Characterizing forest fragmentation: Distinguishing change in composition from configuration

Long, Jed A.; Nelson, Trisalyn A.; Wulder, Michael A.

doi:10.1016/j.apgeog.2009.12.002

Cited by 51 publications

(38 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Factor data were subsequently standardized into a range from 0 to 1 using a minimum-maximum linear transformation. This normalization is necessary to ensure comparison of values of the same range, which is an important prerequisite for multiple variable analysis with continuous variables (Long et al 2010).…”

Section: Data Preparationmentioning

confidence: 99%

Regional climate sensitivity of wetland environments in Rwanda: the need for a location-specific approach

2015

View full text Add to dashboard Cite

Wetlands are sustaining large communities of people in Rwanda where 10 % of its land surface consists of many local wetlands. Sustainable future management of these numerous wetlands requires a reliable inventory of their location and a dynamic quantitative characterization that allows assessment of their climate change sensitivity. The aim of this study was to assess the importance of climatic factors for determining wetland location at different regional scales. Wetland locations were analyzed and statistically modeled using their location factors with logistic regression. Wetland location probability was determined using topographic (elevation, slope), hydrological (contributing area) and climatic (temperature and rainfall) location factors. A wetland location probability map was made that demonstrated a calibration accuracy of 87.9 % correct at national level compared to an existing inventory, displaying even better fits at subnational level (reaching up to 98 % correct). A validation accuracy of 86.2 % was obtained using an independently collected dataset. A sensitivity analysis was applied to the threshold values used as cutoff value between wetland/non-wetland, demonstrating a robust performance. The developed models were used in a sensitivity scenario analysis to assess future wetland location probability to changes in temperature and rainfall. In particular, wetlands in the central regions of Rwanda demonstrate a high sensitivity to changes in temperature (1 % increase causes a net probable wetland area decline by 12.4 %) and rainfall (?1 % causes a net increase by 1.6 %). This potentially significant impact on wetland number and location probability indicates that climate-sensitive future planning of wetland use is required in Rwanda.

show abstract

Section: Data Preparationmentioning

confidence: 99%

Regional climate sensitivity of wetland environments in Rwanda: the need for a location-specific approach

2015

View full text Add to dashboard Cite

show abstract

“…After recording the total number of patches of forest (NP), we focused on the other three class landscape metrics, which had been previously reported as ecologically meaningful [22,63] and have been proven to be useful in describing patch spatial structure in a forested landscape context [64][65][66][67][68][69]: largest forest patch index (LFP), area weighted mean shape index (AWMSI) and patch cohesion index (COHESION). NP and LFP were selected, because they are related to forest fragmentation [33,66,67,70], defined as the breaking up of one large forest area into many smaller patches [71]. The largest forest patch index (LFP) quantifies the percentage of total landscape area comprised by the largest forest patch.…”

Section: Lpi Calculationmentioning

confidence: 99%

“…Current research suggests that the comparison of raw LPI values should be avoided [26], since they are sensitive to scale [27], land cover proportions [28], spatial resolution [29,30], spatial extent [31] and land-cover misclassification [32]. New methods for comparing LPI values may be useful in order to add statistical context to landscape pattern analysis [27,33]. As at present it is possible to clearly identify and quantify the differences between any two given dates, the statistical significance of the observed changes remains the most important and complex challenge with which to deal [28,34].…”

Section: Introductionmentioning

confidence: 99%

Quantifying Landscape-Scale Patterns of Temperate Forests over Time by Means of Neutral Simulation Models

Frate

Carranza

2013

IJGI

View full text Add to dashboard Cite

Several studies attempt to describe changes in the spatial patterns of forests over time, resorting to the comparison of landscape pattern indices (LPI), but new methods for quantifying landscape differences in a statistical context are necessary. In this paper, we quantified and assessed the statistical significance of the forests pattern changes, which have occurred since the end of WWII in Central Italy (Isernia). To do this; based on the proportion of forest cover (pi) and contagion (H) of three land cover maps (1954-1981-2006); we generated 100 forest maps with predictable results through the midpoint displacement algorithm. Then, for both observed and simulated maps, we computed a set of LPI (number of patches, cohesion, largest forest patch index and area weighted mean shape index) and we derived their empirical distributions; finally, we compared the empirical distributions using the non-parametric Kruskal-Wallis test. Our results show significant changes in the spatial pattern of forests and underline the process of natural forest re-growth, which, in the area, is constrained by -remnants‖ of traditional activities. The adopted approach could be extended to a large ensemble of landscapes and spatial scales and could become a standard procedure when comparing patterns in time.

show abstract

“…Forest maps provide explicit information on forest distribution [Stehman, 2012], which is the first step of fragmentation analysis. Forest, fragmentation consists of two interdependent components: forest loss and changes in spatial configuration [Neel et al, 2004;Long et al, 2010]. Accordingly, a proper interpretation of forest fragmentation needs to consider the interdependencies among these aspects [Neel et al, 2004].…”

Section: Introductionmentioning

confidence: 99%

Measuring forest fragmentation using multitemporal remotely sensed data: three decades of change in the dry Chaco

Carranza

Frate

Acosta

et al. 2014

European Journal of Remote Sensing

View full text Add to dashboard Cite

We introduce an approach based on remotely sensed data to summarize forest fragmentation over time, which specifically accounts for the interdependencies between landscape composition and configuration changes. The proposed method consists of five steps: i) multitemporal landscape sampling, ii) calculation of selected landscape pattern indices, iii) statistical comparison, iv) construction of sampled-based relationship spaces, and v) trajectory analysis. To show how the proposed method works in practice we examined the multitemporal fragmentation of the Arid Chaco forest in central Argentina during the period 1979-2010 using forest maps derived from Landsat images. As shown by our results, the approach provides a consistent framework for the interpretation of landscape structural changes over time.

show abstract

Characterizing forest fragmentation: Distinguishing change in composition from configuration

Cited by 51 publications

References 34 publications

Regional climate sensitivity of wetland environments in Rwanda: the need for a location-specific approach

Regional climate sensitivity of wetland environments in Rwanda: the need for a location-specific approach

Quantifying Landscape-Scale Patterns of Temperate Forests over Time by Means of Neutral Simulation Models

Measuring forest fragmentation using multitemporal remotely sensed data: three decades of change in the dry Chaco

Contact Info

Product

Resources

About