Generally the most obvious effect of the fragmentation process is the loss of native habitat. This has led many researchers to measure the degree of impact simply as the amount of remaining habitat in the landscape, ignoring that fragmentation can occur in various ways generating very different spatial patterns. The truth is that habitat loss always accompanies landscape fragmentation and that is why fragmentation indices strongly correlate with the proportion of habitat lost; however, they are different phenomena that must be discerned.[17]​[18]​ In fact, fragmentation per se generates important impacts on biodiversity that are independent of habitat loss.[19]​[18]​.

To understand the process of habitat fragmentation (which occurs as a consequence of its destruction) and its consequences on the populations, maps generated by random models were obtained that show that the size and isolation of the patches do not vary linearly with respect to the amount of habitat lost.[6]​[20]​[14]​ This determines that certain proportions of destroyed habitat are important thresholds, beyond which the decrease in size and increase in the isolation of the patches is accelerated, with the remaining populations being in the same non-viable.[6]​ In fact, small losses of habitat near the thresholds can mean large precipitations in the survival probability of metapopulations.[21]​ However, some empirical studies detected certain discrepancies with the levels of fragmentation expected on the basis of said theoretical models.[22]​[23]​.

These thresholds in the physical fragmentation process have important consequences for the conservation of biodiversity. The truth is that different species disappear at different points along the habitat destruction gradient,[24]​ with the extinction threshold being considered the minimum amount of habitat required for a population of a given species to persist in the environment.[25]​ This suggests that although extinction thresholds are affected by the way in which fragmentation varies in relation to the degree of habitat destruction, they also depend on the characteristics of the species (being species-specific) and the environments in which they occur. those that are found.[4]​ In fact, With and King[26]​ discovered that the extinction threshold can be found from 5% of habitat destroyed up to 90%, depending on the species and the spatial arrangement of the fragments.

Environmental science has established measures of the destruction of biodiversity caused by disturbances such as deforestation, climate change, the introduction of invasive species or the overexploitation of natural resources, with the Extinction Debt.[27]​.

The expected genetic consequences of fragmentation, which creates small, discrete and isolated populations, are based on traditional theories of island and metapopulation biogeography.[12]​[15]​ Thus, fragmentation is expected to erode genetic variation and increase divergence between populations due to increased genetic drift and inbreeding, and reduced gene flow in remaining populations.[28]​.

Theoretically, fragmentation can decrease the genetic variety of a population in two ways. First, reductions in population size at the time of fragmentation create a "Bottleneck (biology)" bottleneck, since only a small sample of the original gene pool is maintained.[28]​ The immediate effects depend on the effective size of the remaining populations and the patterns of variation present in the original population.[29]​.

After the bottleneck, if the remaining populations remain small and isolated for several generations, heterozygosity due to genetic drift and mainly due to inbreeding will decrease, resulting in the accumulation of deleterious "Dominance (genetic)" recessive alleles and decreased fecundity, leading the population to eventual extinction.[30][28]​ The loss of variety can reduce the ability of populations to respond to future environmental changes, increasing the probability. of extinction, or at least decreasing opportunities for adaptation.[28][29]​.

However, empirical evidence provides inconsistent results, since not all episodes of fragmentation in plants necessarily led to genetic erosion,[28]​ which can be explained by their particular attributes (generation time, ploidy level, vegetative reproduction, etc.) that can confer different vulnerability.[31]​ Likewise, factors such as the time elapsed since fragmentation and ecological processes can shape the levels of demographic and genetic connection.[31]​.

Contenido

El proceso de fragmentación genera paisajes y fragmentos con determinadas características, que en su conjunto describen los atributos espaciales del hábitat. Estas pueden ser agrupadas dentro de cinco categorías: (1) área del fragmento, (2) efecto de borde, (3) forma del fragmento, (4) aislamiento del fragmento y (5) estructura de la matriz.

Fragment area

The size of some fragments may be smaller than necessary for some populations to be viable, and therefore generate notable reductions in their size. This eventually generates local extinctions, and therefore, small patches generally have fewer species.[16]​[32]​ For Wilcove[1]​ This is frequently the case in large animals, however different studies have predicted and observed positive, negative and absent relationships between body size and the probability of extinction (reviewed by Johst and Brandl).[33]​.

The fact that the area of the fragment imposes a maximum limit on the population size increases the vulnerability of the species to local extinctions,[34]​ since they are exposed to genetic erosion due to genetic drift and inbreeding depression.[35]​ The mechanisms underlying the area/vulnerability relationship can be divided into four categories: environmental randomness, demographic randomness, natural catastrophes and reduction of genetic variability;[36]​ and although it is worth distinguishing them, It is important to note that they rarely act independently.[4]​.

At the same time, the habitat is not homogeneous, so there is a risk that the individual fragments do not cover all the microhabitats found in the large original fragment. Therefore, species that need several microhabitats are especially vulnerable since they must move from one fragment to another to meet their requirements, which in some cases is impossible.[1]​.

Furthermore, it has been shown that in small fragments various aspects of the structure of populations and communities are impacted;[18]​ for example, the reduction in the length of food chains,[37]​ in the number of specialist and large species,[38]​ and the alteration of interspecific interactions.[39]​ In turn, fragmentation negatively affects reproductive success,[40]​ the success of dispersal,[41]​[21]​ the rate of predation[42]​ and aspects of animal behavior that affect feeding success.[43]​.

Border effects

Edge effects are changes in physical and biological conditions associated with artificial and abrupt boundaries in habitat fragments.[13] Abiotic effects refer to changes in physical variables such as radiation, humidity, temperature, wind speed and availability of nutrients in the soil. While biotic effects are changes in biological variables such as species composition or patterns of competition, predation and parasitism.[44]​ Their greater or lesser penetration into the fragment can vary widely, from tens of meters to several kilometers,[45]​ as a consequence of various factors. Within which we can include the structural contrast with the matrix, wind speeds, temperatures and the presence or absence of invasive exotic species.[46]​ Additionally, the edges that are in contact with urban areas experience modifications due to human activity.[16]​.

Edge habitat can alter the nature of interactions and therefore modify ecological processes and dynamics at a wide range of scales.[47] The structure, composition and diversity of communities are affected in various ways in edge zones. In invertebrate communities, for example, species richness typically decreases as one enters the center of the patch. The most common explanation for this is that the overlapping faunas of both sides outnumber the lost species.[48]​[49]​ However, this pattern is not universal since several studies have detected positive[50]​ or absent[51]​ correlations between species richness and the distance to the edge of the patch; which suggests that in some systems many species avoid the edges, and that there are not always enough species in the matrix to compensate for the loss in them.[4]​ Curiously, some animals appear to select or prefer the edges to reproduce, even though the mortality rates there can be much higher, which is why in those cases they are called ecological traps.[52][53]​.

Fragment shape

Basically, shape is determined by the ratio of area to perimeter, which in turn determines the amount of remaining, unaltered habitat (or core). This is why fragments with complex shapes have a greater proportion of their area as an edge.[54]​[16]​ Likewise, highly twisted shapes can generate additional divisions in the core of the fragment, creating multiple disjoint cores (separated by edge habitat) in the same fragment.[55]​ However, the effect of shape is not independent of size. The impact of edge effects increases as fragment size decreases, as any deviation from circularity significantly reduces the amount of interior habitat, creating a confusing intercorrelation between area and shape effects in fragmented landscapes.[54]​.

Perhaps the most consistent pattern that emerges from the shape of the fragments is that those with complex shapes are colonized more frequently.[16] This is because the proportional amount of edge is greater, increasing the probability that a moving individual encounters the patch.[56] In turn, this probability may be moderated by the relative orientation of the patch. When species movement occurs in a predictable direction, such as migrations[57]​ or tidal movements,[58]​ long, narrow patches are more exposed to colonization, but only if they are oriented perpendicular to the direction of movement. However, just as colonization increases, so does emigration. The result of this dynamic is the reduction in the probability of populations remaining in these fragments.[59]​.

Fragment isolation

One of the most obvious spatial consequences of fragmentation is the isolation in time and space of habitat patches. Such isolation is almost universally understood as a measure of habitat configuration. However, it is more accurate to interpret it as a measure of the amount of habitat removed.[18]​.

Isolation disrupts the distribution patterns of species and forces the dispersal of individuals through the matrix.[4]​ Given that matrix habitat may inhibit the dispersal of some species more than others,[60]​ and that species differ in their willingness to traverse it,[61]​ controversial results have been documented. Particularly in insects, genetic differentiation between populations was clearly related to isolation in some studies,[62][63]​ but not in others.[64]​.

In conclusion, interruptions can fractionate metapopulations, decreasing the total population size, increasing the probability of eventual local extinctions, and preventing new colonizations from reestablishing extinct populations. This is why one of the most intuitive conservation measures is the creation of biological corridors.[65]​.

Matrix structure

The interpretation of modified landscapes as "islands" represents much of the literature on fragmentation.[11] However, the traditional theory of Wilson and McArthur fails to recognize that edge effects can alter the characteristics of the fragment, and that the matrix may not be completely inhospitable to native fauna. Indeed, a growing body of evidence suggests that matrix characteristics are crucially important in determining the abundance and composition of species within fragments,[61]​ primarily affecting edge characteristics and dispersal and connectivity between patches.[4]​.

Dispersion between fragments is essential for the long-term permanence of metapopulations,[66]​ and depends largely on the characteristics of the matrix.[60]​ This acts as a selective filter for the movement of propagules and animals, affecting the composition of species in the fragments.[13]​ Its structure, for example, can alter the colonization-extinction dynamics and give rise to changes in population density,[66]​ the structure of the fragments[67]​ and inter-specific interactions.[68]​ On the other hand, the matrix can influence the nature and magnitude of edge effects.[13]​ Highly contrasting fragments and matrices generate more severe edge effects[69]​ and are more impermeable to animal movement.[56]​.

On the other hand, the remnants of native habitat are not necessarily the only parts of the landscape that provide relevant resources for the permanence of the species, since some are capable of compensating for the loss of habitat by making use of the resources present in the matrix.[4]​ Through simulation models it has been shown that generalist species (i.e. those that can use resources from the matrix) reach larger population sizes than specialist species.[70]​ Likewise, similar models show that increasing the quality of the matrix (i.e. increasing the resources it contributes to the system) the population density within the fragments increases.[71]​ In this way, the ability of species to use resources from the matrix can alter the intensity of the effects of habitat fragmentation.

Studies on the impact of fragmentation can vary widely in their results according to the components included in the research. This explains why the literature on habitat fragmentation is replete with examples with seemingly contradictory results that cannot be explained by taking into account only the differences between environments or methods used.[4] These differences highlight the wide number of factors that can hide or enhance the effects of fragmentation.

Examples of such factors may be both the mechanisms that explain the patterns observed at the species level, as well as the attributes of the species that determine their susceptibility to the fragmentation process.[4]​ In general, there are a series of attributes that are commonly thought to increase the vulnerability of species to fragmentation, including large body sizes, low mobility or dispersal, tolerance to the matrix, dependence on mutualistic interactions "Mutualism (biology)") and trophic levels. high.[19]​[61]​[72]​ In fact, species extinction has been shown to occur in a consistent and sequential pattern as fragment area decreases, indicating a gradient in vulnerability to extinction across species.[4]​ Thus, studying the impacts of fragmentation on species with different extinction thresholds can generate seemingly contradictory results.

Likewise, species comprise a set of attributes that are strongly correlated[61]​ and that can interact with each other, and with those of other species, synergistically modifying or magnifying the impacts of fragmentation. An example of this is the plant-pollinator interaction, since it has been documented that habitat loss and fragmentation can significantly alter the nature of pollinator communities and interrupt the interaction.[72]​.

On the other hand, there is a large temporal component in the manifestation of species responses after fragmentation, and given that most studies of landscapes fragmented by human action have been carried out in less than one hundred years after fragmentation, the long-term effects are relatively unknown. This mode increases population and species density.[73]​ However, the fragments are not capable of supporting all individuals and species in the long term, which causes a progressive reduction in the abundance of species over time.[32]​ These reductions are commonly called extinction debts, and occur in the medium and long term,[74]​ describing a temporal delay between the loss of habitat and the eventual collapse of populations. The extinction debt is illustrated by negative correlations between species richness and the age of the fragment.[75]​ However, several studies have found the opposite in fragments of intermediate[76]​ and long age,[77]​ which suggests that after the impact the communities could reach new “equilibria”.

Another factor that can influence the impacts of fragmentation is the biogeography and history of the environments. The sensitivity of species is different according to the different biomes, being particularly low in the temperate zones of the northern hemisphere.[78]​ A possible explanation is that of the extinction filter,[79]​ which assumes that fragmentation in these biomes is old enough that the most vulnerable species have already become extinct. Communities from temperate regions are also believed to be more resistant to the effects of habitat fragmentation than communities from tropical regions, because the former are more widely distributed, occur at higher densities, and have greater dispersal powers than their tropical counterparts.[1]​.

Finally, it should be noted that fragmentation can have positive effects on biodiversity, although these can be masked by the negative effects of habitat loss.[18]​ For example, Huffaker[80]​ carried out experiments suggesting that subdivision of the same amount of habitat into many small patches can favor the permanence of predator-prey systems. He hypothesized that the subdivision provides temporary refuges for prey, where they can increase in number and disperse to other sites before the predator or parasite finds them; idea that can be transferred to the fragmentation process.

Recent models have indicated that habitat loss and fragmentation can increase the susceptibility of species to increased temperatures and the frequency of extreme climate events resulting from global climate change,[81]​ since when populations are isolated their area of expansion is restricted.[82]​[83]​ This restriction does not allow species to escape from unfavorable conditions, a typical survival strategy of certain insects.[82]​ As a consequence of this strategy, recently in Europe there has been expanded distribution at the northern limit of several butterfly species as a result of global warming; and even greater expansion is expected in the future.[82]​ This change in the species' range is actually the result of the extinction of populations at the warm limits of the range, and the colonization and growth of populations at the cooler limits.[83]​ Such dynamics are believed to be faster in regions where the landscape structure allows dispersal, compared to regions where the landscape is fragmented. That is, from the assumed interaction between climate change and fragmentation, large rates of expansion would be generated in landscapes that have high spatial cohesion; although there are no studies that support this prediction.[83]​.

An important factor to take into account is that climate change must occur at a speed that allows the expansion and colonization of new environments before the propagule-providing population becomes extinct. This coordination depends in turn on the characteristics of the species, such as the range of propagule dispersal. However, if species require long periods of time or numerous colonization events to establish new settlements, the expansion rate will be much lower than expected based on the species' dispersal capacity.[83] Likewise, the extinction-recolonization process causes a less effective distribution throughout the habitat, and the mortality of propagules is high, which extends the recovery time of metapopulations. In conclusion, it can be said that habitat fragmentation multiplies the impact of climate change.[83]​.

As previously mentioned, landscape modification and fragmentation constitute one of the most serious and urgent issues within conservation ecology.[11]​ Therefore, it is not surprising that there is a large number of theoretical and observational studies.[32]​ Based on these, different measures have been proposed, many of them potentially complementary, to mitigate or avoid the negative consequences of the fragmentation process.

One of the most widespread measures is the creation of nature reserves. Ideally, their purpose is to maintain a complex set of physical, ecological, genetic, behavioral and evolutionary processes, and the compatible populations that participate in them.[84] Faced with such complexity, the basic difficulties in planning are: their quantity, form and proximity; and the solution varies depending on the scale of the conservation effort.[1]​ In trying to resolve these difficulties, the theory of fragmentation formed the basis of the acronym SLOSS (single large or serveral small), arousing great controversies.[13]​ Despite the intense debate, large reserves are considered to be superior in terms of the long-term persistence of species,[85]​ since they retain larger populations and the habitat is less exposed to degradation through time.[86]​.

On the other hand, one of the obvious methods to slow down and compensate for the advance of local extinctions is artificial recolonization, and for this purpose genetic resources constitute an extremely useful tool. These interventions, in addition to being necessary to avoid extinctions, ensure the permanence of a sufficient number of local populations and preserve global genetic variability; and there are two basic strategies: recolonize with animals from a single large population, or with a mixture of animals from several populations.[35]​ The latter can restore variability at the local level, but since introduced populations are generally small, this variation can be quickly lost due to genetic drift. But even a temporary infusion of genetic variability can be beneficial. Other factors to take into account are outbreeding depression (i.e. the breakdown of adapted genotypes by recombination) and inbreeding depression (as a consequence of the small population size), which would endanger the population introduced in the first generations until adapted genotypes are selected and established again.[35]​.

Despite the risks in the first generations, the introduction of admixed populations allows deleterious alleles to be eliminated by natural selection. On the other hand, the introduction of individuals from a single population does not allow it, therefore any characteristic can be fixed by inbreeding, putting the population at risk.[35] This suggests that the mixed release of a number of individuals ideally close to the carrying capacity of the system is preferable. But the recolonization process continues, since the released population must be genetically and demographically monitored in the first generations to detect possible depressions due to exo- or inbreeding.[35]​.

From another perspective, it has been proposed to moderate the negative effects of the isolation of the fragments by maintaining the connectivity of the landscape through soft matrices, cornerstones, and mainly biological corridors.[87]​[44]​ In the ecological context of fragmentation, the term corridor refers to a linear landscape element composed of native vegetation that unites patches of similar native vegetation.[87]​ These connections facilitate the movement of plants and animals between the different patches, allowing a greater number of species to exist and/or remain for a longer time than expected. according to their size; That is, the effective size of the fragments increases.[16]​.

The suitability of corridors as habitat or possible dispersal routes varies depending on the species. In the case of animals it depends on feeding patterns, body size, territory size, degree of dietary specialization, mobility and social behavior. Likewise, the perception and use of corridors by animals differs according to the physical dimensions and the context of the landscape.[88]​.

However, not all the consequences of corridors are positive, since they, for example, can facilitate the expansion of the distribution of invasive exotic species.[89] Likewise, through simulations it has been shown that when the host-parasite interaction is important in host dynamics, the increase in landscape connectivity promotes parasite transmission, increasing the probability of extinction of the metapopulation.[90] It is worth clarifying that later models have indicated that the benefits of corridors typically offset or outweigh the risks of parasite transmission.[91]​ Therefore, despite some ecological risks, increasing landscape connectivity is more likely to find positive than negative effects on native species and ecological processes.[11]​.

Finally, it is worth mentioning that it has been recommended to leave aside the conservationist approach at the species level, to focus on the conditions in which communities develop. These conditions must allow populations to respond to large-scale disturbances and genetically adapt to such changes.[83]​ To do this, it must be taken into account that the factors that influence populations are numerous, and therefore, those conservation plans that are only based on the area to be preserved will probably have limited success.[92]​.

Generally the most obvious effect of the fragmentation process is the loss of native habitat. This has led many researchers to measure the degree of impact simply as the amount of remaining habitat in the landscape, ignoring that fragmentation can occur in various ways generating very different spatial patterns. The truth is that habitat loss always accompanies landscape fragmentation and that is why fragmentation indices strongly correlate with the proportion of habitat lost; however, they are different phenomena that must be discerned.[17]​[18]​ In fact, fragmentation per se generates important impacts on biodiversity that are independent of habitat loss.[19]​[18]​.

To understand the process of habitat fragmentation (which occurs as a consequence of its destruction) and its consequences on the populations, maps generated by random models were obtained that show that the size and isolation of the patches do not vary linearly with respect to the amount of habitat lost.[6]​[20]​[14]​ This determines that certain proportions of destroyed habitat are important thresholds, beyond which the decrease in size and increase in the isolation of the patches is accelerated, with the remaining populations being in the same non-viable.[6]​ In fact, small losses of habitat near the thresholds can mean large precipitations in the survival probability of metapopulations.[21]​ However, some empirical studies detected certain discrepancies with the levels of fragmentation expected on the basis of said theoretical models.[22]​[23]​.

These thresholds in the physical fragmentation process have important consequences for the conservation of biodiversity. The truth is that different species disappear at different points along the habitat destruction gradient,[24]​ with the extinction threshold being considered the minimum amount of habitat required for a population of a given species to persist in the environment.[25]​ This suggests that although extinction thresholds are affected by the way in which fragmentation varies in relation to the degree of habitat destruction, they also depend on the characteristics of the species (being species-specific) and the environments in which they occur. those that are found.[4]​ In fact, With and King[26]​ discovered that the extinction threshold can be found from 5% of habitat destroyed up to 90%, depending on the species and the spatial arrangement of the fragments.

Environmental science has established measures of the destruction of biodiversity caused by disturbances such as deforestation, climate change, the introduction of invasive species or the overexploitation of natural resources, with the Extinction Debt.[27]​.

The expected genetic consequences of fragmentation, which creates small, discrete and isolated populations, are based on traditional theories of island and metapopulation biogeography.[12]​[15]​ Thus, fragmentation is expected to erode genetic variation and increase divergence between populations due to increased genetic drift and inbreeding, and reduced gene flow in remaining populations.[28]​.

Theoretically, fragmentation can decrease the genetic variety of a population in two ways. First, reductions in population size at the time of fragmentation create a "Bottleneck (biology)" bottleneck, since only a small sample of the original gene pool is maintained.[28]​ The immediate effects depend on the effective size of the remaining populations and the patterns of variation present in the original population.[29]​.

After the bottleneck, if the remaining populations remain small and isolated for several generations, heterozygosity due to genetic drift and mainly due to inbreeding will decrease, resulting in the accumulation of deleterious "Dominance (genetic)" recessive alleles and decreased fecundity, leading the population to eventual extinction.[30][28]​ The loss of variety can reduce the ability of populations to respond to future environmental changes, increasing the probability. of extinction, or at least decreasing opportunities for adaptation.[28][29]​.

However, empirical evidence provides inconsistent results, since not all episodes of fragmentation in plants necessarily led to genetic erosion,[28]​ which can be explained by their particular attributes (generation time, ploidy level, vegetative reproduction, etc.) that can confer different vulnerability.[31]​ Likewise, factors such as the time elapsed since fragmentation and ecological processes can shape the levels of demographic and genetic connection.[31]​.

Contenido

El proceso de fragmentación genera paisajes y fragmentos con determinadas características, que en su conjunto describen los atributos espaciales del hábitat. Estas pueden ser agrupadas dentro de cinco categorías: (1) área del fragmento, (2) efecto de borde, (3) forma del fragmento, (4) aislamiento del fragmento y (5) estructura de la matriz.

Fragment area

The size of some fragments may be smaller than necessary for some populations to be viable, and therefore generate notable reductions in their size. This eventually generates local extinctions, and therefore, small patches generally have fewer species.[16]​[32]​ For Wilcove[1]​ This is frequently the case in large animals, however different studies have predicted and observed positive, negative and absent relationships between body size and the probability of extinction (reviewed by Johst and Brandl).[33]​.

The fact that the area of the fragment imposes a maximum limit on the population size increases the vulnerability of the species to local extinctions,[34]​ since they are exposed to genetic erosion due to genetic drift and inbreeding depression.[35]​ The mechanisms underlying the area/vulnerability relationship can be divided into four categories: environmental randomness, demographic randomness, natural catastrophes and reduction of genetic variability;[36]​ and although it is worth distinguishing them, It is important to note that they rarely act independently.[4]​.

At the same time, the habitat is not homogeneous, so there is a risk that the individual fragments do not cover all the microhabitats found in the large original fragment. Therefore, species that need several microhabitats are especially vulnerable since they must move from one fragment to another to meet their requirements, which in some cases is impossible.[1]​.

Furthermore, it has been shown that in small fragments various aspects of the structure of populations and communities are impacted;[18]​ for example, the reduction in the length of food chains,[37]​ in the number of specialist and large species,[38]​ and the alteration of interspecific interactions.[39]​ In turn, fragmentation negatively affects reproductive success,[40]​ the success of dispersal,[41]​[21]​ the rate of predation[42]​ and aspects of animal behavior that affect feeding success.[43]​.

Border effects

Edge effects are changes in physical and biological conditions associated with artificial and abrupt boundaries in habitat fragments.[13] Abiotic effects refer to changes in physical variables such as radiation, humidity, temperature, wind speed and availability of nutrients in the soil. While biotic effects are changes in biological variables such as species composition or patterns of competition, predation and parasitism.[44]​ Their greater or lesser penetration into the fragment can vary widely, from tens of meters to several kilometers,[45]​ as a consequence of various factors. Within which we can include the structural contrast with the matrix, wind speeds, temperatures and the presence or absence of invasive exotic species.[46]​ Additionally, the edges that are in contact with urban areas experience modifications due to human activity.[16]​.

Edge habitat can alter the nature of interactions and therefore modify ecological processes and dynamics at a wide range of scales.[47] The structure, composition and diversity of communities are affected in various ways in edge zones. In invertebrate communities, for example, species richness typically decreases as one enters the center of the patch. The most common explanation for this is that the overlapping faunas of both sides outnumber the lost species.[48]​[49]​ However, this pattern is not universal since several studies have detected positive[50]​ or absent[51]​ correlations between species richness and the distance to the edge of the patch; which suggests that in some systems many species avoid the edges, and that there are not always enough species in the matrix to compensate for the loss in them.[4]​ Curiously, some animals appear to select or prefer the edges to reproduce, even though the mortality rates there can be much higher, which is why in those cases they are called ecological traps.[52][53]​.

Fragment shape

Basically, shape is determined by the ratio of area to perimeter, which in turn determines the amount of remaining, unaltered habitat (or core). This is why fragments with complex shapes have a greater proportion of their area as an edge.[54]​[16]​ Likewise, highly twisted shapes can generate additional divisions in the core of the fragment, creating multiple disjoint cores (separated by edge habitat) in the same fragment.[55]​ However, the effect of shape is not independent of size. The impact of edge effects increases as fragment size decreases, as any deviation from circularity significantly reduces the amount of interior habitat, creating a confusing intercorrelation between area and shape effects in fragmented landscapes.[54]​.

Perhaps the most consistent pattern that emerges from the shape of the fragments is that those with complex shapes are colonized more frequently.[16] This is because the proportional amount of edge is greater, increasing the probability that a moving individual encounters the patch.[56] In turn, this probability may be moderated by the relative orientation of the patch. When species movement occurs in a predictable direction, such as migrations[57]​ or tidal movements,[58]​ long, narrow patches are more exposed to colonization, but only if they are oriented perpendicular to the direction of movement. However, just as colonization increases, so does emigration. The result of this dynamic is the reduction in the probability of populations remaining in these fragments.[59]​.

Fragment isolation

One of the most obvious spatial consequences of fragmentation is the isolation in time and space of habitat patches. Such isolation is almost universally understood as a measure of habitat configuration. However, it is more accurate to interpret it as a measure of the amount of habitat removed.[18]​.

Isolation disrupts the distribution patterns of species and forces the dispersal of individuals through the matrix.[4]​ Given that matrix habitat may inhibit the dispersal of some species more than others,[60]​ and that species differ in their willingness to traverse it,[61]​ controversial results have been documented. Particularly in insects, genetic differentiation between populations was clearly related to isolation in some studies,[62][63]​ but not in others.[64]​.

In conclusion, interruptions can fractionate metapopulations, decreasing the total population size, increasing the probability of eventual local extinctions, and preventing new colonizations from reestablishing extinct populations. This is why one of the most intuitive conservation measures is the creation of biological corridors.[65]​.

Matrix structure

The interpretation of modified landscapes as "islands" represents much of the literature on fragmentation.[11] However, the traditional theory of Wilson and McArthur fails to recognize that edge effects can alter the characteristics of the fragment, and that the matrix may not be completely inhospitable to native fauna. Indeed, a growing body of evidence suggests that matrix characteristics are crucially important in determining the abundance and composition of species within fragments,[61]​ primarily affecting edge characteristics and dispersal and connectivity between patches.[4]​.

Dispersion between fragments is essential for the long-term permanence of metapopulations,[66]​ and depends largely on the characteristics of the matrix.[60]​ This acts as a selective filter for the movement of propagules and animals, affecting the composition of species in the fragments.[13]​ Its structure, for example, can alter the colonization-extinction dynamics and give rise to changes in population density,[66]​ the structure of the fragments[67]​ and inter-specific interactions.[68]​ On the other hand, the matrix can influence the nature and magnitude of edge effects.[13]​ Highly contrasting fragments and matrices generate more severe edge effects[69]​ and are more impermeable to animal movement.[56]​.

On the other hand, the remnants of native habitat are not necessarily the only parts of the landscape that provide relevant resources for the permanence of the species, since some are capable of compensating for the loss of habitat by making use of the resources present in the matrix.[4]​ Through simulation models it has been shown that generalist species (i.e. those that can use resources from the matrix) reach larger population sizes than specialist species.[70]​ Likewise, similar models show that increasing the quality of the matrix (i.e. increasing the resources it contributes to the system) the population density within the fragments increases.[71]​ In this way, the ability of species to use resources from the matrix can alter the intensity of the effects of habitat fragmentation.

Studies on the impact of fragmentation can vary widely in their results according to the components included in the research. This explains why the literature on habitat fragmentation is replete with examples with seemingly contradictory results that cannot be explained by taking into account only the differences between environments or methods used.[4] These differences highlight the wide number of factors that can hide or enhance the effects of fragmentation.

Examples of such factors may be both the mechanisms that explain the patterns observed at the species level, as well as the attributes of the species that determine their susceptibility to the fragmentation process.[4]​ In general, there are a series of attributes that are commonly thought to increase the vulnerability of species to fragmentation, including large body sizes, low mobility or dispersal, tolerance to the matrix, dependence on mutualistic interactions "Mutualism (biology)") and trophic levels. high.[19]​[61]​[72]​ In fact, species extinction has been shown to occur in a consistent and sequential pattern as fragment area decreases, indicating a gradient in vulnerability to extinction across species.[4]​ Thus, studying the impacts of fragmentation on species with different extinction thresholds can generate seemingly contradictory results.

Likewise, species comprise a set of attributes that are strongly correlated[61]​ and that can interact with each other, and with those of other species, synergistically modifying or magnifying the impacts of fragmentation. An example of this is the plant-pollinator interaction, since it has been documented that habitat loss and fragmentation can significantly alter the nature of pollinator communities and interrupt the interaction.[72]​.

On the other hand, there is a large temporal component in the manifestation of species responses after fragmentation, and given that most studies of landscapes fragmented by human action have been carried out in less than one hundred years after fragmentation, the long-term effects are relatively unknown. This mode increases population and species density.[73]​ However, the fragments are not capable of supporting all individuals and species in the long term, which causes a progressive reduction in the abundance of species over time.[32]​ These reductions are commonly called extinction debts, and occur in the medium and long term,[74]​ describing a temporal delay between the loss of habitat and the eventual collapse of populations. The extinction debt is illustrated by negative correlations between species richness and the age of the fragment.[75]​ However, several studies have found the opposite in fragments of intermediate[76]​ and long age,[77]​ which suggests that after the impact the communities could reach new “equilibria”.

Another factor that can influence the impacts of fragmentation is the biogeography and history of the environments. The sensitivity of species is different according to the different biomes, being particularly low in the temperate zones of the northern hemisphere.[78]​ A possible explanation is that of the extinction filter,[79]​ which assumes that fragmentation in these biomes is old enough that the most vulnerable species have already become extinct. Communities from temperate regions are also believed to be more resistant to the effects of habitat fragmentation than communities from tropical regions, because the former are more widely distributed, occur at higher densities, and have greater dispersal powers than their tropical counterparts.[1]​.

Finally, it should be noted that fragmentation can have positive effects on biodiversity, although these can be masked by the negative effects of habitat loss.[18]​ For example, Huffaker[80]​ carried out experiments suggesting that subdivision of the same amount of habitat into many small patches can favor the permanence of predator-prey systems. He hypothesized that the subdivision provides temporary refuges for prey, where they can increase in number and disperse to other sites before the predator or parasite finds them; idea that can be transferred to the fragmentation process.

Recent models have indicated that habitat loss and fragmentation can increase the susceptibility of species to increased temperatures and the frequency of extreme climate events resulting from global climate change,[81]​ since when populations are isolated their area of expansion is restricted.[82]​[83]​ This restriction does not allow species to escape from unfavorable conditions, a typical survival strategy of certain insects.[82]​ As a consequence of this strategy, recently in Europe there has been expanded distribution at the northern limit of several butterfly species as a result of global warming; and even greater expansion is expected in the future.[82]​ This change in the species' range is actually the result of the extinction of populations at the warm limits of the range, and the colonization and growth of populations at the cooler limits.[83]​ Such dynamics are believed to be faster in regions where the landscape structure allows dispersal, compared to regions where the landscape is fragmented. That is, from the assumed interaction between climate change and fragmentation, large rates of expansion would be generated in landscapes that have high spatial cohesion; although there are no studies that support this prediction.[83]​.

An important factor to take into account is that climate change must occur at a speed that allows the expansion and colonization of new environments before the propagule-providing population becomes extinct. This coordination depends in turn on the characteristics of the species, such as the range of propagule dispersal. However, if species require long periods of time or numerous colonization events to establish new settlements, the expansion rate will be much lower than expected based on the species' dispersal capacity.[83] Likewise, the extinction-recolonization process causes a less effective distribution throughout the habitat, and the mortality of propagules is high, which extends the recovery time of metapopulations. In conclusion, it can be said that habitat fragmentation multiplies the impact of climate change.[83]​.

As previously mentioned, landscape modification and fragmentation constitute one of the most serious and urgent issues within conservation ecology.[11]​ Therefore, it is not surprising that there is a large number of theoretical and observational studies.[32]​ Based on these, different measures have been proposed, many of them potentially complementary, to mitigate or avoid the negative consequences of the fragmentation process.

One of the most widespread measures is the creation of nature reserves. Ideally, their purpose is to maintain a complex set of physical, ecological, genetic, behavioral and evolutionary processes, and the compatible populations that participate in them.[84] Faced with such complexity, the basic difficulties in planning are: their quantity, form and proximity; and the solution varies depending on the scale of the conservation effort.[1]​ In trying to resolve these difficulties, the theory of fragmentation formed the basis of the acronym SLOSS (single large or serveral small), arousing great controversies.[13]​ Despite the intense debate, large reserves are considered to be superior in terms of the long-term persistence of species,[85]​ since they retain larger populations and the habitat is less exposed to degradation through time.[86]​.

On the other hand, one of the obvious methods to slow down and compensate for the advance of local extinctions is artificial recolonization, and for this purpose genetic resources constitute an extremely useful tool. These interventions, in addition to being necessary to avoid extinctions, ensure the permanence of a sufficient number of local populations and preserve global genetic variability; and there are two basic strategies: recolonize with animals from a single large population, or with a mixture of animals from several populations.[35]​ The latter can restore variability at the local level, but since introduced populations are generally small, this variation can be quickly lost due to genetic drift. But even a temporary infusion of genetic variability can be beneficial. Other factors to take into account are outbreeding depression (i.e. the breakdown of adapted genotypes by recombination) and inbreeding depression (as a consequence of the small population size), which would endanger the population introduced in the first generations until adapted genotypes are selected and established again.[35]​.

Despite the risks in the first generations, the introduction of admixed populations allows deleterious alleles to be eliminated by natural selection. On the other hand, the introduction of individuals from a single population does not allow it, therefore any characteristic can be fixed by inbreeding, putting the population at risk.[35] This suggests that the mixed release of a number of individuals ideally close to the carrying capacity of the system is preferable. But the recolonization process continues, since the released population must be genetically and demographically monitored in the first generations to detect possible depressions due to exo- or inbreeding.[35]​.

From another perspective, it has been proposed to moderate the negative effects of the isolation of the fragments by maintaining the connectivity of the landscape through soft matrices, cornerstones, and mainly biological corridors.[87]​[44]​ In the ecological context of fragmentation, the term corridor refers to a linear landscape element composed of native vegetation that unites patches of similar native vegetation.[87]​ These connections facilitate the movement of plants and animals between the different patches, allowing a greater number of species to exist and/or remain for a longer time than expected. according to their size; That is, the effective size of the fragments increases.[16]​.

The suitability of corridors as habitat or possible dispersal routes varies depending on the species. In the case of animals it depends on feeding patterns, body size, territory size, degree of dietary specialization, mobility and social behavior. Likewise, the perception and use of corridors by animals differs according to the physical dimensions and the context of the landscape.[88]​.

However, not all the consequences of corridors are positive, since they, for example, can facilitate the expansion of the distribution of invasive exotic species.[89] Likewise, through simulations it has been shown that when the host-parasite interaction is important in host dynamics, the increase in landscape connectivity promotes parasite transmission, increasing the probability of extinction of the metapopulation.[90] It is worth clarifying that later models have indicated that the benefits of corridors typically offset or outweigh the risks of parasite transmission.[91]​ Therefore, despite some ecological risks, increasing landscape connectivity is more likely to find positive than negative effects on native species and ecological processes.[11]​.

Finally, it is worth mentioning that it has been recommended to leave aside the conservationist approach at the species level, to focus on the conditions in which communities develop. These conditions must allow populations to respond to large-scale disturbances and genetically adapt to such changes.[83]​ To do this, it must be taken into account that the factors that influence populations are numerous, and therefore, those conservation plans that are only based on the area to be preserved will probably have limited success.[92]​.