What is the difference between habitat loss and habitat fragmentation

Deep-water trawlers use heavy rock-hopping equipment, which has been reported to cause long-term to seabed habitats such as cold-water coral reefs in Norwegian, Scottish and Irish waters. Habitat loss often leads species to get endangered or threatened, and eventually extincted leading to significant loss of overall diversity and changes in ecosystem functioning. Conservation efforts of coastal and marine habitats have been driven in part by the effects of habitat loss on declines in species richness.

However, looking at the effect of habitats loss on species richness is not sufficient. Conservation efforts must consider the effects habitat loss and fragmentation on all components of species diversity and the ecosystem functioning and services provided by habitas:.

There is an urgent need for the implementation of long-term and large scale monitoring programs of changes to marine habitats and species distributions. This monitoring should be coupled with studies designed to treat management actions that result in habitat loss as large-scale experiments. For example, the use of marine protected areas MPAs and marine reserves as tools to underpinned the relationships between the functionality of habitats and the distribution and abundance of target taxa.

According to the Global Biodiversity Asessment [8] the most effective way to conserve biodiversity, by almost any reckoning is to prevent the conversion or degradation of habitat. There is an ongoing debate among conservationist biologist about whether is preferable to protect several already fragmented patches of habitats or a single large area, often referred as the SLOSS Single Large or Several Small debate.

One important aspect that also needs to be considered is habitat restoration. On land there is a long tradition of restoring habitats, such as mining waste tips. There are some examples of habitat restoration in the marine environment, such as the well-publicized clean-up of the River Thames in the UK where salmon can now be found in London. The developing science of restoration ecology should be a part of a strategy for conservation of coastal biodiversity.

Log in. Page Discussion. Read View source View history. Jump to: navigation , search. Marine biodiversity: patterns, threats and conservation needs. Loss, status and trends for coastal marine habitats of Europe. In addition, the relationships between habitat loss and habitat fragmentation based on simulated landscapes reported in eight articles were summarized to compare with our results in real urbanizing landscapes see S1 Appendix for detail.

Specifically, we used the built-up areas, the administrative boundaries, locations of central business districts CBD , and urban population of 16 cities around the world, i. These data were produced using both remote sensing imageries and historical maps Angel et al. The locations of the 16 study cities a and, as an example, the habitat loss during urbanization in Paris from to b. After the data acquisition, we implemented a time series correction for the built-up areas from to to improve their continuity and comparability, and then employed an indirect accuracy assessment based on urban population to test the consistency of the corrected data [ 12 ].

We found that the built-up area with correction was highly consistent with urban population See Figs A and B in S3 Appendix for details , which could represent the trend of urbanization in a reliable way. The details of the time series correction and the accuracy assessment had been described in our previous study [ 12 ].

Our method for evaluating the relationships between habitat loss and habitat fragmentation per se during urbanization included three steps: 1 extracting habitats; 2 quantifying habitat fragmentation per se; 3 analyzing relationships between habitat loss and habitat fragmentation per se. First of all, in consideration of data availability, we regarded all the non-built-up areas as habitats—similar to the habitats in the wider sense defined by IUCN—including forest, grassland, wetlands, cropland, and so forth [ 1 ].

Then, the habitats were extracted in each city from to Fig 1. At the second step, we chose several landscape metrics used in previous studies to quantify habitat fragmentation per se for facilitating comparison [ 23 , 24 , 26 — 29 ]. Particularly, we selected ten landscape metrics within four groups including: 1 area metrics, i. After that, we carried out two commonly used approaches—historical data based approach and space-for-time based approach [ 23 , 28 , 29 , 34 ]—to evaluate relationships between habitat loss and habitat fragmentation per se.

First, we performed the historical analysis at two extents, including the smaller extent of central city area dominated by built-up area [see details in 12 ] and the larger extent of urban region defined by administrative boundaries [ 11 , 12 ] Fig 1B , to explore possible effects of changing spatial scales [ 35 , 36 ].

At two extents, we respectively developed regression models of different types of function for habitat proportion and each landscape metric using historical data of habitat in each city from to The regression model with the highest value of R 2 was selected for representing the relationship between the habitat amount and the corresponding landscape metric.

Based on space-for-time perspective, we built three types of grid with different spatial extents i. At three extents, we calculated the habitat proportion and the ten landscape metrics in each grid in each city in , and then analyzed relationships between habitat proportion and landscape metrics using their values for all the grids.

In addition, we reviewed 22 relevant papers between and to summarize hypotheses on relationship between habitat loss and fragmentation Text A in S1 Appendix. Among them, eight papers reported the relationships between habitat loss and fragmentation based on simulated landscapes [ 7 , 23 , 26 — 30 , 38 ], and we summarized them into 14 forms Figs A—J in S1 Appendix.

By comparing these relationships in simulated landscapes with our results in real urbanizing landscapes, we classified the 14 forms into the following three groups: relationships found in both simulated landscapes and real urbanizing landscapes, relationships only found in real urbanizing landscapes, and relationships only found in simulated landscapes. As urbanization unfolded, habitat area decreased slowly during the first century, and then accelerated rapidly since about for all the 16 cities, while built-up area showed an opposite trend Fig 2.

This general pattern was consistent for the two spatial extents i. In the following section, we describe how habitat fragmentation, as measured by 10 different landscape metrics, changed with habitat loss. Changes in built-up area and habitat in 16 study cities from to at two extents: a the urban region and b the central city area. The study cities are ordered by urban population in In the period of —, the percentage of habitat was significantly correlated with nine landscape metrics measuring habitat fragmentation per se, with R 2 greater than 0.

Area metrics—mean patch size, total core area, and normalized total core area—were monotonically correlated with the percentage of habitat Fig 3A—3C , Table 2. The values of mean patch size decreased exponentially with continuing habitat loss over the years Fig 3A , Table 2 , while the values of total core area linearly decreased Fig 3B , Table 2 and the values of normalized total core area decreased with a logarithmic curve Fig 3C , Table 2. Density metrics, i.

With the process of habitat loss, patch and edge density both increased linearly from to Among two shape metrics, landscape shape index continuously increased with habitat loss. Specifically, both the linear function and the exponential function were found in terms of the relationships between landscape shape index and the percentage of habitat Fig 3F1 and 3F2 , Table 2.

In addition, fractal dimension generally increased at the beginning of habitat loss, then peaked and finally decreased with continuing habitat loss, revealing quadratic relationships with the percentage of habitat Fig 3G , Table 2.

Cohesion and normalized nearest neighbor distance—two connectivity metrics—showed logarithmic and quadratic relationships with the percentage of habitat respectively Fig 3H and 3I , Table 2. With the habitat loss from to , cohesion decreased monotonically Fig 3H , Table 2 , while normalized nearest neighbor distance increased at the first and then decreased Fig 3I , Table 2.

In addition to the general relationships, several cities and landscape metrics showed some idiosyncratic relationships. For example, when nearest neighbor distance was used to indicate habitat fragmentation per se, the significant relationships were not found in most cases Fig I in S4 Appendix. Also, fractal dimension in Beijing and London did not reveal the quadratic functions, which were found in other cities at the central city area extent Fig G-b in S4 Appendix. Based on the space-for-time analysis in , we found that all the ten metrics measuring habitat fragmentation revealed significant correlations with the percentage of habitat, and these relationships were exactly the same while different sample extent sizes 64 by 64 pixels, by pixels, and by pixels were used in the 16 world cities Fig 4 , Table 3 , Figs K—AD in S4 Appendix.

Three area metrics, including mean patch size, total core area, and normalized total core area, all decreased exponentially with habitat loss Fig 4A—4C , Table 3. In addition, density metrics i. Among three connectivity metrics, cohesion and nearest neighbor distance respectively revealed positively logarithmic and negatively power relationships with the percentage of habitat Fig 4H and 4I , Table 3 , whereas normalized nearest neighbor distance and the percentage of habitat represented quadratic relationships Fig 4J , Table 3.

Four forms of the HLHF relationship were shared by simulated landscapes and real urbanizing landscapes. Other ten forms of relationship were only found in simulated landscapes Fig 5.

For instance, the quadratic relationships between the percentage of habitat and density metrics patch density and edge density Fig 5D and 5E , the relationships between the percentage of habitat and the area metrics of total core area and normalized total core area Fig 5B, 5C1 and 5C2 , the relationships between the percentage of habitat and the shape metric of landscape shape index Fig 5F , and the relationships between the percentage of habitat and the connectivity metric of nearest neighbor distance Fig 5I.

In addition, six new forms of relationship were found in urbanizing landscapes, i. The consistent forms included the exponential relationships between the percentage of habitat and area metrics mean patch size, total core area, and normalized total core area , the quadratic relationships between the percentage of habitat and five metrics i.

Moreover, other four forms of relationship, e. Overall, the 16 study cities revealed similar relationships between habitat loss and fragmentation during urbanization. From to , the continuing habitat loss in 16 cities resulted in decreases in mean patch size, total core area, normalized total core area, cohesion, and increases in patch density, edge density, and landscape shape index, suggesting increasing habitat fragmentation and shape complexity, and decreasing habitat connectivity Fig 3.

Besides the general relationships, some idiosyncrasies existed in several cities as well. For example, Beijing and London did not reveal the quadratic function between fractal dimension and the percentage of habitat at the central city area extent Fig G-b in S4 Appendix.

In addition, the various urbanization patterns might result in idiosyncrasies on HLHF relationships. For example, the urban land changed slightly in Beijing from to , which was restricted by the city walls [ 39 ]. The particular urbanization patterns might constrain the habitat loss and fragmentation in Beijing during that period Fig 2.

The abnormal process of habitat fragmentation in Mumbai in the period of — might be attributable to the land policy change after the independence of India in [ 40 ]. It implied that the relevant policies on land use would be important roles in quantifying the HLHF relationships during urbanization.

It is well known that many landscape metrics are closely related and several metrics are often endogenously correlated with habitat abundance, resulting redundancy and inaccuracy in quantifying HLHF relationships [ 29 , 32 ].

To eliminate related metrics and quantify habitat fragmentation without redundancy, Frohn and Hao [ 41 ] classified 16 landscape metrics into four individual groups i. To remove the endogenous correlations between landscape metrics and habitat abundance, Wang and Cumming [ 29 ] proposed an approach to normalize landscape metrics by habitat abundance, and found that the normalization markedly reduced correlations with habitat abundance on natural landscapes.

In our research, we selected 10 landscape metrics to quantify habitat fragmentation from the literature on HLHF relationship to facilitate comparison. Specifically, three area metrics mean patch size, total core area and normalized total core area , two density metrics patch density and edge density , two shape metrics landscape shape index and fractal dimension , and three connectivity metrics nearest neighbor distance, normalized nearest neighbor distance and cohesion were used.

We found that the choice of landscape metrics matters in quantifying the habitat loss-fragmentation relationship during urbanization. For example, total core area and the percentage of habitat were linearly correlated Fig 3B , while nearest neighbor distance was not significantly correlated with the percentage of habitat Fig I in S4 Appendix.

After normalizing by habitat abundance [ 29 ], normalized total core area and the percentage of habitat showed logarithmic relationships Fig 3C , and normalized nearest neighbor distance and the percentage of habitat revealed quadratic relationships Fig 3I. In addition, fractal dimension and normalized nearest neighbor distance—two metrics calculated using values at patch level [ 32 ]—represented different relationships compared with landscape shape index and cohesion, which were directly calculated at class level Fig 3F—3I.

To avoid redundancy and endogenous correlations, we suggested that the normalized metrics and at least one of metrics representing different aspects of habitat fragmentation should be selected in terms of the previous studies [ 29 , 41 ].

In addition, two or three metrics representing habitat fragmentation in the same aspects could be used for confirming each other as well. Thus, four metrics i.

Mean patch size, edge density, fractal dimension and cohesion were recommended to confirm these in quantifying HLHF relationships. For example, patch density increased linearly with habitat loss from to Fig 3D , whereas patch density and the percentage of habitat revealed quadratic relationships when space-for-time analysis was utilized Fig 4D.

Space-for-time analysis is problematic when it is used to estimate relationships between habitat loss and fragmentation during urbanization. The space-for-time analysis, which "assumes that spatial and temporal variation are equivalent", is a commonly used approach to study long-term phenomena in ecology according to a series of different-aged samples [ 42 ], and has been used widely to evaluate habitat loss-fragmentation relationships during deforestation [ 27 , 28 , 37 ].

However, this approach has several problems when it is performed in landscapes across much environmental variance [ 42 ], e. It represented that even the percentage of habitat was low, a large number of small patches of habitat existed in the highly urbanized area. This phenomenon was found in several cities, where many small green spaces were kept to satisfy urban residents' requirements for cultural services from urban landscapes [ 43 ].

Thus, the patch density of habitat increased even though the area of habitat decreased during urbanization. However, the space-for-time analysis based on the samples in various places cannot well capture the artificially dominated dynamics of habitat fragmentation in the process of urbanization.

In addition, since the extent used in the space-for-time analysis is much smaller than the whole urban region, this approach would result in HLHF relationships different from the historical analysis due to the scaling effects in measuring habitat fragmentation [ 35 , 36 ]. The habitats in the space-for-time analysis were derived from data in Paris in with the extent of 64 by 64 pixels. In our study, we found that four forms of the HLHF relationship were shared by real urbanizing landscapes and simulated landscapes.

Among them, three forms were also consistent with that in the real landscapes with deforestation, which included exponential relationships between mean patch size and the percentage of habitat, logarithmic relationships between cohesion and the percentage of habitat, and quadratic relationships between normalized nearest neighbor distance and the percentage of habitat [ 23 , 26 , 28 — 30 , 44 ]. Habitat destruction is the process by which natural habitat is damaged or destroyed to such an extent that it no longer is capable of supporting the species and ecological communities that naturally occur there.

It often results in the extinction of species and, as a result, the loss of biodiversity. Habitat can be destroyed directly by many human activities, most of which involve the clearing of land for uses such as agriculture, mining, logging, hydroelectric dams, and urbanization. Although much habitat destruction can be attributed to human activity, it is not an exclusively man-made phenomenon. Habitat loss also occurs as a result of natural events such as floods, volcanic eruptions, earthquakes, and climate fluctuations.

For the most part, habitat destruction leads to species extinctions, but it can also open up new habitat that might provide an environment in which new species can evolve, thus demonstrating the resiliency of life on Earth. Sadly, humans are destroying natural habitats at a rate and on spatial scales that exceed what most species and communities can cope with. Habitat degradation is another consequence of human development. Humans indirectly cause habitat degradation through pollution, climate change, and the introduction of invasive species, all of which reduce the quality of the environment, making it difficult for native plants and animals to thrive.

Habitat degradation is fueled by a fast-growing human population. As the population increases, humans use more land for agriculture and for the development of cities and towns spread out over ever-widening areas.

The effects of habitat degradation not only affect native species and communities but human populations as well. Degraded lands are frequently lost to erosion, desertification, and nutrient depletion. Human development also leads to habitat fragmentation , as wild areas are carved up and split into smaller pieces. Fragmentation reduces animal ranges and restricts movement, placing animals in these areas at higher risk of extinction. Breaking up habitat can also separate animal populations, reducing genetic diversity.

Conservationists often seek to protect habitat in order to save individual animal species. For example, Conservation International invests in the Critical Ecosystem Partnership Fund , an initiative of multiple international organizations which provides grants to non-profit and private sector environmental groups to protect fragile habitats around the world. The groups' aim is to protect "biodiversity hotspots" that contain high concentrations of threatened species, such as Madagascar and the Guinean Forests of West Africa.

These areas are home to a unique array of plants and animals found nowhere else in the world. Conservation International believes that saving these "hotspots" is key to protecting the planet's biodiversity.