasible for the supply of forest products in a shorter time than provided for in Brazilian regulations; however, they must respect the speciﬁcities of the species and also of the site.


Introduction
The management of natural forests seeks to conserve forest resources and to perpetuate forest production with sustainable feedstock for industry [1][2][3]. However, discussions show the confrontation and difficulties in establishing rational parameters that enable environmental, social, and economic interests in the use of natural forests [4][5][6]. Studies in the region show mainly the ecological character of The climate in the region is Aw (Köpen). Tropical climate with summer rains and average annual rainfall of 2500 mm [36,37]. The average annual temperature is 25 °C [38]. The vegetation type of the municipality is dense submontane ombrophylous forest with emerging canopy [39]. The predominant soils are dystrophic yellow latosol and dystrophic red-yellow argisol [40]. Shet Farm is represented by a degraded natural forest, which is a characteristic of the region of the deforestation arc in the Amazon [41]. Such forest was degraded due to illegal logging activity that took place between the 70 s and 90 s, with no information on the volume of extracted wood. The first legal logging activity in the area occurred in 1993 and 1994, allowed by the Brazilian forestry authorities, with an average harvest volume of 65 m 3 ha −1 . The current regulation that prescribes a maximum harvest of 30 m 3 ha −1 for the Amazon rainforest began only in 2006 [14]. Enrichment planting was carried out sowing paricá (Schizolobium parahyba var. amazonicum) in the clearings formed by logging activity between 1993 and 1994 [33].
A forest inventory was carried out in 2008, in which all trees with the diameter at breast height (dbh, the diameter at 1.3 m in height) ≥ 25 cm were evaluated. The methodology proposed by [42] was used to perform the forest census. The following procedures were performed: botanical identification and vines cut off; attachment of a numbered label at the base of the tree trunk; dbh measurement; visual estimate of commercial height; classification regarding shaft quality (shaft 1-straight and The climate in the region is Aw (Köpen). Tropical climate with summer rains and average annual rainfall of 2500 mm [36,37]. The average annual temperature is 25 • C [38]. The vegetation type of the municipality is dense submontane ombrophylous forest with emerging canopy [39]. The predominant soils are dystrophic yellow latosol and dystrophic red-yellow argisol [40]. Shet Farm is represented by a degraded natural forest, which is a characteristic of the region of the deforestation arc in the Amazon [41]. Such forest was degraded due to illegal logging activity that took place between the 70 s and 90 s, with no information on the volume of extracted wood. The first legal logging activity in the area occurred in 1993 and 1994, allowed by the Brazilian forestry authorities, with an average harvest volume of 65 m 3 ha −1 . The current regulation that prescribes a maximum harvest of 30 m 3 ha −1 for the Amazon rainforest began only in 2006 [14]. Enrichment planting was carried out sowing paricá (Schizolobium parahyba var. amazonicum) in the clearings formed by logging activity between 1993 and 1994 [33].
A forest inventory was carried out in 2008, in which all trees with the diameter at breast height (dbh, the diameter at 1.3 m in height) ≥ 25 cm were evaluated. The methodology proposed by [42] was used to perform the forest census. The following procedures were performed: botanical identification and vines cut off; attachment of a numbered label at the base of the tree trunk; dbh measurement; visual estimate of commercial height; classification regarding shaft quality (shaft 1-straight and cylindrical shaft; shaft 2-slightly tortuous and/or ribbed; shaft 3-crooked, strongly ribbed or forked); and health (rot, senescence, broken, dead, and/or fallen tree top).
The census quantified 46,010 trees and 106 species divided into three market value groups that were defined by the company. For the groups 1, 2, and 3, 13, 41, and 52 species were included, Diversity 2020, 12, 373 4 of 14 respectively. The trees were classified as "for harvest" or "remaining" (next harvest). For harvesting resulted in a total of 10,671 trees of 53 species, with 13 m average height, 37 cm average dbh, 16,067.53 m 3 volume, and production for harvest of 29.99 m 3 ha −1 (Table 1). Table 1. Species (commom and scientific name) listed by market value group, forest management area (535.6 ha) at Shet Farm, Dom Eliseu, Pará, Brazil.

Harvesting Criteria
According to Brazilian regulations for harvesting in natural forests, only trees with dbh ≥ 50 cm can be harvested [14]. To support this research, an authorization for wood extraction was issued by the Pará State Department of Environment and Sustainability (SEMAS-PA) to harvest trees with dbh ≥ 25 cm.
The harvesting criteria were established based on the harvest history of the area, considering trees' silvicultural conditions and the ecological behavior of the species. The criteria were applied together or separately to select the trees. With these parameters, the following criteria were defined for the harvest, in this order: Density: tree harvest of species with higher density in the management area, aiming at the conservation of low-density species and the maintenance of biodiversity. The density was obtained using the equation [43]: where Da absolute density; n i = number of inventoried trees of the i-th species; and A = total sampled area, in hectare. Health: harvest of trees identified with rot, senescence, broken top, signs of disease, or death were selected for harvest. Traditionally, trees with compromised health are not selected for harvest. However, the permanence of these trees influences negatively the quality of the future forest as they are more susceptible to pests and diseases, facilitating their proliferation. Furthermore, trees in these conditions use growth resources (space, light, water, and nutrients) that could be made available for healthy and productive trees. In general, trees with compromised health have a reduced life cycle.
Maximum dbh (>100 cm): harvesting trees with dbh > 100 cm. This criterion aims to increase the population of trees with smaller diameters. The harvest of trees with larger diameters helps to make the future industrial plant suitable for a larger number of trees of smaller diameters compatible.
Minimum dbh (≥25 and <50 cm): harvesting trees with dbh (≥25 and <50 cm) below the one provided in Brazilian regulations was performed. It includes species that, due to biological characteristics, do not reach the minimum dbh recommended by law in the industrial production process.
Tree stem: harvesting trees with tree stem type 3 for the maintenance of trees with tree stem types 1 and 2.
Conservation: all inventoried trees of low-density species (≤0.15 tree ha −1 ) were maintained for seed production, natural regeneration, and species diversity maintainance. On the other hand, the species A. lecointei, C. odorata, C. goeldiana, Copaifera sp., H. courbaril, H. petraeum, H. serratifolius, and M. elata that had been pressured in previous harvests, regardless of density, were then selected as remaining. However, the trees of these species classified by the aforementioned criteria (health, shape shaft, and maximum dbh) were destined for harvest.

Cost-Benefit and Sensitivity Analysis
Net present value (NPV) was used to estimate profitability of criteria for harvesting in degraded forests. NPV is a tool to calculate profitability of projects through discounted cash flow analysis [44]: where B t is the revenues for the sale of standing wood in year t, C t is the total cost in year t, r is the discount rate per year, t is the year when revenue or cost occurs, and n is the time demanded for revenues. Only NPVs larger than zero indicate profitability of the investment [45]. The profitability of the sale of standing wood was calculated based on the 13 years of forest growth, after the last logging at Shet farm. Cash flow was based on field worksheets provided by the Arboris Group, with a total cost of USD 37.82 ha −1 , which is the sum of the annual cost plus the costs of forest census, planning for harvest, and management. The revenues considered for the sale of standing wood is the result of the volume of wood multiplied by the price determined by the business group for the inventoried species, according to the market value group (1, 2, and 3). Based on the values of costs and revenues, a cash flow was created using current values ( Table 2). Scenarios with different interest rates were constructed through a sensitivity analysis to project the profitability of the commercialization of trees authorized for harvest. In addition to the basic rate of 7%, other scenarios were built with interest rates of 4% and 10% per year. These scenarios aim to encompass, through interest rates, possible economic fluctuations in wood-based products and forest production inputs/operations and, consequently, in the cost of production, wood selling price and interest rates, and to project how they can affect the NPV if they occur.
Cash flow was based on field worksheets provided by the Arboris Group, with a total cost of USD 37.82 ha −1 , which is the sum of the annual cost plus the costs of forest census, planning for harvest, and management. The revenues considered for the sale of standing wood is the result of the volume of wood multiplied by the price determined by the business group for the inventoried species, according to the market value group (1, 2, and 3). Based on the values of costs and revenues, a cash flow was created using current values (Table 2). Scenarios with different interest rates were constructed through a sensitivity analysis to project the profitability of the commercialization of trees authorized for harvest. In addition to the basic rate of 7%, other scenarios were built with interest rates of 4% and 10% per year. These scenarios aim to encompass, through interest rates, possible economic fluctuations in wood-based products and forest production inputs/operations and, consequently, in the cost of production, wood selling price and interest rates, and to project how they can affect the NPV if they occur.

Technical Analysis
In the forest census, 85.907 trees ha −1 (100.8566 m 3 ha −1 ) were inventoried with dbh ≥ 25 cm, belonging to 106 species. When applying the harvest criteria, 19.923 trees ha −1 (29.99 m 3 ha −1 ), referring to 53 species, were destined for harvest. Some trees were selected by more than one criterion,   The trees classified by the health criterion represented 2.16% (1.854 trees ha −1 ) of the total inventoried, totalizing 48 species and 1.4032 m 3 ha −1 . Of this total, 98.79% (1.832 trees ha −1 ) were destined for harvest totalizing 1.3943 m 3 ha −1 (99.37% m 3 ha −1 ). The species that had compromised health were more frequently genus of Cecropia, Inga spp., T. burseraefolia, and P. guianensis, representing 67.18% of the population destined for harvest by this criterion.
The population inventoried with dbh > 100 cm was 0.269 trees ha −1 (0.31%), 28 species, and a volume of 3.6182 m 3 ha −1 . According to the maximum dbh criterion, 0.213 trees ha −1 (79.17%) of 20 species with a volume of 2.9250 m 3 ha −1 (80.84%) were destined for harvest. The species C. tocantinum, P. suaveolens and T. burseraefolia, contributed 64.04% of the trees destined for harvest by the maximum dbh criterion.
Among the inventoried population, 76.387 trees ha −1 (88.92%) belong to 104 species and were present in the dbh classes between 25 and 55 cm. When the minimum dbh criterion (≥25 and <55 cm) was applied, 17.657 trees ha −1 (21.7044 m 3 ha −1 ), distributed in 47 species, were destined for harvest. In the middle of the species with the highest frequency of trees destined for harvest by the minimum dbh criterion are genus Cecropia, Inga spp., P. guianensis, S. parahyba var. amazonicum, S. pruriens, T. burseraefolia, and Z. ekmanii, representing 89.77% (16.882 trees ha −1 ).
The conservation criterion inventoried 4.221 trees ha −1 that correspond to 4.91% (7.5167 m 3 ha −1 ) and belong to 13 species classified as having density below 0.15 trees ha −1 and pressed in the previous logging. Of these trees, 4.141 trees ha −1 (7.3532 m 3 ha −1 ) were set as remainings and only 0.080 trees ha −1 were harvested.

Cost-Benefit and Sensitivity Analysis
Cost-benefit analysis indicates the application of the harvest criteria in a lucrative activity, proving that this forest management system is financially viable under the nominal interest rate of 6% per year, since their NPVs were higher than zero in all simulations (Figure 3). trees (17.985 trees ha −1 ) and 80.29% of volume (24.0863 m 3 ha −1 ) of the harvest. S. parahyba var. amazonicum (12.881 trees ha −1 ) and genus Cecropia (12.705 trees ha −1 ) were the species with the highest tree density (29.78% of trees) and the highest harvested volume (58.30%).
The trees classified by the health criterion represented 2.16% (1.854 trees ha −1 ) of the total inventoried, totalizing 48 species and 1.4032 m 3 ha −1 . Of this total, 98.79% (1.832 trees ha −1 ) were destined for harvest totalizing 1.3943 m 3 ha −1 (99.37% m 3 ha −1 ). The species that had compromised health were more frequently genus of Cecropia, Inga spp., T. burseraefolia, and P. guianensis, representing 67.18% of the population destined for harvest by this criterion.
The population inventoried with dbh > 100 cm was 0.269 trees ha −1 (0.31%), 28 species, and a volume of 3.6182 m 3 ha −1 . According to the maximum dbh criterion, 0.213 trees ha −1 (79.17%) of 20 species with a volume of 2.9250 m 3 ha −1 (80.84%) were destined for harvest. The species C. tocantinum, P. suaveolens and T. burseraefolia, contributed 64.04% of the trees destined for harvest by the maximum dbh criterion.
Among the inventoried population, 76.387 trees ha −1 (88.92%) belong to 104 species and were present in the dbh classes between 25 and 55 cm. When the minimum dbh criterion (≥ 25 and < 55 cm) was applied, 17.657 trees ha −1 (21.7044 m 3 ha −1 ), distributed in 47 species, were destined for harvest. In the middle of the species with the highest frequency of trees destined for harvest by the minimum dbh criterion are genus Cecropia, Inga spp., P. guianensis, S. parahyba var. amazonicum, S. pruriens, T. burseraefolia, and Z. ekmanii, representing 89.77% (16.882 trees ha −1 ).
The conservation criterion inventoried 4.221 trees ha −1 that correspond to 4.91% (7.5167 m 3 ha −1 ) and belong to 13 species classified as having density below 0.15 trees ha −1 and pressed in the previous logging. Of these trees, 4.141 trees ha −1 (7.3532 m 3 ha −1 ) were set as remainings and only 0.080 trees ha −1 were harvested.

Cost-Benefit and Sensitivity Analysis
Cost-benefit analysis indicates the application of the harvest criteria in a lucrative activity, proving that this forest management system is financially viable under the nominal interest rate of 6% per year, since their NPVs were higher than zero in all simulations (Figure 3). Maintenance cost of the area is relatively low in the period of forest growth, with 95% of the costs associated with the activities of the year in which the forest census was carried out to sale of standing wood. This fact positively disfavors the other feasible activities for the area which, according to Brazilian regulations, would need at least two more decades to carry out standing wood sale activities.
Scenarios with interest rates below and above the rate of 7% per year showed positive NPV in all simulations. It shows that with a fluctuation in the discount rate between 4% and 10% there is no probability of obtaining an NPV below USD 425.62 ha −1 . This fact weighs in favor of forest owners who have invested in these forest profiles in recent decades but still could not economically enjoy the wood products of the respective forests.

Technical Analysis
It is known that harvesting intervention in intensely pressured forests is one of the factors that influences the composition of future species in the forest, without compromising species diversity, mainly promoting the abundance of pioneer species, with no significant negative effect on shade-tolerant species [3, 46,47]. In this case, the criteria defined for harvesting involve a rational management hypothesis aimed at maintaining the abundance of species in the forest, from the conservation of species and the use of trees with compromised health, senescents, and promoting species of greater and lower value in the forest.
In order to issue a logging authorization in natural forests in the Amazon, Brazilian authorities require trees with dbh ≥ 50 cm to be selected for harvest, directing the use of trees of larger diameter and species of high commercial value. However, this management pressures a reduced number of species in successive harvesting events, reducing the density and dominance of these species and compromising the perpetuation of forest production and abundance [14,[47][48][49].
Some species do not reach the minimum diameter determined by Brazilian regulations (dbh > 50 cm) due to their own natural characteristics. These species are included in the forest productive chain by the destination for logging trees with dbh ≥ 25 cm and with a high population density [14,15,21]. One example is the Cecropia that is part of the plywood panel production chain at Adeco Indústria e Comércio de Compensados Ltd.a. industry located in Dom Eliseu-PA, Brazil [50]. The Cecropia is highlighted for being an important species in the recovery of the forest [51] and traditional medicine as well [52]. Other species such as Manilkara elata, Swietenia macrophylla, and Dipteryx odorata need more time to develop and become economically and ecologically productive [8,10,[53][54][55].
Species submitted to enrichment planting in clearings and S. parahyba var. amazonicum (from this study) tend to form high density at the site [33]. Therefore, the result of forest clearing with S. parahyba var. amazonicum added to the natural regeneration of Cecropia genus specimens, both pioneers with short life cycle, promoted the restructuring and productive viability of the forest in approximately 10 years [21,[56][57][58]. Consequently, it is suggested that these populations are preferably used for harvesting, supposing the maintenance of remaining trees of low-density species and favor the diversity of the forest. Researches carried out with the planting of C. odorata, C. goeldiana, H. serratifolius, M. itauba, and H. courbaril in clearings identified a relatively slower development [59,60]. It takes more time for the recovery process of forest productivity. However, such technique has potential for use in other species that have regeneration difficulties or in cases where density is intended to increase.
The health level of the trees is defined as one of the main causes of tree mortality [61]. Although this loss generaly does not have economic value, it is only evaluated as a priority when there is an importance of nutrient cycling [62]. It excludes the hypothesis that these trees are precursors of diseases that, biologically, could be transmitted to their descendants [63].
In this study, harvesting trees with compromised health is potential alternative to eliminate hosts of pests and diseases and to make available a greater amount of resources (e.g., space, light, water, and nutrients). Is is known that pests and diseases can lead to changes in the composition, structure, Diversity 2020, 12, 373 9 of 14 functions, and productivity of forests [64][65][66]. This criterion represented a relatively low percentage in relation to the total population inventoried (2.16%; 1.854 trees ha −1 ). However, 98.79% of these trees with compromised health were destined for harvest and used in the industry.
The Amazon wealth is characterized by the rarity of species [67], and knowledge about the density per species to define the degree of their conservation is limited [68]. The minimum population density (>0.15 trees ha −1 ) defined in this study takes into account the stages of plant development [69] so that the richness, diversity, and productivity of multiple species are maintained. The conservation of low-density species and those defined as being widely pressured in previous harvests was applied to favor natural regeneration and to expand the population in future harvests, according to the concept proposed of population control. The same could be applied to the tree stem criterion, which indicates that priority should be given to harvesting trees with tree stem quality 3 and senescent in the forest, seeking to improve the quality of the forest based on the concept of population genetics that defines matrix trees to generate descendants with better quality.

Cost-Benefit and Sensitivity Analysis
Successful the application of criteria for harvesting in this study confirms the results from other experiments [24][25][26][27][28][29]. The authors suggest the economic efficiency and ecological potential of applying harvesting criteria to degraded tropical forests, which presents itself as a potential silvicultural alternative for the management of the tropical forest.
The use of silvicultural practices aiming profitability at short, medium and long term is one of the factors that makes it possible to make the management of tropical forests truly sustainable for maintaining wealth and productivity. This study demonstrates the application of technical and economic tools that aim to increase the health and productivity of these forest profiles, contribute significantly to breake paradigms and encourage effective practices of sustainable management of intensively exploited tropical forests.
Applying the criteria for harvesting makes it possible to have financial profitability from the potential conservation techniques of species and the use of trees with compromised health, senescents, and promoting species of greater and lower value in the forest. Furthermore, the application of these criteria for harvesting tends to promote continuous forest production, which is an essential factor for this type of forest to become desirable for conservation.
It is suggested that tree harvesting in Amazon forests that are considered intensively exploited should be treated as silvicultural interventions aiming to maintain biodiversity and timber productivity. Remaining trees could be benefitted from the low canopy height (20 m) of the degraded forest in Fazenda Shet. Lower canopy heights contribute for a larger illumination reaching the forest floor [27]. Dealing with the forest as an environmentally and financially productive structure throughout the interventions process is a promising purpose to awaken the people's desire and motivation to care for and to maintain this standing forest profile.
As a viable silvicultural alternative, the application of criteria for harvesting can be applied in tropical forests under risk of land use changes [29]. Such treatment can provide a high timber production while the treated forest maintains its environmental services. This is particularly important in the arc of deforestation region [21,48], in the Brazilian Amazon, where the experiment was developed. The arc of deforestation, a 500,000 km 2 area in south and southeast Amazon, presents high levels of forest losses and degradation due to decades of uncontrolled logging. Its landscapes are formed by a mosaic of lands covered by crop fields, pastures, and degraded forests.
At this moment, it is necessary to comment on the management of natural forests in the Brazilian Amazon and forests that have been intensively explored in the past. Financial returns, however, are not the only benefit of the application of criteria for harvesting in degraded tropical forests. Fazenda Shet typically represents private areas with degraded forests in the arc of deforestation. It is not part of a public protected area (conservation unit) but a private area under significant deforestation risks due to high pressures of land use changes to more financially competitive activities such as agriculture and livestock.
Under this scenario, the application of criteria for harvesting with a minimum cutting diameter of 25 cm permits shorter cycles (13 years in this experiment) and has the potential to promote degraded tropical forests in a more competitive land use in opposition to the long harvesting cycles allowed in the Brazilian Amazon (30-35 years). In a shorter cutting cycle, based on an assortment of planted species and/or less commercially known species, but which are abundant in the diameter classes ≥ 25 cm, as well as trees of compromised health, of low-quality tree stem and/or for presenting biological characteristics that make them suitable for harvest, depending on the dbh ≥ 25 cm, makes the harvest more profitable. However, specific legal regulations addressed to the management of degraded natural forests are necessary to guarantee their conservation and economic viability.

Conclusions
The case study indicated that the application of criteria for harvesting trees with a minimum cutting diameter of 25 cm in degraded tropical forests it's a lucrative activity with potential conservationist and that proposes to make a more competitive land use in opposition to the long harvesting cycles allowed in the Brazilian Amazon, mainly, for private forests that have been heavily exploited in the past.
The selection of trees with a minimum cut diameter of 25 cm in shorter cycles is proposed, based on the assumption of promoting new commercial species, diversifying the income of forest owners and reducing pressure on species that have been the most pressured in the past.
In all scenarios, the application of criteria for harvesting presented excellent cost-benefit ratios, being little sensitive to discount rates.
The set of actions presented has potential for favors the maintenance of biodiversity and expansion of low-density populations and health of forest. In addition, it is more feasible for the supply of forest products in a shorter time than the provided for in Brazilian regulations.
Degraded tropical forests similar to this case study tend to be able to be subjected to silvicultural interventions in shorter cycles than those recommended by Brazilian regulations, respecting the specificities of the species and also of the location, as long as the harvest is carried out in enriched clearings, regeneration natural of species with high tree density, considering the health conditions and species suitable for harvesting due to biological characteristics.
With the recommendation of potential forest management strategies in the Amazon, it is expected to motivate in a practical way the maintenance of the forest standing, making it a sustainable financial asset from an environmental, social and economic point of view.