FRACTAL ANALYSIS IN ESTIMATING THE FRAGMENTATION DEGREE OF AGRICULTURAL LANDS
Fractal analysis was used to evaluate the degree of agricultural lands fragmentation. An area in the Western Plain, Romania was studied. The image was taken with the RapidEye satellite system. From the basic image, 10 polygons with equal resolution of 735 x 840 pixels were selected. For each studied...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
University of Agricultural Sciences and Veterinary Medicine, Bucharest
2020-01-01
|
| Series: | Scientific Papers Series : Management, Economic Engineering in Agriculture and Rural Development |
| Online Access: | https://managementjournal.usamv.ro/pdf/vol.20_3/Art56.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Fractal analysis was used to evaluate the degree of agricultural lands fragmentation. An area in the Western Plain, Romania was studied. The image was taken with the RapidEye satellite system. From the basic image, 10 polygons with equal resolution of 735 x 840 pixels were selected. For each studied polygon, the total surface (TS), the number of plots (PN), the average plot area (APA), and the fractal dimension (D) were determined. Fractal analysis was performed using the box counting method. The correlation analysis revealed a moderate, negative, correlation between PN and APA (r=-0.776), strong negative correlation between D and PN (r=-0.871), respectively a very strong, positive, correlation between D and APA (r=0.935). APA variation according to PN was most faithfully described by a smoothing spline model. Variation of fractal dimension D according to PN was described by a polynomial equation of degree 2, in conditions of R2=0.946, p<< 0.01, and the variation of D according to the APA was described by a polynomial equation of degree 2 in conditions of R2=0.939, p<< 0.01. Based on fractal dimension (D), regression analysis made it possible to estimate PN under conditions of R2=0.818, p=0.0025, F=15.782, respectively APA variation under conditions of R2=0.984, p<< 0.001, F=214.86. Based on PCA, PC1 explained 89.441% of variance, and PC2 explained 10.559% of variance. Cluster analysis led to the grouping of the studied cases, in condition of Coph.corr=0.988. |
|---|---|
| ISSN: | 2284-7995 2285-3952 |