Scholarly article on topic 'Identification of flash flood hazard zones in mountainous small watershed of Aceh Besar Regency, Aceh Province, Indonesia'

Identification of flash flood hazard zones in mountainous small watershed of Aceh Besar Regency, Aceh Province, Indonesia Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Flash flood" / "Hazard zones" / "Disaster management" / "Krueng Teungku watershed"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Azmeri, Iwan K. Hadihardaja, Rika Vadiya

Abstract Flash-floods develop at space and time scales that conventional observation systems were not able to monitor for rainfall, stream flow and sediment discharge. This condition resulted in greater casualties and tremendous economic losses. One of the regions in Indonesia affected by the flash floods was Aceh Besar Regency. It was located in Krueng Teungku watershed. The flash floods were recurring events, which occurred in 1987 and 2000. The disaster reocurred on January 2, 2013 at 19:30P.M. with the huge impact. This study aimed to analyze the factors affecting flash flood hazards and to obtain flash flood hazard zones at the Krueng Teungku watershed. The method used in this study was weighted overlay technique through Geographic Information System (GIS). The result revealed the information about the flash flood hazard zones at Krueng Teungku watershed as the model for the early warning. Through the development of this model, flood forecasting capabilities in the watershed without measuring devices can be improved. This paper provided the review of factors that affect the incidence of flash flooding, including the factors of peak discharge, slope, watershed shape, stream gradient, damming, drainage density, erosion, slope stability and reservoir volume. The information factors were expected as a contribution for research agencies and government (Aceh Disaster Management Agency) to guide the disaster risk reduction (DRR) activities of flash floods.

Academic research paper on topic "Identification of flash flood hazard zones in mountainous small watershed of Aceh Besar Regency, Aceh Province, Indonesia"

The Egyptian Journal of Remote Sensing and Space Sciences (2016) xxx, xxx-xxx

HOSTED BY

.«ïtmjj.-ggs&MÊm*

ELSEVIER

National Authority for Remote Sensing and Space Sciences

The Egyptian Journal of Remote Sensing and Space

Sciences

www.elsevier.com/locate/ejrs www.sciencedirect.com

RESEARCH PAPER

Identification of flash flood hazard zones in mountainous small watershed of Aceh Besar Regency, Aceh Province, Indonesia

Azmeria,% Iwan K. Hadihardajab, Rika Vadiyaa

aDepartment of Civil Engineering, Syiah Kuala University, Jl. Syech Abdur Rauf No. 7, Banda Aceh 23111, Aceh Province, Indonesia

b Faculty of Civil and Environmental Engineering, Institute of Technology Bandung (ITB), Jl. Ganesha No. 10, Bandung 40132, West Java Province, Indonesia

Received 22 July 2015; revised 19 October 2015; accepted 14 November 2015

KEYWORDS

Flash flood; Hazard zones; Disaster management; Krueng Teungku watershed

Abstract Flash-floods develop at space and time scales that conventional observation systems were not able to monitor for rainfall, stream flow and sediment discharge. This condition resulted in greater casualties and tremendous economic losses. One of the regions in Indonesia affected by the flash floods was Great Aceh Regency. It was located in Krueng Teungku watershed. The flash floods were recurring events, which occurred in 1987 and 2000. The disaster reocurred on January 2, 2013 at 19:30 P.M. with the huge impact. This study aimed to analyze the factors affecting flash flood hazards and to obtain flash flood hazard zones at the Krueng Teungku watershed. The method used in this study was weighted overlay technique through Geographic Information System (GIS). The result revealed the information about the flash flood hazard zones at Krueng Teungku watershed as the model for the early warning. Through the development of this model, flood forecasting capabilities in the watershed without measuring devices can be improved. This paper provided the review of factors that affect the incidence of flash flooding, including the factors of peak discharge, slope, watershed shape, stream gradient, damming, drainage density, erosion, slope stability and reservoir volume. The information factors were expected as a contribution for research agencies and government (Aceh Disaster Management Agency) to guide the disaster risk reduction (DRR) activities of flash floods.

© 2015 Authority for Remote Sensing and Space Sciences. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/

4.0/).

* Corresponding author. Tel.: +62 85260691773. E-mail addresses: azmeri@unsyiah.ac.id ( Azmeri), hardaja@si.itb. ac.id (I.K. Hadihardaja), rika.vadiya@yahoo.com (R. Vadiya). Peer review under responsibility of National Authority for Remote Sensing and Space Sciences.

1. Introduction

Indonesia, especially at most of the production sectors, has suffered heavily from the impacts of climate change and climate variability. Aceh was particularly vulnerable to multiple disaster-related shocks and climate change. This happened

http://dx.doi.org/10.1016/j.ejrs.2015.11.001

1110-9823 © 2015 Authority for Remote Sensing and Space Sciences. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

LEGEND

- KRUENG TENGKU RIVER

WATERSHED i WATERSHED 2 WATERSHED 3

WATERSHED 4 WATERSHED 5 WATERSHED 6 WATERSHED 7

Figure 1 The Study area - Krueng Teungku drainage watershed.

due to its location, density of population, high levels of poverty, high dependence on climate-sensitive resources, lack of awareness climate risks, and unplanned urbanization coupled with poor infrastructure (Islam et al., 2010). Most farmers, fishermen, small-businesses holders and other communities living in the low lying waterlogged areas were experiencing a wide range of climate variability because the global warming and

flood, especially flash floods, which were found to be more pronounced in those areas (Mirza, 2002).

Flash flooding was caused by a set of preliminary and triggering factors which determine their locations, frequency and magnitude. Excessive rainfall with a high intensity was the main source of flash flood in the hilly area, particularly resultant landslide in the area composed of unconsolidated rocks

95°31'30"E 95°32'30"E 95°33'30"E 95°34'30"E 95°35'30"E 95°36'30"E 95°37'30"E 95°38'30"E 95°39'30"E 95°40'30"E

_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_I_l_

5°38'30"N-

T-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-r

95°31'30"E 95°32'30"E 95°33'30"E 95°34'30"E 95°35'30"E 95°36'30"E 95°37'30"E 95°38'30"E 95°39'30"E 95°40'30"E

DEM MAP LEGEND

KRUENG TENGKU WATERSHED KRUENG TENGKU RIVER

ELEVATION

N 4 H 12,5 - 169,4346144 H 619,7687252 - 967,7541743

00,51 2 3 4 □ 169,4346145 - 374,1319374 ^M 967,7541744 - 1.752,427246

Kilometers i 374,1319375 - 619,7687251

Figure 2 Digital Elevation Model (DEM) of Krueng Teungku watershed.

5°38'0"N

5°38'0"N

5°37'0"N

5°37'0"N

5°36'0"N

5°36'0"N

5°35'0"N

5°35'0"N

5°34'0"N

5°34'0"N

5°33'0"N

5°33'0"N

5°32'0"N

5°32'0"N

5°31'0"N

5°31'0"N

5°30'0"N

5°30'0"N

5°29'0"N

5°29'0"N

5°28'0"N

5°28'0"N

5°27'0"N

5°27'0"N

(Sarker and Rashid, 2013). Flash flooding was one of the most common forms of natural disasters at the hilly region of Seu-lawah Mountains in Aceh Province, which occurred almost every year. The hilly area specifically at Krueng Teungku watershed almost annually suffered from the sudden inundation of water during the monsoon period due to the torrential rain. The daily rainfall occurred in the headwaters of Krueng

Teungku watershed during the flash flood reach to 125 mm height (Azmeri et al., 2015). This sudden inundation of water also carried silt, rocks, sediments and other debris causing damage to agricultural productions and causalities (Sarker and Rashid, 2013). There has been a relatively limited research on regional flash flood occurrence in a climatological sense. Although the occurrence of flash flooding was often associated

Table 1 Parameters of flash flood hazard (1).

Specific peak Average slope Watershed shape River gradient (%) Branching river Storage volume (m3) Class

discharge watershed (%) damming

(m3/dt/km2)

<0.58 <8 Elongated <0.5 No branching 0-49,018 Low

0.58-1.00 8-15 Elongated-Medium 0.5-1.0 Tributary of 49,018-98,036 Low-Moderate

main river branching

1.01-1.50 16-25 Medium 1.1-1.5 Main river branching 98,036-147,055 Moderate

1.51-5.00 26-45 Medium- Rounded 1.5-2.0 Main river/Bottle Neck 147,055-196,073 High-moderate

>5.00 >45 Rounded >2.0 Flood Tide 196,073-245,092 High

Source: Paimin et al. (2010) and analysis.

with "heavy precipitation," it has been acknowledged that this reason did not occurred alone. Hoyt and Langbein (1939) noted meteorological, climatic and physiographic influence on flood occurrence, including precipitation intensity, topography and soil characteristics (Modrick and Georgakakos, 2015).

This research utilized a case study on the recurrent flash floods at Krueng Teungku watershed in Great Aceh Regency of Aceh Province. The flash floods occurred repeatedly in the period of 1987, 2000 and 2013. The flash flood reocurred on January 2, 2013 at 19:30 P.M. The flood brought out soil and rock materials with a surface runoff of 1-3 m height. The flood that had occurred in the downstream of watershed (Beureunut Village) came from the amount of water from Krueng Teungku watershed. This disaster resulted in greater casualties and tremendous economic losses. As recurring disasters, this study aimed to identify flood hazard zones at Krueng Teungku watershed. Through zone identification, this study was expected to be used as disaster mitigation measures for the recurring floods in that region.

Most of the hydrological studies focusing on flash floods were based on morphometric parameters of the catchment area (Gabr and El Bastawey, 2015). According to Sumi et al.

(2013), developing an advanced methodology for the developing flood management issues through setting-up the flash flood potential hazard map was considered important.

Flash floods are one of the most destructive natural disasters in the world. Globally, flash floods caused more than 5000 deaths annually with mortality rate (computed as the number of fatalities divided by the number of affected persons). Flash floods also describe deaths attributed which were 50% of the flood-related to damage to property, infrastructure, and industry. The deadly nature of flash floods was caused by the incident that occurred in a short time associated with the characteristics of a small watershed (e.g. the occurrence of rainfall with high intensity produces a very large and fast discharge runoff from the mountains areas) (Modrick and Georgakakos, 2015). The location, topography and climate of the region lead to hydro meteorological hazards including the frequency of flash floods. It was rather hard to predict the occurrence of flash flood and very short duration events causing trouble to take mitigation measures. It resulted in an increase in the damage and loss.

2. Study area

The study area was Krueng Teungku watershed in Great Aceh Regency. The area focused on the mountain-to-foothill watersheds draining to the coast of Indian Ocean. Geographically, Krueng Teungku watershed is at 5°26'40"-5°38'20" North Latitude coordinates and 95°32'30''-95°40'50'' Longitude Cage coordinate (Fig 1). This region was elected as the area of study because of the influence of driving factors for the flash flood occurrence that were: climatology, geomorphology, and hydrology conditions.

3. Materials and methods

3.1. Digital elevation model (DEM)

Digital elevation model (DEM) is the digital representation of the earth surface terrain. It is an essential component in the hydrological models. Modern techniques of remote sensing provide tremendous potential for monitoring and managing dynamic changes in large surface water bodies; extracting hydrological parameters, and modeling the water balance (Memon et al., 2015). Digital elevation model (DEM) is the most important input of the hydrological modeling to get

Table 2 Parameters of flash flood hazard (2).

River density (km/km) Erosion hazard level Class

<0.25 <1.0 0.25-10 1.1-4.0 10-25 4.1-10.0 >25 >10.01 Low Moderate High-Moderate High

Source: Rahayu et al. (2009).

Table 3 Parameters of flash flood hazard (3).

Slope stability (Safety Condition Class

Factor)

F < 1.07 The likelihood of slope High

failure

1.07 < F < 1.25 The slope failures occurred Moderate

F > 1.25 The slope failure rarely Low

happened

Source: Bowles (1993:547).

LEGEND

- KRUENG TENGKU RIVER

| | KRUENG TENGKU WATERSHED BOUNDARY RIVER DISCHARGE CLASS

5°37'0"N

5°37'0"N

5°36'0"N

5°36'0"N

5°35'0"N

5°35'0"N

5°34'0"N

5°34'0"N

5°33'0"N

5°33'0"N

5°32'0"N

5°32'0"N

5°31'0"N

5°31'0"N

5°30'0"N

5°30'0"N

5°29'0"N

5°29'0"N

5°28'0"N

5°28'0"N

5°27'0"N

5°27'0"N

Figure 3 (a) Map of Krueng Teungku watershed specific peak discharge; (b) map of Krueng Teungku watershed slope. (c) Map of Krueng Teungku watershed shape; (d) map of Krueng Teungku watershed gradient.

Flood hazard maps. The precision of watershed calculation is directly dependent on the scale and precision of topographic maps (Elkhrachy, 2015). The DEM can significantly calculate both topographic parameters such as slopes, slope length and shape and aspects as well as hydrologic parameters such as flow direction, flow accumulation, watershed delineation,

stream networks, and flow length. Consequently, both flow accumulation and flow length would be used to extract the surface runoff (Islam et al., 2010). The DEM of Krueng Teungku drainage watershed was generated from contour maps as the digitized result from topographic maps 1:50.000 from Coordination Agency National Survey and Mapping

Fig. 3 (continued)

(Fig. 2). The manufacture of DEM Krueng Teungku was used to obtain the contour maps in raster format through Triangulated Irregular Networks (TIN) process. The data DEM was then used as input for the manufacturing process of watershed and sub-catchment of Krueng Teungku. Slope was estimated from the DEM using the embedded functions in ARCGIS.

3.2. Synthetic hydrographs and runoff generation

The annual maximum daily rainfall data during the period of 1982-2011 was obtained from rain gauge stations at Blang Bintang in Great Aceh Regency. Statistical parameter of average rainfall data (x) was 111.490 mm with the standard

95°33'30"E 95°34'30"E 95°35'30"E 95°36'30"E 95°37'30"E 95°38'30"E 95°39'30"E 95°40'30"E _i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_i_

И-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1—

95°33'30"E 95°34'30"E 95°35'30"E 95°36'30"E 95°37'30"E 95°38'30"E 95°39'30"E 95°40'30"E

LEGEND

- KRUENG TENGKU RIVER

| | KRUENG TENGKU WATERSHAED BOUNDARY WATERSHED SHAPE CLASS

LOW-MODERATE LOW HIGH

HIGH-MODERATE MODERATE

Fig. 3 (continued)

5°37'0"N

5°37'0"N

5°36'0"N

5°36'0"N

5°35'0"N

5°35'0"N

5°34'0"N

5°34'0"N

5°33'0"N

5°33'0"N

5°32'0"N

5°32'0"N

5°31'0"N

5°31'0"N

5°30'0"N

5°30'0"N

5°29'0"N

5°29'0"N

5°28'0"N

5°28'0"N

5°27'0"N

5°27'0"N

deviation 37.559 mm. Based on Chi-Square test, the rainfall distribution followed a normal distribution. The normal distribution required that the variance located between (x — s) and (x + s) was 68.27% and that between (x — 2s) and (x + 2s) was 95.44% also met the requirement.

The method used in calculating the flow rate was a method of synthetic unit hydrograph SCS. This method can estimate the flow rate to describe the characteristics combination of the watershed. Due to the limitation of observed discharge data at the field, it was expected that flood plan

KRUENG TEUNGKU RIVER GRADIEN MAP

Scale 1:90.000 0 0,5 1 2 3

I Kilometers

LEGEND

. KRUENG TENGKU RIVER KRUENG TENGKU WATERSHED BOUNDARY

GRADIEN RIVER CLASS

LOW-MODERATE

HIGH-MODERATE HIGH

Fig. 3 (continued)

5°37'0"N

5°37'0"N

5°36'0"N

5°36'0"N

5°35'0"N

5°35'0"N

5°34'0"N

5°34'0"N

5°33'0"N

5°33'0"N

5°32'0"N

5°32'0"N

5°31'0"N

5°31'0"N

5°30'0"N

5°30'0"N

5°29'0"N

5°29'0"N

5°28'0"N

5°28'0"N

5°27'0"N

5°27'0"N

was done using the unit hydrograph. The unit hydrograph analysis required rainfall hyetograph planning on certain rain durations as it only provides the daily rainfall data. Then, rainfall hyetograph was formed using Alternate Block Method (ABM). The rainfall in the upper watersheds during floods was about 125 mm/day. Based on the analysis of rainfall plan with specific return period, the rainfall causing flash

flooding was rainfall with a return period of 5 years in the estimation of flood discharge that occurs. The rain causing direct run off is called effective rainfall or excess rainfall which means the rain is not intercepting or infiltrating into the soil. Excess Rainfall Hyetograph (ERH) is a graph showing the relationship between effective rainfall and time. Infiltration model with SCS Curve Number (CN) estimates

Table 4 Classes of flash flood hazard at Krueng Teungku watershed (1).

Sub Specific peak Class Average Class Watershed shape Class Gradient Class

catchments discharge slope of the river

(m3/dt/km2) watershed (%)

1 6.176 High 6.1588 Low Elongated Low 0.1 Low

2 9.866 High 6.5002 Low Elongated Low 0.91 Low-Moderate

3 7.417 High 13.4807 Low-Moderate Elongated-Medium Low-Moderate 1.586 High-Moderate

4 12.398 High 8.9895 Low-Moderate Rounded High 2.611 High

5 15.363 High 7.4293 Low Medium Moderate 2.298 High

6 22.047 High 14.4701 Low-Moderate Medium-Rounded High-Moderate 4.473 High

7 25.526 High 25.179 High-Moderate Rounded High 8.801 High

Table 5 Classes of flash flood hazard at Krueng Teungku watershed (2).

Sub Branching river Class River density Class Erosion Class Slope Class

catchments damming hazard Level stability (F)

1 No branching Low 1.69 Moderate 6.02 High-Moderate 1.035 High

2 Main river branching Moderate 1.05 Moderate 10.16 High 1.035 High

3 Main river High-Moderate 0.75 Moderate 8.06 High-Moderate 1.035 High

4 Main river branching Moderate 0.35 Moderate 23.74 High 1.023 High

5 Main river branching Moderate 0.52 Moderate 9.94 High-Moderate 1.108 Moderate

6 Tributary of main Low-Moderate 0.72 Moderate 3.86 Moderate 1.108 Moderate

river branching

7 Tributary of main Low-Moderate 0.50 Moderate 3.83 Moderate 1.108 Moderate

river branching

rainfall in particular as a function of its cumulative rainfall, land use and soil type (Chow et al., 1988).

E, = (200 + 87logwI, )P,

Qn = £ PmUn-

where Qn is direct run off of water discharge on a pulse to n; Pm is effective rainfall in pulses to m; U is ordinate hydrograph unit.

3.3. Erosion

USLE (Universal Soil Loss Equation) method was developed by Wischmeier and Smith (1978) where the USLE method was used to estimate the annual average erosion by using kinetic energy of the rainfall approach. Azmeri et al. (2015) stated that the large erosion can be calculated using USLE method as followed:

A = R.K.LS.C.P (2)

where A is the amount of soil loss per unit area (ton/ha/year); R is rainfall erosivity factor; K is soils erodibility index; LS is slope length factor; C is crop management factor; and P is soil conservation factor.

Wischmeier and Smith (1958) found that the product of the kinetic energy of the raindrop and the maximum intensity of rainfall over the duration of 30 min, in a storm, was the best estimator of soil loss. Rainfall factor (R) is also expressed as:

R = REJ30 (3)

where Ei is rainfall kinetic energy (J/m2); 130 is rainfall intensity (mm/h).

According to Das (2002) the rain fall kinetic energy (Ei) can be calculated using this equation:

where I¡ is average rainfall intensity (cm/h); Pi is rainfall depth (cm).

3.4. Slope stability

According to Bowles (1993) on a non-horizontal ground surface, component of gravity is likely to move down to the ground. If the gravity component is so great that resistance to fricative which could be deployed by ground in the field of avalanche is exceeded, there will be a sliding slope. Krueng Teungku watershed is largely consisted of steep slopes, so it is important to analyze the slope stability. Parameters resulted in slope stability analysis is a form of plane failure and the safety factor. Safety factor with the Bishop method used the following equation

RM _ ELtc'ß +(N — uß) tan u'] DM ~ Pn=1 [W sin a + kW (cos a — |)] 4

Table б Classes of flash flood hazard at Krueng Teungku watershed (3).

Sub catchments

Storage volume

0.00 168.500 13.110 110.291 245.092 20.122 23.081

High-Moderate Low

Moderate High Low Low

95°32'0"E 95°33'0"E

95°37'0"E 95°38'0"E 95°39'0"E 95°40'0"E

95°33'0"E

95°34'0"E

95°35'0"E

1-1-1-1-1-1-1-1-1-1-1-

95°36'0"E 95°37'0"E 95°38'0"E 95°39'0"E 95°40'0"E 95°41'

KRUENG TENGKU WATERSHED DAMMING MAP

Scale 1:100.000 00,51 2 3

I Kilometers

LEGEND

._. KRUENG TENGKU RIVER

| | KRUENG TENGKU WATERSHED BOUNDARY

DAMMING CLASS

^ HIGH-MODERATE MODERATE

LOW-MODERATE LOW

Figure 4 (a) Damming map of Krueng Teungku watershed; (b) river density map of Krueng Teungku watershed. (c) Erosion hazard level map of Krueng Teungku watershed; (d) slope stability of the riverbank map of Krueng Teungku watershed.

95°41'0"E

5°38'0"N

5°37'30"N

5°37'30"N

5°37'0"N

5°37'0"N

5°36'30"N

5°36'30"N

5°36'0"N

5°36'0"N

5°35'30"N

5°35'30"N

5°35'0"N

5°35'0"N

5°34'30"N

5°34'30"N

5°34'0"N

5°34'0"N

5°33'30"N

5°33'30"N

5°33'0"N

5°33'0"N

5°32'30"N

5°32'30"N

5°32'0"N

5°32'0"N

5°31'30"N

5°31'30"N

5°31'0"N

5°31'0"N

5°30'30"N

5°30'30"N

5°30'0"N

5°30'0"N

5°29'30"N

5°29'30"N

5°29'0"N

5°29'0"N

5°28'30"N

5°28'30"N

5°28'0"N

5°28'0"N

5°27'30"N

5°27'30"N

5°27'0"N

5°27'0"N

5°26'30"N

5°26'30"N

where F is safety factor; RM is resistance moment (kN/m2); DM is tension moment (kN/m2); c' is soil effective cohesion (kN/m2); b is width of slice to-i (kN); N is normal force at slice (kN); u is pores water pressure (kN/m2); /' is angle of effective stress friction (o); W is weight of slice (kN); a is angle of inclination of the tangent that surface to the middle of wedge side

to the horizontal line (o); kW is horizontal seismic force (kN); hc is the altitude of the centroid of the slice from midpoint of the base of the slice (m); R is the radius of the circular arc (m); A is hydrostatic force (kN); a is the vertical distance from the hydrostatic force against the central moments (m); n is the amount of slices.

Fig. 4 (continued)

3.5. Parameters of flash flood hazard

Map making of flood disaster required parameter analysis of the flood hazards. Among the causes were specific peak discharge, average slope watershed, watershed shape, review gradient, branching river damming, storage volume, river density, erosion hazard level, and slope stability. Each of these param-

eters formed a different class. Tables 1-3 give the parameters of flash flood hazard as follows.

4. Results and discussion

This model has subjectively subdivided the whole drainage watershed into various sub-catchments that were linked

Fig. 4 (continued)

together at a number of outlets that were located along the main channel (Fig. 3). The flash flood zoning model has defined seven main outlets along the main channel of Krueng Teungku drainage watershed. Each outlet receives surface runoff water from a number of sub-catchments. Accordingly, the whole fluvial system of Krueng Teungku was classified into seven major accumulative hydrologic zones. These zones vary

in sizes and morphologic characteristics, which in due course vary in the quantity of runoff they deliver and the depth of runoff in the main channel. Besides run off, all factors that affect the occurrence of flash floods were also analyzed from the seven zones.

Flash flooding at Krueng Teungku watershed was influenced by several factors, namely: peak discharge, slope, size

Fig. 4 (continued)

and shape of watershed, gradient of the river, damming, density of the river, erosion, slope stability and the storage volume. Tables 4-6 give the classes of flash flood hazard at Krueng Teungku watershed.

As the sample, it was given the estimation of specific peak discharge for sub-catchments 7 (upstream). Discharge of

Synthetic Unit Hydrograph (SUH) SCS was obtained in between 0 and 52.847 m3/s, from the comparison of discharge q with a peak discharge qp and time t with peak time Tp. Excess Rainfall Hyetograph (ERH) was obtained from cumulative of rainfall and abstraction with the value between 0 and 8.382 mm. Use equation (1) and base flow of Krueng Tengku

95°33inilF Q5°33'30"F Q5°34'0"F Q5°34

Q5°3£'30"F Q5°37'0"F Q5°37'30"F Q5°3S'n"F Q5°3S'3n"F Q5°3Q'0"F Q5°3Q'30"F Q5°4n'n"F Q5°40

Q5°33'0"F Q5°33'30"F Q5°34'0"F Q5°34'30"F Q5°35'0"F Q5°35'30"F Q5°3R'n"F Q5°3R'3n"F Q5°37'0"F Q5°37'30"F Q5°3S'n"F Q5°3S'3n"F Q5°3Q'0"F Q5°3Q'30"F Q5°4n'n"F Q5°4n'3n"F Q5°41

DAMMING VOLUME OF KRUENG TEUNGKU WATERSHED MAP

о 0,5 1

I Kilometers

LEGEND

-KRUENG TEUNGKU RIVER

I I KRUENG TEUNGKU WATERSHED BOUNDARY

MODERATE

I HIGH-MODERATE HIGH

35'30 "N

32'3n"N

2Q'3n"N

Figure 5 Storage volume map of Krueng Teungku watershed.

river basin of 0.90 m3/s to obtain flood peak discharge with the value of 452.179 m3/s. Area of sub-watershed 7 by the value of 17.742 km2 was generating the specific peak discharge of 25.633 m3/s/km2. This peak discharge value was at the level of high-class flash flood threat. The specific peak discharger

for all sub-watershed was in the range of 6.176-25.633 m3/s/ km2, which led each sub-catchment was at the high-class level of flash flood threat.

Specific peak discharge at each of sub-catchments was at high level of flash flooding hazard. The peak discharge was

Table 7 The values of parameters that determine the value of Safety factor (F) at soil sample A.

c' B N u W a kW hc R A A

(kN/m2) (kN) (kN) (kN/m2) (°) (kN) (°) (kN) (m) (M) (kN) (M)

1.49 0.479 0.004 0 26.83 1.48 30 0 0.23 4.76 0 4.54

also affected by the time of concentration (tc) at very fast each of sub-catchments resulting in a rapid accumulation of runoff water at each outlet of sub-catchments.

This was due to the length and slope of the flow at each Krueng Teungku sub-catchment which was very high (Fig 3a). Moreover, the mining activities require the establishment of local community needs including housing, domestic water pipelines, roads and other facilities. Therefore, it is essential to identify the environmental threats and quantify the surface runoff on the purpose of safety activities. Climatic conditions in the upstream Krueng Teungku watershed trigger rainfall intensity, and therefore also affect the flash flood hazard. This condition requires identification of existing infrastructure as appropriate mitigation measures (Montz and Gruntfest, 2002). The result of the peak discharge parameters was appropriate with the study result which was conducted by Firmansyah and Kadarsetia (2010) that show peak discharge affects the flash flooding hazard at Jember regency, East Java where rainfall at some rain stations at Dinoyo watershed included in the high rainfall. A sharp increase in river discharge in a short of time due to the high rainfall can potentially cause flash floods.

Slope of each sub-catchments were obtained from watershed delineation (Fig 2). According to the slope, the level of flash flood hazard in Krueng Teungku divided into three classes namely low, low-moderate and high-moderate classes. On the downstream and middle (sub-catchments 1, 2 and 5) were flat lands with a slope of <8% and in the category of low-moderate class. In the central area (sub-catchments 3, 4 and 6) were wavy and undulating lands with slope of 8-15% so it was included in the category of low-moderate class. The headwaters (sub-catchment 7) mountainous areas with slopes of between 16% and 25% thus it included in the category of high-moderate class. Viewed from the slope side, flash flood potentially happened at sub-catchment 7 (Fig 3b). The steep slope watershed accelerates the accumulation of flow rate. Furthermore, steep and hilly area has the potential in moving the ground and clogging up the river which results in flash floods.

The result of this study was confirmed by the result of the research conducted by Firmansyah and Kadarsetia (2010)

which showed that the worst affected areas due to flash floods in Jember regency East Java was caused by the presence of undulating areas with slope of 15-20%. Krueng Teungku watersheds shape were varied including oval, slightly oval, round, moderate, somewhat round and round, so in general was dominated by oval and round shapes. According to the shape of watershed, sub-catchments of 4 and 7 were at the high level of flash flood threat with a round shape of watershed (Fig 3c). Round shape may increase the flow rate because the rain fall distance at the observation point to the watershed outlet was shorter. The time required by the rain water to the outlet was also shorter, so it increases the time for peak discharge and flow accumulation thus has the potential for flash floods.

Gradient of Krueng Teungku for each sub-catchments were obtained from the result of watershed delineation. According to the gradient of the river, the levels of flash flood hazard were divided into four classes, namely low, low-moderate and highmoderate and high classes (Fig 3d). On the downstream and middle (sub-catchments 4, 5, 6 and 7) Krueng Teungku river has gradient of river about >2% included into steep gradients and high-class categories.

In the area slightly to the downstream (sub-catchment 3) has gradient of river between 1.5% and 2% and it includes the category of high-moderate class. Regions slightly downstream (sub-catchment 2) have a gradient of river between 0.5% and 1% and it includes the category of low-moderate class. The downstream area (sub-catchment 1) has a gradient of river about <0.5% and it includes the category of low class. Based on general gradient of river category, Krueng Teungku watershed has a high level of flash flood hazard with extensive sub-catchment about 79.967 km2 (74.32 %). The steep slope of the river can accelerate the rate of water flow to the downstream and potentially cause flash floods.

Based on the potential for damming, sub-catchment 1 which was in the downstream (estuary) does not occur as natural damming so that there was no threat to flash floods. The natural damming at sub-catchments 2, 4 and 5 occurred at the branch of the main river. The natural damming at sub-catchment 3 can happen at the main river. At sub-catchments

Table 8 Flash flood hazard weight at Krueng Teungku watershed.

No. Parameters Parameters weight (%) Indicators Indicators weight (%) Total weight (%)

1 Characteristics of the river 33.33 River gradient 70 23.33

River density 30 10.00

2 The accumulation of river water 33.33 Specific peak discharge 30 10.00

Storage volume 30 10.00

Branching river damming 20 6.67

Watershed shape 20 6.67

3 Erosion 33.33 Average slope watershed 40 13.33

Erosion hazard level 30 10.00

Slope stability 30 10.00

Total 100.00

FLASH FLOOD HAZARD LEVEL OF KRUENG TEUNGKU WATERSHED MAP N 0 0,5 1 2 3 4 ^ LEGEND KRUENG TENGKU RIVER | | KRUENG TENGKU WATERSHED BOUNDARY FLASH FLOOD HAZARD CLASS Щ HIGH I HGH-MODERATE | MODERATE

Figure 6 Flash flood hazard level map of Krueng Teungku watershed.

У5"350"Ь

У5" 36'0"t

У5"37'0"Ь

У5" 3У'0"Ь

Ц5"40'0"Ь

а5"41'0"Ь

5°37'0"N

5°37'0"N

5°36'30"N

5°36'30"N

5°36'0"N

5°36'0"N

5°35'30"N

5°35'30"N

5°35'0"N

5°35'0"N

5°34'30"N

5°34'30"N

5°34'0"N

5°34'0"N

5°33'30"N

5°33'30"N

5°33'0"N

5°33'0"N

5°32'30"N

5°32'30"N

5°32'0"N

5°32'0"N

5°31'30"N

5°31'30"N

5°31'0"N

5°31'0"N

5°30'30"N

5°30'30"N

5°30'0"N

5°30'0"N

5"2y'30"N

5°2y'30"N

5°2y'0"N

5°2y'0"N

5°28'30"N

5°28'30"N

5°28'0"N

5°28'0"N

5°27'30"N

5°27'30"N

5°27'0"N

5°27'0"N

5°26'30"N

5°26'30"N

6 and 7, natural damming can happen at the branch of the main river (Fig 4a). The area of 5,687.28 ha (52.86%) allows the natural damming at the river branching.

Teungku Krueng river system consists of the main river and tributaries that drain water from downstream to upstream, and it is included in the third-order. The more the order of the river illustrates the more the branching of tributaries.

The river density at Krueng Teungku watershed was varied, from 0.35 to 1.69. River density value is included in the medium density class. The river density makes the river flow over the rocks with softer resistance and accumulates transport (Fig 4b).

As a sample count of erosion rate value at sub-catchments 7, according to rainfall intensity /30 was about 5.51 mm/h and

the kinetic energy of the rain fall E¡ was around 438.47 J/m2 from the maximum rainfall depth P¡ occurred from durations of 5-100 min, resulted in the rainfall erosivity factor R was 1100.51 (Eqs. (3) and (4)). The soils erodibility index Kfor this kind of alluvial soil was 0.47, with a length of slopes L was 2,978.065 m and slope S was 6.159% which produced slope length factor LS was 0.79. Land use in the form of dry land agriculture providing crop management factor C was 0.001, and land without processing giving soil conservation factor P was 1. Based on the five parameters of erosion (Eq. (2)) the obtained the erosion rate was 162.58 ton/ha/year. Erosion Hazard Level (EHL) of Krueng Teungku sub-catchment was obtained by comparing the value of erosion rate that occurs with tolerable erosion (27 ton/ha/year) (Azmeri et al., 2015). Erosion Hazard Level at sub-catchments 1 was about 6.02 including high-moderate level of classes.

The flood hazard level was analyzed based on erosion rate using Universal Soil Loss Equation (USLE) and Geographic Information System (GIS). High-level of flash flood hazard were at the area of sub-catchments 2 and 4. This can be caused by a steep slope, the type of soil susceptible to erosion, land-use of the area was dominated by dry farming land which were susceptible to erosion as well (Fig 4c).

Sliding slope of the riverbank was obtained by collecting three samples of undisturbed soil at 3.1 m height of the riverbed for each with a distance of 3.7 for samples A soil, a height of 3.5 m and a distance of 2 m for sample B soil, and a height of 3.5 m and a distance of 1.2 m for sample C soil. Soil samples were tested in Soil Mechanics Laboratory at Faculty of Engineering, Syiah Kuala University, Indonesia on the purpose of obtaining soil density parameters, friction angle and cohesion. The slope angle for the three soil samples was about <54°, then the whole landslide occurred on base circle slope. Table 7 shows the parameter value that determined the value of Safety factor (F) as in Eq. (5) on the slice 1 for soil sample A.

By using Bishop, the values of these parameters were iterated between the resistance moment and the tension moment in order to obtain safety factor (F). Safety factor value 1.035 of SF (A), 1.023 of SF (B) and 1.108 of SF (C). The value of SF for those three soil sample ranged from 0.6 up to 1.058 (Bowles, 1993) so Krueng Teungku riverbank has a great potential of collapse occurrence (Fig 4d). The collapse riverbank can close the river channel of Krueng Teungku which potentially creates natural damming. High water discharge may cause a collapse of the natural dam and cause flash floods.

The drainage network analysis is generally performed to understand the prevailing geological variation, topographic information and structural set of a basin and their interrelationship. Remote sensing and GIS based drainage basin evaluation has been carried out by the number of researchers for different terrains and it is proved to be a very scientific tool for the generation of precise and updated information for characterization of drainage basin parameters (Singh et al., 2014).

Dam height for each sub-catchment took the minimum height of riverbank. Natural damming can lead to the accumulation flow at any time the dam was destroyed and cause the flash floods. Flash flood hazard level based on the volume of water dammed in the sub-catchments Krueng Teungku was a high level of flash flood hazard (69.18 of the total sub-catchment (Fig 5)).

Flash flood at the Krueng Teungku watershed was affected by several factors, namely peak discharge, slope, watershed shape, river gradient, damming, river density, erosion, slope stability and the storage volume. According to TDMRC (2013), flash flood hazard parameters distinguished on the river slope parameter by weight of 33.33%, the accumulation of river water by weight of 33.33%, and landslide 33.33%. Table 8 gives the weight parameters and weight indicators of flash floods that can be used in this study.

Krueng Teungku watershed has a high potential of flash flood treat with the percentage of 11.08% of the wide watershed, high-moderate level of 87.22% of broad watersheds and moderate level of 1.70%. The biggest flash flood hazard level was at sub-catchment 4 which has a level erosion hazard on sub-watershed 4 included in the high category (Fig 6).

5. Conclusions and recommendations

Flash flooding at Krueng Teungku watershead was affected by several factors, namely peak discharge, slope, watershed shape, river gradient, damming, river density, erosion, slope stability and the storage volume. Those factors almost have the same points. Krueng Teungku watershed has high level of flash flooding hazard with the percentage of 11.08%, highmoderate level of 87.22% and moderate level about 1.70%. Distribution of high-level flash flood hazard was at the villages of Paya Kameng, Beurandeh, Meunasah Kulam which lied on sub-watershed 4. Efforts to address the flash flood hazard in the form of conservation either mechanically or chemically and can be done at the vegetative sub-catchment with high erosion hazard level (sub-catchments 2 and 4). The sub-catchment with the high level of gradient of river (sub-catchments of 4, 5, 6 and7) needs site planning and groundsill detail to flatten the bottom slope so as to reduce the flash flood hazard.

Conflict of interest

The authors declare no conflict of interest. Acknowledgments

Deep thanks are dedicated to The Postgraduate Incentive Grants No. 526/UN11/S/LK-PNBP/2014 on June 5, 2014 from Syiah Kuala University for all the funding and facilities provided for this research.

References

Azmeri, Yulianur, A., Rizalihadi, M., Bachtiar, S., 2015. Hydrologic response unit analysis using avswat 2000 for keuliling reservoir watershed. Aceh Int. J. Sci. Technol. 4 (1), 32-40. Bowles, J.E., 1993. Physical and Geotechnical Properties of Soils.

Erlangga, Jakarta. Chow, V.T., Maidment, D.R., Mays, L.W., 1988. Applied Hydrology, International Edition. McGraw-Hill Book Company Inc, New York.

Das, G., 2002. Hydrology and Soil Conservation Engineering.

Prentice Hall of India, Private Limited, New Delhi. Elkhrachy, I., 2015. Flash flood hazard mapping using satellite images and GIS tools: a case study of Najran City, Kingdom of Saudi Arabia (KSA). Egypt. J. Remote Sens. Space Sci.

Firmansyah, M.N., Kadarsetia, E., 2010. Risk identification of flash flood disaster in Jember Regency, East Java.

Gabr, S., Bastawesy, M.E., 2015. Estimating the flash flood quantitative parameters affecting the oil-fields infrastructures in Ras Sudr, Sinai, Egypt, during the January 2010 event. Egypt. J. Remote Sens. Space Sci.

Islam, A.E., Hermas, E.S., Bastawesy, M.E., 2010. GIS-modelling of the spatial variability of flash floodhazard in Abu Dabbab catchment, Red Sea Region, Egypt. Egypt. J. Remote Sens. Space Sci. 13, 81-88.

Memon, A.A., Muhammad, S., Rahman, S., Haq, M., 2015. Flood monitoring and damage assessment using water indices: a case study of Pakistan flood-2012. Egypt. J. Remote Sens. Space Sci. 18, 99-106.

Mirza, M.M.Q., 2002. Global warming and changes in the probability of occurrence of floods in Bangladesh and implications. Global Environ. Change 12, 127-138.

Modrick, T.M., Georgakakos, K.P., 2015. The character and causes of flash flood occurrence changes in mountainous small basins of Southern California under projected climatic change. J. Hydrol. Reg. Stud. 3, 312-336.

Montz, B.E., Gruntfest, E., 2002. Flash flood mitigation: recommendations for research and applications. Environ. Hazard 4, 15-22.

Paimin, Sukresno, Purwanto, 2010. Quick Detection of Sub watershed Degradation, Research and Development Center for Conservation

and Rehabilitation of Forestry Research and Development Agency (FORDA), Bogor.

Rahayu, S., Widodo, R.H., Van Noordwijk, M., Suryadi, I., Verbist, B., 2009. Water Monitoring of Watershed. World Agroforestry Centre, Bogor.

Sarker, A.A., Rashid, A.K.M.M., 2013. Landslide and Flashflood in Bangladesh. In: Disaster Risk Reduction Approaches in Bangladesh, Disaster Risk Reduction. Springer, Japan.

Singh, P., Gupta, A., Singh, M., 2014. Hydrological inferences from watershed analysis for water resource management using remote sensing and GIS techniques. Egypt. J. Remote Sens. Space Sci. 17, 111-121.

Sumi, T., Saber, M., Kantoush, S.A., 2013. Japan-Egypt hydro network: science and technology collaborative research for flash flood management. J. Disaster Res. 8 (1), 28-36.

Tsunami and Disaster Mitigation Research Center (TDMRC), 2013. Final Report of Disaster Areas Mapping in Gayo Lues Regency. Syiah Kuala University.

Wischmeier, W.H., Smith, D.D., 1958. Rainfall energy and its relationship to soil loss. Trans. Am. Geophys. Union.

Wischmeier, W.H., Smith, D.D., 1978. Predicting Rainfall Erosion Losses. A guide to conservation planning, USDA Hanbook, 537, Washington DC.