Contents lists available at ScienceDirect

Acta Astronautica

Acta Astronautica 67 (2010) 1280–1288

0094-57

doi:10.1

$ Thi� Cor

E-m

journal homepage: www.elsevier.com/locate/actaastro

An analysis of coastal topography and land cover changes atHaeundae Beach, South Korea$

Jiyeon Yang a,n, Doochun Seo a, Hyosuk Lim a, Chuluong Choi b

a Satellite Information Research Institute, Korea Aerospace Research Institute (KARI), 115 Gwahangno, Yuseong-Gu, Daejeon 305-333, Republic of Koreab Department of Geoinformatic Engineering, Pukyong National University (PKNU), 599-1, Daeyeon 3-Dong, Nam-Gu, Busan 608-737, Republic of Korea

a r t i c l e i n f o

Article history:

Received 16 February 2010

Received in revised form

3 May 2010

Accepted 9 June 2010Available online 8 July 2010

Keywords:

Aerial photograph

Coastline

Coastal monitoring

Land cover map

Soil loss

65/$ - see front matter & 2010 Elsevier Ltd. A

016/j.actaastro.2010.06.013

s paper was presented during the 60th IAC i

responding author. Tel.: +82 42 870 3952; fax

ail addresses: sp0731@hanmail.net, sp0731@n

a b s t r a c t

Environmental changes and increased development have led to sharp rises in coastal

erosion. To understand and address this important social, environmental, and economic

problem, a long-term data collection and monitoring system is necessary. This study

analyzed changes in coastal topography and land cover for the past 60 years at

Haeundae Beach, the most well-known beach in South Korea. First, aerial photographs

were used to analyze the coastal topography. The coastline was extracted from digitized

images, and tide levels were corrected using sounding and global positioning system

(GPS) survey data. The beach area was then calculated and analyzed in terms of change

in area over the 60-year study period. Second, land-cover changes were analyzed using

detailed land-cover maps created by on-screen digitizing techniques and soil loss in the

area near Haeundae Beach was estimated and compared with land use. Results showed

that the coastline has moved farther inland and that in general the beach area has

decreased. The beach area in 2005 was 35% less than that in 1947. The interception of

the sand supply by artificial developments is thought to be a major cause of the coastal

erosion. The study results can be applied to long-term coastal monitoring and

environmental change projects and are expected to be available as part of a coastal

information system.

& 2010 Elsevier Ltd. All rights reserved.

1. Introduction

The problem of coastal erosion has sparked bothinterest and controversy. Landslides and soil loss con-tribute to coastal erosion and have led to costly losses ofproperty and recreational lands. The decline in beach sandhas particularly harmed the tourism industry in Busan.Although tourism has been increasing annually at Busan’sHaeundae Beach, the most popular beach in South Korea,sand loss has progressively decreased the width of thebeach. While the Busan Metropolitan City Hall andSuyeong ward office expend vast funds annually to

ll rights reserved.

n Daejeon.

: +82 42 860 2605.

ate.com (J. Yang).

preserve the beach, accurate data on beach erosion, whichis necessary to formulate an effective plan for combatingcoastal erosion are lacking.

The present study used aerial photographs to quanti-tatively analyze changes in the coastal environment atHaeundae Beach over the past 60 years. Aerial photo-graphs have many advantages over satellite-derivedimagery, including high resolution and lower costs.

Several previous studies have examined coastal erosion.Choi et al. [1] analyzed coastal topography using topographicdata, digital elevation models (DEM), and image processingmethods. Park and Jeong [2] extracted a boundary line fortidelands using both aerial photographs and Landsat-5satellite images and generated a DEM using tide-level data.Frihy et al. [3] analyzed the coastlines of lagoons and islandscovering a period of 86 years using Landsat satellite images(from 1978, 1983, 1990, 1993, and 1995) and topographic

J. Yang et al. / Acta Astronautica 67 (2010) 1280–1288 1281

maps (from 1909, 1922, 1944, and 1955). El Raey et al. [4]used long-term surveys and hydrographic profiles to examinechanges in beach accumulation/erosion patterns and volume,demonstrating the effectiveness of using images and topo-graphic maps to analyze long-term changes. Ron et al. [5]extracted a three-dimensional representation of the Lake Erieshoreline using high-resolution IKONOS satellite panchro-matic stereo images and compared the extracted coastlinewith the real coastline. Kim [6] analyzed patterns of coastaltopographic changes at Kawanganri Beach, South Korea usingsounding data, global positioning system (GPS) survey data,and aerial photographs.

The present study aims at the following: (1) monitoringlong-term change of Haeundae Beach through analyses of itscoastal topography and land cover changes; and (2) ananalysis of coastal erosion and it causes. This study attemptsto construct a database related to long-term changes in thecoastal topography at Haeundae Beach by extracting thecoastline and computing the beach area based on aerialphotographs from the past 60 years. The correlation betweencoastal topographic change and land-cover change is alsoexamined using information on land cover and soil loss inareas near Haeundae Beach.

Fig. 1. Study area is Haeundae Beach in South Korea. Haeundae Beach is

located at the southeast end of the city of Busan, and it is the famous

beach where over the ten million people visit every year. Unfortunately,

beach has been eroded steadily so the Busan Metropolitan City Hall

expends vast funds annually to preserve the beach.

2. Data and methods

Haeundae Beach is located on the eastern part of Busanwhich is situated around a mountain (Jangsan Mountain)having peak elevation of 634 m (Fig. 1). Haeundae Beachis the most well known domestic beach, with over 10million tourists visiting every year. The sandy beach isabout 1.8 km in length about 35–50 m wide, having anarea of 72,000 m2.

The beach is artificially filled with new sand annuallyto battle continuous erosion. In spite of these efforts, thebeach area continuously decreases.

In this study, we used thirteen aerial photographs(from 1947 to 2005) to analyze the change of HaeundaeBeach during the last 60 years. This study can be dividedinto two analyses, coastal topographic change and landcover change. A flow chart illustrating the study metho-dology is presented in Fig. 2.

In this study we analyzed coastal topographic changesquantitatively by detecting changes in the beach’s areaand width using aerial photographs. We also analyzed thetrend in land cover change through statistical, spatial andthree-dimensional analyses after creating a land covermap, and we calculated soil loss using the land cover map.In this manner, we could monitor Haeundae Beach’senvironmental changes by analyzing coastal topographyand land cover changes together.

3. Results and considerations

3.1. Analysis of coastal topographic changes

Aerial photographs were obtained from the Depart-ment of the Cadastral Survey, Busan City Hall, and thenscanned. Table 1 provides information on the photos.

Image processing, including interior and exteriororientation, was conducted using ERDAS Imagine 8.6(Leica Geosystems Geospatial Imaging, Norcross, GA,USA) based on the generated coverage and DEMs createdby Arcview 3.3 and ArcInfo 8.0.1 (ESRI, Redlands, CA,USA). Image Analyst was used to produce digitized orthoimages of the coastline. Fig. 3 shows the extractedcoastlines.

It was necessary to correct the coastline according tothe tide levels, as each aerial photo was taken at adifferent time. To compute the correction value, contoursassociated with tide level and mean sea level (MSL) wereextracted from sounding survey data and calculated todetermine the correction value. Assuming a gentle coast-line slope, correction values for 1982, 1985, and 1988were calculated from sounding survey data in 1980 and1992, and correction values for 1995 and 1997 werecalculated from sounding survey data in 1994. Thecorrection values for 2000, 2002, and 2005 were com-puted from sounding survey data in 2004 and GPS surveydata. Because sounding and GPS survey data wereunavailable in 1947, 1970, 1975, and 1979, aerial photoswere used for these years. Fig. 4 shows the sounding andGPS survey points.

Fig. 5 presents an example of the extracted contoursbased on the sounding and GPS survey data. The coastline

Fig. 2. Methodology of the study consists of two analyses, coastal topographic change and land cover change analysis. Coastline and beach area change

are analyzed through the coastal topography change analysis, and soil loss computed through the land cover change analysis. Finally, the correlation of

two results was analyzed.

Table 1Index of aerial photos used in the study (FL: focal length; TL: tide level; MSL: mean sea level).

Date Time Scale No. Camera type FL (mm) TL (cm) TL-MSL (cm)

25 October 1947 1

25 July 1970 10:00 1 151.246

25 May 1975 11:22 1:6000 4 Wild UAG 152.00

14 July 1979 10:19 1:6000 4 Wild UAG 152.00

2 March 1982 11:08 1:6000 4 Wild UAG 152.00 77 +12

2 May 1985 9:56 1:6000 4 Wild 152.36 30 �35

5 May 1988 13:27 1:6000 3 Wild 15/4 UAG 153.35 66 +1

24 January 1992 11:55 1:6000 3 Wild 15/4 UAG 153.40 117 +52

24 May 1995 10:34 1:6000 3 Zeiss 153.10 33 �32

25 May 1997 11:51 1:6000 4 Wild 15/4 UAGA-F 153.59 100 +35

13 April 2000 9:00 1:6000 3 Wild 15/4 UAG-S 153.71 48 �17

26 May 2002 10:11 1:6000 7 Zeiss 152.54 105 +40

14 May 2005 13:55 1:6000 3 Wild 15/4 UAG-S 152.84 74 +9

J. Yang et al. / Acta Astronautica 67 (2010) 1280–12881282

extracted from the aerial photos was corrected based oncalculated values that considered tide levels and yearlycoastline change.

Fig. 6 illustrates the corrected coastlines and coastlinechanges for the western and eastern beaches. Generally,the coastline has moved farther inland, and the width of

J. Yang et al. / Acta Astronautica 67 (2010) 1280–1288 1283

the beach has decreased. At the center of the beach, thecoastline has remained relatively unchanged, but majorchanges have occurred on both the western and easternsides. The width of the beach has also decreased greatly infront of the Chosun Beach Hotel in west side.

Fig. 3. Each coastline was extracted on the aerial photo using only visual

analysis without consideration of tide level. These are examples of

extracted coastlines in 1947 (a), 1970 (b), 1979 (c), 1982 (d), 1988 (e),

1995 (f), 2000 (g), and 2005 (h).

Fig. 4. Survey data is needed for extract the coastline with consideration of the

1980 (a), 1994 (b), and 2004 (c) and GPS survey points in 2005 (d).

The beach area was analyzed based on standardizedboundary lines. The back of the beach was considered tobe located at the shore-protection structures built in1975. The coastline was considered to be the front of thebeach. Analysis results indicate that the beach area hasgenerally decreased since 1947, as presented in Table 2.The beach area in 2005 was 35% less than that in 1947.

Fig. 7 graphs the change in beach area by year. A largearea of beach has disappeared, and the trend line suggeststhat the decrease will continue. In 1992, the beach areaincreased sharply. It is difficult to determine the cause ofthis phenomenon based on the data used here. The beacharea may experience seasonal changes; the photo in 1992was taken in winter (January) whereas those for the otheryears were taken in summer.

3.2. Analysis of land-cover changes

A land cover map was created to analyze changes inland use near Haeundae Beach over the 60-year studyperiod. To analyze yearly changes in land cover, aerialphotographs were orthorectified, and orthophoto mosaicswith equal boundaries were created. As shown in Fig. 1,the study area is near Haeundae Beach and has a total areaof 2,713,260.5438 m2. A detailed land-cover map wascreated by visual analysis of the mosaics from 1947 to2005 (Fig. 8). Thirteen land-cover maps were made basedon classifications by the Ministry of Environment of Korea.Changes in each land-cover type were examined by areaand for trends. Fig. 9 shows changes in main land-covertypes by area.

Analysis of changes by land-cover type from 1947 to2005 showed that the urban area has increased con-tinuously and greatly due to development around theChuncheon Stream and within the overall urban zone.

The area of barren land increased from 1947 to 1975,decreased from 1975 to 1982, increased greatly from 1982to 1997, and has decreased since 1997. The area of ricepaddies has decreased since 1970. Fields increased from1947 to 1979 but have decreased since 1979. The stream

tide level. These figures are distributions of the sounding survey points in

Fig. 5. This figure shows that the process to extract contour lines between mean sea level (MSL) and tide level from the sounding survey data;

(a) sounding survey analysis, (b) TIN generation, and (c) contour line extraction.

Fig. 6. This figure shows that the results of the coastline correction with consideration of tide level and the coastline change; (a) in whole beach, (b)

zoomed in west side, and (c) zoomed in east side. Upper lines face downtown (north direction), and lower lines face the sea (south direction). All upper

lines fixed on same location to compare each coastline. As time passed, the coastline moved toward inland. This means that beach area has been steadily

eroded.

Table 2Haeundae Beach area by year.

Year Photo (m2) Tide (m2) Total (m2)

1947 88658.5 – 88658.5

1970 68751.3 – 68751.3

1975 68334.6 – 68334.6

1979 59854.1 – 59854.1

1882 56847.9 2081.1 58929.0

1985 64967.6 �6123.6 58844.0

1988 55775.9 142.9 55918.8

1992 61400.9 14118.5 75519.3

1995 59963.7 �8767.4 51196.3

1997 56958.6 9316.4 66275.0

2000 61734.7 �2453.5 59281.2

2002 55295.8 5102.9 60398.7

2005 56687.5 1206.1 57893.6

Fig. 7. Beach area which is computed from extracted coastline is on the

decrease. Unusually, beach area of 1992 is increased. It is assumed to be

due to the seasonal effect. Among 13 experimental data, only an aerial

photo in 1992 was taken on winter season (refer to Table 1 for detail).

J. Yang et al. / Acta Astronautica 67 (2010) 1280–12881284

area has decreased continuously and now accounts foronly a small area.

Three-dimensional (3D) visualization allows for aneasier visual analysis of land cover. A building layer was

created by digitizing buildings from each aerial photo andoverlaying this layer onto aerial photos using ArcScene 8.3(ESRI), as shown in Fig. 10. In 1947, the study area was

Fig. 8. Land cover maps were generated by digitizing on each aerial photo. These are examples of generated land cover map in 1947(a) and 2005(b).

Fig. 9. This graph shows change in area of various land types. In 1947, most of area is farmland such as rice paddy, barley field, vegetable garden, etc. On

the other hand, in 2005, most farmland has been disappeared and many buildings have been built in that area. And Chuncheon Stream which served as a

natural sand source to the beach has been disappeared because of road construction.

J. Yang et al. / Acta Astronautica 67 (2010) 1280–1288 1285

predominantly covered by fields and a few buildings.However, urban development surged around 1970,expanding the area covered by buildings. Since 1990,urban development has included the building of high-riseapartments, multiplexes, and resort accommodations,making Busan a main seaside resort city.

Renard et al. [7] computed soil loss using therevised universal soil loss equation (RUSLE) to compen-sate for factors connected to the soil surface and soilmoisture in the widely used universal soil loss equation(USLE). The present study used RUSLE to investigatethe effects of soil loss on beach area based on changes inland cover.

Here, from constructed thematic maps (5�5-m reso-lution) of RULSE factors, soil loss was computed using anArcView 3.3 Spatial Analysis and 3D Analysis (ESRI). TheRUSLE equation to compute soil loss (ton/ha/yr) is asfollows, Eq. (1):

A¼ R� K � LS� C � P ð1Þ

A is the average soil loss per unit area, R the rainfall factor,K the soil erodibility factor, LS the topographic factors,C the crop-management factor, and P the erosion-controlpractices factor.

This study used the rainfall factor (R) estimated byPark et al. [8] based on data from the Korean Meteor-ological Administration from 1973 to 1996 and the soil-erodibility factor (K) issued by the Korea Water ResourcesAssociation [9]. Topographic factors (LS) were computedto explain the effect on soil loss; this computationincluded the slope length factor (L) and slope gradientfactor (S) from the topographic factors (LS) equationprovided by Moore and Burch [10]. The crop-managementfactor (C) was taken from Oh and Jung [11]. The P value forslope angles was determined using the National Institutefor Disaster Prevention’s [12] erosion-control practicefactor (P) to represent the effects of soil erosion accordingto field cultivation type and slope.

Fig. 11 shows thematic maps for each RULSE factor.Fig. 12 and Table 3 illustrate the results of soil loss and a

Fig. 10. 3D building mapping was carried out using aerial photos and

digital topographic maps. These are examples of generated 3D building

map in 1945 (a), 1970 (b), 1975 (c), 1985 (d), 1995 (e), and 2005 (f).

Topography was described using digital elevation model (DEM) and

height of each building was approximately computed using only

building floor information from digital topographic maps.

Fig. 11. Thematic maps were generated for the calculation of soil loss;

(a) K-factor, (b) LS-factor, (c) C-factor, and (d) P-factor. K-factor was

computed using 1:25,000 digital detailed soil map and LS-factor was

calculated using DEM extracted from 1:5,000 scale digital topographic

map. C-factor and P-factor was estimated using land cover map

generated by digitizing on each aerial photo.

Fig. 12. This figure shows that distribution patterns of potential soil loss

(a) and change in soil loss by year (b). According to the Bathymetric

survey results (refer to Table 3 for detail), soil was deposited between

1980 and 1994, on the contrary, eroded between 1994 and 2005 at the

bottom of the sea. Bathymetric survey result is corresponding with

change pattern of soil loss on the graph (b).

J. Yang et al. / Acta Astronautica 67 (2010) 1280–12881286

bathymetric survey. Because the study area is a resortarea located in the coastal zone, and the slope of the studyarea is gentle with little forest cover, total soil loss wasless than 1 ton.

By year, the lowest amount of soil loss occurred in1947. This value indicates that soil loss under naturalconditions was less than soil loss caused by artificialdevelopments. In other words, soil loss increased becauseforest cover decreased and barren lands increased duringurban development. Soil likely flowed out from construc-tion sites, and soil loss increased with greater urbandevelopment. When compared to soil loss in 1947, soilloss in 1970 showed a sharp, five-fold increase due toactive urban development.

Completion of urban construction projects then led toa respite in soil loss, although increases again occurred

from 1985 to 1997 with the construction of numerousmultiplexes, high-rise apartments, and hotels as well asexpansion of the highway network. Analyzing the totalchange in soil loss, construction projects partly increasedsoil loss in some periods. In general, soil loss appears tobe decreasing, and artificial factors have largely affectedsoil loss.

Sounding data from the study area show an accumula-tion in marine topography from 1980 to 1994 but erosionfrom 1994 to 2005. These results indicate that soil loss isstrongly related to marine erosion, as shown in Fig. 12(b)and Table 3. Although marine topography eroded from1980 to 2005 as indicated in Table 3, soil loss notdecreased, as shown in Fig. 12(b). This situation mayhave resulted from the construction of sewage pipe anddisposal facilities; soil in the stream that may otherwisehave accumulated in the marine area was instead held inthe sewage facility. The sewage disposal system is thusconsidered to have affected marine sediment accumula-tion and erosion.

3.3. Comparative analysis between coastal topography and

land-cover changes

The relationship between coastal topography andland-cover changes was examined for the 60-year studyperiod. Data from 1992 were excluded because the beacharea in that year was much higher than in all other years.Seasonal effects and urban development in 1992 mayaccount for this higher value. Fig. 13 shows the analysisresults for the beach area, stream area, and soil loss.

Table 3Bathymetric survey results.

Year Points Study area (m2) Cell size (m2) Avg. depth (m) Water area (m2) Water vol. (m2)

1980 11245 2798�2435 2�2 14.49 6039196 87555332

1994 3772 2798�2435 2�2 14.29 6048012 86471876

2005 21257 2798�2435 2�2 14.89 6028168 89763216

Fig. 13. This figure shows that the correlations of beach area versus

stream area (a) and soil loss (b). The correlation (R2=0.7814) between

the stream area and beach area, and the correlation (R2=0.695) between

soil loss and beach area suggest that changes in land cover are related to

changes in beach area.

J. Yang et al. / Acta Astronautica 67 (2010) 1280–1288 1287

A high correlation was observed between the beacharea and stream area (R2=0.7814) from 1947 to 1995,during which time the stream area decreased greatly. Thisresult suggests that the Chuncheon Stream had served asa natural sand source to the beach and that the stream’sdecline led to a decrease in beach replenishment and,accordingly, a decrease in the beach area.

Beach area and soil loss were correlated (R2=0.695)from 1970, when urban development began, to 2005,suggesting that changes in land cover near HaeundaeBeach led to changes in the beach area. That is, the beach’ssand is influenced by soil carried from the land near thebeach, as well as soil (silt) that washes in from sea. Duringthis process, the amount of soil that is carried from thenearby land is influenced by changes in land cover.

4. Conclusions

Employing digital aerial photogrammetry techniques,this study examined changes in coastal topography andland cover near Haeundae Beach and obtained thefollowing main results:

�

The coastline has moved farther inland over the past60 years. The western and eastern sides of HaeundaeBeach in particular have experienced more erosion andcoastline changes than the central area of the beach. � The beach area in 2005 was 35% less than that in 1947,

based on an analysis of the coastline extracted fromdigitized aerial photos. Therefore, it is estimated thatHaeundae Beach has eroded continuously over the pastsix decades.

� Urban and barren land use types have increased 65%

and 55%, respectively, since 1947. In contrast, forestshave decreased by 50% since 1947. An even moredramatic change has occurred in the area of fields,which have decreased 99% since 1947.

� Total soil loss was less than 1 ton because the slope of

the study area is gentle and there are few forests.Compared to soil loss in 1947, soil loss in 1970 sharplyincreased five-fold due to active urban development. Inother words, soil loss increased because forest areadecreased and barren land increased with urbandevelopment. These results suggest that soil loss hasbeen deleteriously affected by artificial development.

� The correlation (R2=0.7814) between the stream area

and beach area, and the correlation (R2=0.695)between soil loss and beach area suggest that changesin land cover are related to changes in beach area.

The use of the geographic information system and remotesensing technology allowed a quantitative analysis of changesin coastal topography and land cover. Because there has beenno precedent analysis of the change in area of HaeundaeBeach over the past 60 years, this study’s data on changes inthe beach’s area and width may be useful for long termanalysis of the changes to Haeundae Beach. The presentresults are expected to be available as part of a coastalinformation system. Also, this study may serve as aconsultable basis for various kinds of research on HaeundaeBeach. Additionally, the developmental process of HaeundaeBeach can be viewed by using this study’s data on long termland cover changes in the region of the beach. We believethat this study will be helpful to the urban planning andtourism industries, it provides evidence that the beach’s areais influenced by land cover changes in nearby areas. We havethus presented another way to study coastal topographicchange.

J. Yang et al. / Acta Astronautica 67 (2010) 1280–12881288

Further study should analyze seasonal changes incoastal topography using aerial photographs taken duringvarious seasons and the study area should be expanded toinclude the Suyeong River and new urban zones.

References

[1] C.U. Choi, J.H. Kwak, S.K. Park, The prediction of coastal topographicdeformation using change, Korean Society of Surveying, Geodesy,Photogrammetry and Cartography 13 (2) (1995) 169–176.

[2] S.W. Park, J.C. Jeong, Extraction of DEM in the southern tidal flat ofKanghwa Island using satellite images, Journal of the GIS Associa-tion of Korea 11 (1) (2003) 13–22.

[3] O.E. Frihy, Kh. Dewidar, M.S.M. Nasr, M.M. El Raey, Changedetection of the northeastern Nile Delta of Egypt shoreline changes,spit evolution, margin changes of Manzlal lagoon and its islands,International Journal of Remote Sensing 19 (10) (1998) 1901–1912.

[4] M. El-Raey, S.H. Sharaf El-Din, A.A. Khafagy, A.I. Abo Zed, Remotesensing of beach erosion/accretion patterns along Damietta-PortSaid shoreline, Egypt, International Journal of Remote Sensing 20(6) (1999) 1087–1106.

[5] L.I. Ron, D.I. Kaichang, M.A. Ruijin, 3-D Shoreline extraction fromIKONOS satellite imagery, Marine Geodesy 26 (2003) 107–115.

[6] H.Y. Kim, Comparison of two-dimensional and three-dimensionalmonitoring techniques to analyze long-term coastal topographicchanges, Master’s Thesis, Graduate School of Pukyong NationalUniversity, 2005, pp. 20–24.

[7] K.G. Renard, G.R. Foster, G.A. Weesies, D.K. McCool D.C. Yoder,Predicting soil erosion by water: a guide to conservation planningwith the revised universal soil loss equation (RUSLE), USDAAgriculture Handbook Number 703. Washington, DC, 1996.

[8] J.H. Park, H.S. Woo, C.K. Pyun, K.G. Kim, A study of distribution ofrainfall erosivity in USLE/RUSLE for estimation of soil loss, in:Proceedings of the Korea Water Resources Association Conference,33(5), 2000, pp. 603–610.

[9] H.S. Woo, C.W. Kim, Prediction of soil loss and plan of sedimenta-tion basin from the development, in: Proceedings of the KoreaWater Resources Association Conference, The Sixth Water ResourceEngineering Workshop, 1998, pp. 1101–1179.

[10] I. Moore, G. Burch, Physical basis of the length-slope in factoruniversal soil loss equation, Soil Science Society of America Journal50 (1986) 1297–1298.

[11] J.H. Oh, S.G. Jung, Potential soil loss prediction for land resourcemanagement in the Nakdong River basin, Korean Society of RuralPlanning 11 (2) (2005) 9–19.

[12] National Institute for Disaster Prevention (Republic of Korea), Studyon Sediment Yield Estimation Due to Land Development (I), 1998,pp. 161–226.