Constraints to Tropical Forest Conservation and Successful Monitoring and Assessment of Land Uses Cover and Change: Do the Forest Definition and Administration Really Matter?

We examine the constraints to conservation and to successful forestlands’ monitoring/assessments in central Togo through GIS spatial analyses and through a critical overview of the current forestry administration’s model. The major findings are that the land classification based on few inventory parameters cannot substitute for “what forest is”, rather these inventory parameters constitute a mean to sound forest management and conservation when relevantly decided. Also as these parameters measured from satellite imagery are supplemented by continuous fine management data they may consistently contribute to the classification of the vegetation cover. This helps to suggest that solution to forest degradation/deforestation, and monitoring/ assessment requires data refinement through local forest management. Else, the actual forestry administration is local communities and indigenous people’s needs biased because it has been negligent of the cultural forestry practices, the major constraints to conservation and the monitoring/assessment of forest lands. As a common pool-resource, the questions relative to forest cannot be addressed at a single environmental concerns level. Interests are multiple and various along the spectrum from the global environment level to the local environmental level that should be accounted for. Thus we recommend a reconsideration of the forestry administration model. What is required are simple policies processes to define forest management plans that promote simultaneously sustainable forest management while accounting for any stakeholder concern, importantly the cultural forestry that addresses specific local communities and indigenous people’s forest related interests.


Introduction
The monitoring and assessment of tropical forest land have been of serious issues for long. The major reason specific to poor countries is the lack of reliable data on the location and the extent of the forest land. The situation is due first to the fact that the forest literature was not provided by the forest professionals, rather the information on forest was quite documented from travelers' books which are unfortunately often biased, exaggerated, or inaccurate as acknowledged by Richards [1] and Longman and Jenik [2] because some authors have been too quick to generalize, and some others have traveled by the rivers and roads where the observed vegetation cover may not represent a real forest ecosystem. Prior to Richards, Chevalier [3] reported that reports from some previous authors encompassed a lot of fantasies. Another reason is as would remark some social scientist authors such as Fairhead and Melissa [4], pioneer scientists in West Africa have systematically applied the assumptions and techniques elaborated during the colonial era to the actual forest conditions. These false assumptions and techniques have blurred our understanding of tropical forest ecosystems functioning. This uncertainty around the forest data has remained unchanged after decades since forest issues had been of concern. According to our investigation, the earliest field works regarding the vegetation cover in Togo as well as in many other countries of the region, are that of Aubreville's from 1936 to 1949 and Hutchinson's in 1936, reported in Lamouroux [5]. Unfortunately, these works are not easily exploitable because of their mapping low resolution. Recently, the confusion on the extent of the forest land areas as well as that of their changes has been exacerbated by the changing in the assessment methodologies and in the forest related definitions. Data, either from a source to another or from the same source but collected at different time periods contradict each other. For instance, in Togo the forest land assessment using the FAO forest classification yielded 384,880 ha [6] [7], while the national classification provided a figure of 1,392,360 ha [8], about four folds the figure issued from the classification based on the FAO's forest definition. All this seems to confirm the fact that the magnitudes of the tropical forests today, their location, and their various changes remain unknown in most cases [9]. Chazdon et al. [10] after providing historical overviews of forest definitions and concepts linked the inconsistency in the forest land assessment results to the changing in the forest definitions. Previously, Sasaki and Putz [11] recognized that the definition default has led to forest degradation and affirmed that an adequate definition of forest will promote the switch from degradation to responsible forest management,  111001 3 Natural Resources which will help mitigate global warming while protecting biodiversity and contributing to sustainable development. Rather this article aims at showing that the discrepancy in the forest assessment results and the inability to conserve tropical forest lands are the consequences of our current forest administration or governance. The deforestation and the forest degradations may be induced by the inadequate choice of the inventory parameters; but the negligence of the cultural forest practices by the current forest administration has more to do with these issues. Specifically, the first part of the article shows that the forest inventory parameters are tools for sound decisions on forest management, thus, must primarily be forest management objective oriented. Defining what forest ecosystem or other types of vegetation covers are, requires additional information notably, the ecosystem species structure and composition, the life history of the site, the mean biomass periodical increment, and so on. Still in this part the article discusses the effects of the inventory parameter thresholds on forest ecosystems.
When properly set these inventory parameter thresholds help to reach sound decision regarding both forest management and conservation, but constrain them when poorly defined or decided. In the second part, the article reviews the actual forestry administration and shows that it has been biased of the local people needs because negligent of the cultural forestry practices. Issues such as agricultural land expansion, wood supply (importantly firewood and charcoal), and many other forestry subsistence practices, that are well documented as factors of deforestation and forest degradation, have not been seriously accounted for in our forestry policy definition. We have focused solely on global warming mitigation as the "common interest". As a common-pool resource the interests on forest must be multiple and embedded on one another from the local users group to the global ones.

Materials and Methods
The study area is the Central region of Togo. Administratively, it is originally divided into five prefectures which are Blitta, the Plain of Mô River, Sotouboua, Tchamba, and Tchaoudjo. The data for the spatial data analyses are obtained from the global forest dataset made available by the University of Maryland. As presented in the technical notes [12] the global dataset is made of 10 × 10 degree tiles, consisting of seven files per tile. to include all the classes considered in the FAO's classification [13], we reset the canopy closure to 10 classes for the mapping. These classes include the percentage canopy closure above 40, the percentages comprised between 10 -40, 5 -10, and finally the percentage canopy closure below 5. The area in hectare for each class (e.g., the percentage of canopy closure ranging from 0 to 100) was calculated using the ArcGIS ZONAL GEOMETRY

Results
The spatial data analysis The Hansen/UMD/Google GIS data have allowed the mapping of the vegetation cover of Central Togo ( Figure 1) and estimating the areas of the vegetation cover (    Source: UMD-Hansen-Google data. *The areas in square meters represent the areas on the map; the ones in hectares represent the areas in the field. These latter are obtained using the mapping resolution of one kilometer. represented throughout the country by five regional offices for DEF, and four regional offices for ODEF. The DEF operates at local levels through the prefec-  [19]. The program has allowed the country to set in place a national monography on biological diversity and a strategic document on the conservation and management of the biodiversity.
It is clear that the REDD+ and the biological conservation programs are the implementation of the international forestry programs whose major objectives are described above. Ivanova [16] acknowledges the UNEP's achievement for each of the mandates it had assigned. Unfortunately, the international forestry programs do not aim at implementing the programs and activities at national or local levels, this constitutes the responsibilities of the countries [15]. The UNFCCC concerns for local communities and indigenous peoples in Agenda 21 [20] were reiterated at the Conference of Parties in 2017 [21]. The most relevance requirement of the COP 23 towards the countries member of REDD+ is to promote the sustainable forest management. One needs to acknowledge that the sustainable forest management does not exclude the traditional role of the forests, rather it must be reinforced, or an alternative that substitutes for this traditional role must be identified and implemented. As we will raise it in the discussions, it is obvious that we cannot successfully implement forestry programs without accounting for local people's interests or needs.
Unlike the CBD and the REDD+ programs the PAFN addresses the forestry issues in a holistic manner regarding its assigned challenges discussed above. But our investigations during this current study reveal that none of these challenges has been realized. For instance both GIS data and panel data [22] analyses are concordant that the deforestation is still ongoing in Central Togo, the country still relies on import to satisfy its wood consumption [14], and finally, the poverty level in the region is the second highest of the country [23]. There has not been an evaluation of the updated 2011 PAFN, but the analysis of its performance from 1994 up to 2011 helps to identify what had hindered its implementation, which concern will be reintroduced in the chapter reserved to the discussions.

Discussions
Forest definition versus the inventory parameters' threshold This subject has been wildly discussed in the literature [11]; and [10], but we have reintroduced it in here to show that what is required is not to improve forest definition, but rather a sound and responsible forest management practice.
Therefore we discuss here the effects of limiting forest land classification to one inventory parameter (or very few) to define forest ecosystems, and the reverse effects of wrong setting the inventory parameters' thresholds on forest ecosystems.

Forest inventory parameters and forest definitions
The forest land area data from Central Togo (Table 1) is recapitulated below ( In fact, recent definitions of forest stress on the inventory parameters as would outline Chazdon et al. [10]. Importantly these parameters are the canopy ground cover, more than 10% -40% coverage or 10% -30% coverage depending on whether one adopts the FAO or the UNFCCC criteria; the minimum diame- for the management purposes. For instance, thinning a stand to a number of trees (e.g., stocking) helps the forester to make sound decision about how to treat the stand [24]. Setting the parameter threshold should be understood as Natural Resources either to optimize the stand production in the case of forest stand management, or to preserve the integrity of the functioning of forest ecosystems. Therefore, like in the traditional sylvicultural practices setting the tree crown closure threshold should aim at reaching the management objectives such as timber management, conservation, climate change mitigation, and earth stewardship as outlined by Chazdon et al. [10]. Attempting to harmonize these parameters' threshold to reach different management objectives is also another source of confusion. The timber manager emphasizes on the tree stem quality and therefore prefers stand with high tree density. The wildlife manager emphasizes on the quantity and quality of the fodder thus manages the stand with abundant grasses and herbs, thus stands with lower tree density. An agro forester manages stands with different patterns of combination of perennials and crops. The forest manager for carbon sequestration seeks to store as much as biomass in the stand, no matter the quality of the wood. All these types of land uses are present during a forest land assessment and cannot be separated using just one or few inventory parameters' threshold.
To decide if an ecosystem is a forest ecosystem or another type of vegetation cover requires the knowledge of additional vegetation cover characteristics principally the stand species structure and composition, the life history of the site, the mean annual biomass increment, and so on. Therefore a forest ecosystem can be defined in respect to its climax prior to the disturbance. We thus agree with Richards [25] definition of tropical rain forest as: "Evergreen, hygrophilous in character, at least 30 m. high, but usually taller, rich in thick-stemmed lianas and in woody as well as herbaceous epiphytes." The same author when relaying Chevalier [3] defines a savanna forest as: "more or less leafless during the dry season, rarely evergreen, xerophilous in character, often much less than 20 m.
high, park-like, very poor in underwood, lianas and epiphytes, rich in terrestrial herbs, especially grasses." The definitions stress on biotic parameters because any other conditions necessary to define an ecosystem including both edaphic and climatic ones are assumed to be fixed and thus cannot be altered as long as the disturbance is less than the threshold. Also, because of its cybernetic or self-regulatory behavior, following a minor disturbance, a forest ecosystem would continuously tend to return to the initial state (if a climactic forest), or initial trajectory (if an ingrowth or a secondary forest) [26] [27]. Thus, the discrepancy in the assessment results is to be attributed to land classification based on few inventory parameters' thresholds which depend on assigned management objectives, but not on the changing in forest definitions. These inventory parameters' thresholds are just management tools when properly developed and implemented helps to reach the objectives of forest management. New classifications using new inventory parameters' thresholds will emerge as long as new policies are developed to address emergent issues. Therefore, a constant and consistent definition (identification) of forest ecosystem, for efficient monitoring and assessment, requires forest management practices that continuously make

The reverse effects of a wrong setting of the inventory parameters' thresholds on forest conservation
In fact, there is evidence in the Central Togo of the existence of land without forest, land where the vegetation cover is a mixture of forest and the savanna (e.g., fragmented and degraded forests), and finally the existence of closed forest ( Figure 2). From an historical perspective, the valleys of the Atakora chain of mountains which used to be forested lands three decades earlier are currently occupied by villages and farms. It is unsure the forests will revert at the abandoning of farming because the erosions have carried away the topsoil leaving in place the bare rocks.
From this example we explore the extent at which these inventory parameter thresholds induce or could induce deforestation or forest degradation. Often countries or policies decide to increase or reduce such or such inventory parameter threshold to optimize some particular goods or services from the forest [10]. Such practices not only lead to the discrepancy on the assessment results discussed previously, but also are they risky for most tropical countries where wild fire, illegal tree cutting, and water and soil conservation still constitute important challenging issues. The wrong setting of the threshold leads to reverse effects on our conservation effort. For instance, setting the crown cover threshold very low may appear as legitimating wood cutting for illegal cutter because he thinks he still has the right to cut as long as the threshold is not attained. The   associated consequences of a heavy removal of the vegetation cover from a landscape are wild fire in area prone to fire, or to water and soil conservation difficulties in the steep sloped mountainous areas wildly discussed in the literature [11]. In fact, from our study area, the Assoukoko forests ( Figure 2) appear homogeneous, but visits to the site show that the understory of most of these forests is occupied with cocoa and coffee farms. Still these forests deserve to be classified as closed forests because the canopy cover is over 40 percent (e.g., FAO classification) or 30 percent (UNFCCC classification). We need to understand that tropical deforestation and forest degradation are issues at high stake because the observed effects cannot be specifically attributed to the shifting in the classification methods; they are cumulative effects of many deforestation and forest degradation factors. To provide a holistic recommendation on deforestation and forest degradation, all these factors must be accounted for. The most important factor that constitutes the next of the concerns in this article is the administration and the governance of the forest resources.
Conservation constraints relative to the forestry administration In this sub-chapter we present the forest resources as common-pool resources and show that their administration cannot be dealt with at one particular level of environmental concerns (e.g., global level as opposed to regional and local environmental concerns).

Forests as the common-pool resources
Our definition of a common-pool resource is borrowed from Ostrom [28] [29] "A common-pool resource, such as a lake or ocean, an irrigation system, a fishing ground, a forest, the internet, or the stratosphere, is a natural or man-made resource from which it is difficult to exclude or limit users once the resource is provided by nature or produced by humans. One person's consumption of resource units removes those units from what is available to others. Thus, the trees or fish harvested by one user are no longer available for others." Ostrom (1994, p3-49;and 1998, p4) To show the common-pool resource character of the forest, we categorize the forest users into three groups (Table 4), even if there may exist many intermediaries along the spectrum.
These groups include: 1) the international communities such as the United Nations, concerned with global environmental issues specifically, global deforestation, the climate change, and some other diverse conservational issues; 2) the Regional community institutions which are concerned with the regional environmental issues like biodiversity conservation, desertification, soil and water conservation, and so on … and finally; 3) the last group, or the forest dwellers or the local people concerned mostly with cultural or subsistence forestry practices.
It is important to realize ( Table 4) that the lines between the users groups are very difficult to draw. The groups may overlap in terms of sharing the resources or services, which is a characteristic of the common-pool resources. It is equally  (Table 4), and the definition of the user groups are not presented here in an exhaustive way. We just intend to show the common-pool resources character of the forest resources (e.g., forest goods as well as services).
The current forestry administration and its flaws In chapter 3) relative to the presentation of the analyses results, we presented some efforts undertaken to save tropical forests. Here we discuss the factors that have impeded the successful implementation of these plans, programs, and projects. A close look on the forest administration in Togo (Table 5) reveals that there are no real incentives available to local users of the forest regarding forestry  policies' implementation besides anniversary celebrations (e.g., Arbor Day and Earth Day …), prizes distribution, some training workshops, and seedling distribution to landowners. The common policy instruments are the regulations and policing to limit access to the forest lands. Even at the national level, significant incentives are very few and available just for limited projects and programs like the management of the Haho-baloe forest plantation's ITTO project, the project to promote sylvicultural practices in Yoto, and some NGOs initiatives [14]. We recapitulate Table 5 saying that the failure to forestry plan, programs, and projects implementation is associated to various sources including socioeconomics, institutional, socio-politics and technical constraints.
At the international level we also recapitulate here ( Figure 3) the mandates of the international forestry program discussed earlier. Since the influential article by Garrett Hardin [30], "the tragedy of the commons" has been used as a metaphor for the problems of overuse and degradation of natural resources including the destruction of many fisheries, the overharvesting of timber, and the degradation of forests. Many policy analysts, scholars, and public officials agree with Hardin's conclusion that the participants in a commons dilemma are trapped in an inexorable process from which they cannot extract themselves. External authorities are presumably needed to impose rules and regulations on local users since they will not do this themselves (Ostrom, 1998, p1).
Ostrom [29] had already contended the central direction governance model's prescription asserting that the theoretical foundations of the model are wrong.
The results from our conservation policies constitute evidence to the visionary position of Ostrom. The actual forestry administration seems not to be able to account for the interest of all. For instance, factors such as agricultural land expansion, fuel wood supply, branches and other parts of the trees collection for various uses, forestry for divinities worshiping, are well documented as major agents of deforestation and forest degradation [11] [31]. These practices for subsistence constitute important preoccupations to individuals who face these problems in a daily basis, but our forestry policies have given limited consideration to these issues urgency. It is necessary to acknowledge that the local user of forestry resources must have an incentive to abandon the traditional practice and adopt an introduced innovation [29], that is, the conservation.
Else, the proponents of the central direction administration model [30] [10] also raised the point that countries had changed the inventory parameter thresholds in their national legal forest classification to benefit more funding from the Clean Development Mechanism (CDM) program. Clearly, in addition to the difficulty of dealing with the multiplicity of the interests around the forest resources, the actual forestry administration faces the concern of the commons dilemma. Like in agricultural goods' supply to the market, the forestry program implementation also suffers from the K. A. Bassan et al. distortion in the implementation of its program. The question of how to make the financial resources available for the implementation of plans, programs, projects and activities remains issues of serious challenge. Even if these two issues, the common interest and the commons dilemma are all of great interest, we prefer not to elaborate further on the latter one regarding the scope of this study.
The question around the implication of all stakeholders in conservation processes constitutes a preoccupation of the forestry administration because of its concern for indigenous people and local communities. But, we need to clearly understand that the dispositions included in the United Nations Declarations on sustainable development [20], principally in Chapter 3.7a and 3.7d, and in to define forest management plans in link with site specific stakeholders' interests, importantly that of the most dependent on forest resources, are highly required. Models exist in the fields of social sciences (e.g., the self-organized resources governance systems of Ostrom) [34] as well as in biological sciences (e.g., hierarchical concept of an ecosystem of O'Neill) [35], that could serve to successfully address the concerns. Instead of being conflictual, the interests need to be connected and embedded into each other through hierarchical patterns.

Conclusions
In this article we assess some constraints to tropical forest conservation and to the successful monitoring and assessment of the land use and cover change, principally the constraints relative to forest definition of the inventory parameters' threshold, and to the governance. The methodologies to study the con- The vegetation cover classification was based primarily on the percentage of canopy closure. The area in hectare for each class was calculated using the ArcGIS ZONAL GEOMETRY Table tools, and on a mapping resolution of 1 kilometer.
Then we compare the classification of the forest land using the FAO's classification based on canopy closure threshold of 10% -40% with that of the UNFCCC criteria of 10% -30% canopy closure threshold. The difference in the assessment results from the two classification types leads to conclude that confusion in forest monitoring and assessments is due to classification based on few inventory parameter thresholds which are management objectives oriented. These inventory parameter thresholds are efficient tools for forest managers to make a sound decision on sylvicultural practices, but they cannot substitute for what forest is.
To define a forest land or to decide if a vegetation cover is forest ecosystem (e.g., a mature forest or ingrowth), a savanna, or other types of vegetation cover, requires the knowledge of additional ecosystem characteristics. This leads to recommend sylvicultural practices that could provide continuous fine local data to supplement for periodical remote sensing surveys in the forest resources assessment.
Our narrative analysis of the effectiveness of the forestry administration has overviewed the forest administration at the regional Central Togo and national levels, and examined its links with the International forestry administration. Our ultimate aim is to determine factors that have hindered the implementations of the forest land conservation policies. The major plan, program, and project at the national level are the National Forest Action Plan, the REDD+, and the Program on Biodiversity Conservation. At the international level the institutionalized program is the United Nations Environmental Program. The failure to forestry policies' implementation has been commonly associated with the so-local level (e.g., counties, villages, private landowners levels), to define forest management plans with the consideration of the interests of all users, primordially, that of the most dependent on forest products. These interests rather being conflictual need to be embedded in one other with harmonious collaboration of the stakeholders.