assessment of the flood occurrence potential in the upper teleajen

AAnnaalleellee UUnniivveerrssiittăăţţiiii ddiinn OOrraaddeeaa,, SSeerriiaa GGeeooggrraaffiiee Year XXXXVVII, no. 11//22001166 (June), pp. 2255--3333ISSN 11222211--11227733, E-ISSN 22006655--33440099 Article no. 226611110022--668877

http://istgeorelint.uoradea.ro/Reviste/Anale/anale.htm

AASSSSEESSSSMMEENNTT OOFF TTHHEE FFLLOOOODD OOCCCCUURRRREENNCCEE PPOOTTEENNTTIIAALL IINN TTHHEE UUPPPPEERR TTEELLEEAAJJEENN RRIIVVEERR BBAASSIINN

MMiihhaaeellaa BBOORRCCAANN National Institute of Hydrology and Water Management, 97 Bucharest – Ploiesti Street,

013686, Bucharest, Romania, e-mail: [email protected]

MMiihhaaii RREETTEEGGAANN

National Institute of Hydrology and Water Management, 97 Bucharest – Ploiesti Street, 013686, Bucharest, Romania, e-mail: [email protected]

Abstract: For a better management of the extreme situations generated by catastrophic hydrological events, we used a methodology regarding the estimation of the flash-flood occurrence potential in this particular river basin which integrates three physical elements (the slope of the terrain, soil texture and vegetation cover) that relate to precipitation run-off. The method also takes into account a fourth element that is the Land Cover/Use. For the purpose of this paper, we have replaced it with the 1% occurrence probability maximum rainfall.The results obtained by using this method is represented by an index which allows users to see which places are more pre-disposed to flash flooding. Key words: flood, rainfall, FFPI method, risk index, Teleajen

* * * * * *

INTRODUCTION AND STUDY AREA The genesis of flash-floods in the Upper Representative Teleajen River Basin is primarily related to

the climatic conditions, but also to other factors such as: geology, through the degree of permeability of rocks, soil, through its temperature and moisture content, vegetation, landforms, through the longitudinal slope of the river beds as well as the form and the size of the river basin. All these features influence the concentration time and the volume of the liquid run-off generated by a specific amount of rain. Within this context, the present paper aims, by integrating the main physical factors controlling the surface flow, to estimate the areas prone to accelerate run-off and therefore with high potential for flood occurance. The integrations has been achieved in GIS and the result is a quasi-statistic numeral index of flash flood potential specific to a geographic area. Four classes reflecting the flood occurence potential were established (very low, low, moderate and high). The FFPI method was initiated by Smith (2003) and applied for case studies in the United States (Abeyeta, 2009; Brewster, 2010; Kruzdlo & Ceru, 2010). In Romania it was adopted by Mătreață and Mătreață (2010), Teodor and Mătreață (2011), Minea (2011), Prăvălie and Costache (2013) and Zaharia & et al., (2013) who used it in order to estimate de flash flood potential on small and medium river basins. Borcan & Achim (2011) applied the method on a larger basin (Ialomița).

Corresponding Author

Mihaela BORCAN, Mihai RETEGAN

26

Figure1. The position within the Romanian Territory of the Upper Teleajen River Basin (Cheia)

Figure 2. The main sub-basins of the Upper Teleajen River Basin (Cheia)

Assessment of the Flood Occurrence Potential …

27

The study area is represented by the experimental Upper Teleajen River Basin (Cheia), located on the southern slopes of the Eastern Carpathians (The Curvature Group), around the sources of Teleajen River, the most importanttributary of Prahova River (figure 1).

The river basin covers an area of 41.3 km², being situated on the southern slopes of Ciucas Massif and ends at approximately 250 m downstream from the confluence of Tâmpa and Cheița creeks, which merge to form the Teleajen river (figure 2). The analyzed basin comprises 8 sub-basins with areas between 1.51 and 21 km² controlled by 8 river stations (table 1).

Cheia River Basin extends between 865 m and 1954 m, with a mean altitude of 1263 m.The river network consists of two main creeks, Cheița and Gropșoare (Tâmpa), along with other small tributaries (figure 2).

Table 1. Physical features of the river sub-basins compounded in the Upper Teleajen

Representative River Basin (Cheia) Source: NIHWM database

River River

stations Area (km2)

Mean alt. (m.a.s.l.)

Cheia Cheia 21 1320 Cucu Cheia 1.2 1096

Gropșoare Cheia 8.82 1290 Zăganu Cheia 2.87 1073 Baicu Cheia 1.18 1004 Tâmpa Cheia 13.9 1213 Ciobu Cheia 1.78 1085

Teleajen Cheia 41.3 1263

DATA AND METHODOLOGY In order to identify the vulnerable areas subject to processes generated by slope run-off, we

have used the Flash Flood Potential Index (FFPI) method. The FFPI method was proposed by Smith (2003) and used in Romania by several scientific

researchers such as Mătreață and Mătreață (2010), Teodor and Mătreață (2011), Minea (2011), Borcan and Achim (2011), Prăvălie and Costache (2013) and Zaharia et al., (2013).

FFPI method is based on cartographic algebra operations that result in the final thematic raster layer representing the arithmetic mean of the four thematic raster layers, which, in their turn, represent: maximum hourly rainfall amounts with an occurrence probability of 1%, the basin slopes, soils texture and land use. These four thematic layers have a spatial resolution of 30 m and represent the main physical-geographical factors that influence the catchment’s response.

We have obtained the values of the FFPI after integrating the spatial data in GIS software (e.g. ArcGis) and completing several steps, as described below:

- the collection of the necesary data (vector and raster) and their geographical referencing were geo-referenced in the Stereographic Projection 1970;

- slope: this layer was created using the Digital Elevation Model with a 30 m cell resolution interpolated on the basis of topographic maps (Constantin, 2011);

- vegetation Cover/Forest density: Using the Corine Land Cover 2006 data set which provides information regarding vegetation cover, we have determined the percentage of the entire Upper Teleajen river basin area which is covered by forests and by other classes of land use, thus obtaining a new thematic layer;

- soil type/Texture:The polygon type thematic layer representing soil texture was converted into a raster type thematic layer with a 30 m cell resolution, as well. For that purpose we used the pedological maps 1: 200 000;

- the maximum rainfall with a 1% occurance probability:the thematic layers representing the maximum hourly rainfall with a 1% exceeding probability were added by the kriging interpolation method;


28

- ArcMap software has the capability to re-classify raster datasets, which simply means assigning a new range of values to the field contained in the raster;

- according to Smith`s classification (2003) and based on the existing literature (Chendeș, 2007) in order to determine the vulnerability to flash-flood occurrence we have associated to each value of the above mentioned factors an integer-type value starting from a minimum value of 1 (representing the lowest risk) and increasing to a maximum value of 5 (representing the highest risk). This has been achieved by a re-classification of the four thematic layers. Class 1 signified a low participation in the flow formation, while class 5 corresponds to a high participation;

- by weight - averaging of all the factors we have already mentioned we obtained a new final thematic layer which helped us determine the flash-flood occurrence possibility thorough the Upper Teleajen river basin.

The integration of these parameters was accomplished in a GIS environment, through multiple operations that include digitization, interpolation, cropping, conversion, classification, re-classification and cartographic algebra.

RESULTS AND DISCUSSION By using the described methodology we estimated the FFPI values at the scale of the Upper

Teleajen river basin and consequently we were able to come up with a regionalization map. Slopes are an important component in the triggering of flash floods, due to their potential

for determining a rapid run-off. Experience suggests that any slope exceeding 30% leads to extremely quick run-off and a rapide response in local creeks and streams.

Figure 3. Slopes classification at 30 m resolution


29

Regarding slopes, Cheia River Basin is characterized by extremes. In the northern part, slopes are steep (25-30°). Very steep slopes can be encountered in the Zăganu Mountains. By contrast, the mountains in the southern part are characterized by less steep slopes, with a tilt of 15-20°.

The least steep slopes (below 5°) are encountered in Cheia Basin as well as on the erosion platforms located at 1500 – 1600 m (Chirușca Platform and Grohotișului Ridge).

Forest cover with land use are combined into one composite dataset. The values of this dataset are binned into 5 classes, each class representing a 20% degree of forest cover.

In this river basin forests represent approximately 67% of the area and are located at altitudes between 850 and 1400 m. Pastures, usually found at elevations higher than 1450 m, cover around 33% of the territory and are mainly found in the Grohotiș Mountains. In the analysed river basin, according to the degree of forest cover, the reduced-risk class for flash-floods occurrence possibility has not been identified.Using GIS methods, we have obtained a classification of the slopes and degree of forest cover correspondent to the associated vulnerability indexes at a spatial resolution of 30 m (figures 3 and 4).

Figure 4. Classification of the degree of forest cover at 30 m spatial resolution


30

The correspondence between these 5 classes of values used for quantifying slopes, degree of forest cover and the associated potential index is shown in table 2.

The classifying of slopes distribution and forest cover has been determined according to Miță P. (1997) classification.

Table 2. The risk categories for slopes, degree of forest cover and Flash-Flood Potential Index

Slopes (°)

Degree of forest cover (%)

Flash Flood Potential

Index 5 – 10 80 – 100 1

10 – 20 60 – 80 2 20 – 30 40 – 60 3 30 – 40 20 – 40 4 40 – 50 5 – 20 5

Sands and clay are very important components of soils when assessing run-off potential. Since texture, considered as a hydro-physical feature of soils, is a major factor with

significant influence on the maximum flow, our analysis has taken into consideration the synthetic table "Adapting the hydrological groups of soils to the Romanian soil texture classification" (Chendeș, 2007, quoted by Mătreaţă, 2011) and has been achieved by adding a new group, group E, composed of clayish textured soils and impervious areas (table 3).

Table 3. The adaptation of the hydrological groups of soils to the Romanian classification of texture (Chendeș, 2007, quoted by Mătreață, 2011)

Group Texture Flash Flood

Potential Transmission index

A Sandy; Sandy – sandy loam; Sandy – loamy sandy;

Sandy loam; Sandy loam – loam sandy; Loam sandy 1

B Sandy – loamy; Sandy loamy – loamy; Loamy sandy –

loamy; Loamy; Varied texture. 2

C Sandy loam – loam clayish; Sandy loam – clayish

loam; Loam clayish – clayish; Loam – loam clayish 3

D Loamy – clayish; Loam clayish; Clayish loam – clayish 4 E Clayish; Areas without infiltration 5

According to this table, soil types have been classified in 5 different classes, depending on

their texture (figure 5), and each class has been assigned an indicative, ranging from 1 (minimum risk) to 5 (maximum risk). The maximum amount of rain fallen during a 24-hour period is a very important characteristic of the pluvial regime, which can intensify the slope run-off and thus cause flash-floods. As a result of the calculations which have been carried out in order to obtain the maximum rainfall with a 1% occurrence probability, GIS helped us to classify the 24-hour rainfall quantities into several categories (figure 6).

By calculating the weighted-average of the four grids, we have obtained the final layer grid at a 30 m spatial resolution which shows the values of the index that help us estimate, taking into account the influence of the main physical-geographic factors, the flash-flood occurrence vulnerability potential (figure 7).

Please note that the results are obtained in the hypothesis of a quantity of rainfall with an exceeding probability of 1%.

In these conditions in the Upper Teleajen River Basin (Cheia) the largest share is held by average (34.7%) and high risk territories (30.3%), which cover extensive areas, being moderately and highly vulnerable to flash-flood occurrence, while the minimum (7%), low (20.5%) and maximum (7.5%) risk areas cover small, isolated territories (figure 7).


31

Figure 5. Soil texture classification at a 30 m spatial resolution

Figure 6. The classification of the hourly maximum rainfall with a 30 m spatial resolution


32

Figure 7. Map of the Flash-Flood Potential Index spatial repartition Assessment of the Flood Occurrence Potential in the Upper Teleajen River

We consider the results of FFPI method as being appropriate, as long as they were validated

by field observation. However, we must be awere of some limitations and possible errors which derive from the density of forest vegetations; the resolution at which the various investigated parameters are spatialized etc.

CONCLUSIONS The estimation of the territories with accelerated surface run-off allows the identification of

areas susceptible to flash floods. A method that can be used for this purpose relays on the determination of the FFPI as a syntetic index that integrates the main controls of surface flow.

The present study applied the FFPI method on the Representativ Upper Teleajen river basin which shows a medium degree of vulnerability to the risk associated with slope run-off and floods.

The lack of forest cover, as well as the very steep slopes and the clayish soils may ease the occurrence of flash-floods, especially if rainfall has a torrential character.

This method helped us to identify the areas which are not monitored from a hydrometrically point of view but which can be affected by severe hydrological phenomena.

This method also helps specialists and authorities to adopt certain measures that would mitigate the negative effects of flash-floods, by informing the exposed population regarding the potential risk the area in which they live has.


33

Mitigating the effects of flash-floods can be achieved by a continuous and viable monitoring of the variations of the liquid flow, especially on small water courses, which are the most susceptible to the occurrence of severe flash-floods.

Acknowledgements The authors wish to thank to the National Institute of Hydrology and Water Management

for the kidness to put at our disposal the hydrological data.

REFERENCES

Abeyta Amanda (2009), An Evaluation of the Flash Flood Potential Index Using Historical Flood Events, National

Weather Service, Albuquerque, NM, The 5th Symposium On Southwest Hydrometeorology, Albuquerque, New Mexico, September 30th to October 1st, 2009, Poster Session.

Borcan Mihaela, Achim Diana (2011), Estimarea potenţialului de producere a viiturilor în bazinul hidrografic Ialomiţa, Lucrările Conferinţei Ştiinţifice Anuale ale INHGA, 1–3 Noiembrie 2011.

Brewster J. (2010), A Discussion on the Utility of the High Resolution Flash Flood Potential Index for Customers and Partners, Eastern Region Flash Flood Conference, 2 – 4 June, Poster Session, NOAA’s National Weather Service, Available on http://www.erh.noaa.gov/bgm/research/ERFFW/presentations/june_04_2010/Concurrent%20 Sessions/NewTools/Brewster_Jim_Hires_FFPI.ppt

Chendeş V. (2007), Scurgerea lichidă şi solidă în Subcarpaţii de la curbură. Rezumat-Teză de doctorat. Institutul de Geografie. Academia Română.

Constantin S. (2011), Modele numerice altitudinale ale terenului disponibile liber, at http://www.geo-spatial.org/articole/modele-numerice-altitudinale-ale-terenului-disponibile-liber

Kruzdlo R., Ceru J. (2010), Flash Flood Potential Index for WFO Mount Holly/Philadelphia, Eastern Region Flash Flood Conference, 2–4 June, Poster Session, NOAA’s National Weather Service, Available on http://www.erh.noaa.gov/bgm/research/ERFFW/posters/kruzdlo_flashFloodPotentialIndexforMountHollyHSA.pdf

Mătreaţă M., Mătreaţă Simona (2011), Metodologie de estimare a potenţialului de producere de viituri rapide în bazine în bazine hidrografice mici, Comunicări de Geografie, Vol. XIV, Editura Universităţii din Bucureşti, Bucureşti.

Minea G. (2011), Bazinul hidrografic al râului Putna. Studiu de hidrogeografie. Rezumat – Teză de doctorat, Universitatea Bucureşti, Facultatea de Geografie.

Miţă P. (1997), Instrucţiuni pentru calculul scurgerii maxime în bazine mici. INMH, Bucureşti. Prăvălie R., Costache R. (2013), The vulnerability of the territorial – admnistrative units to the hydrological phenomena of

risk (flash floods). Case study: The Subcarpathian sector of Buzău catchment, Annals of the Univerisity of Oradea – Geography Series, TOM XXIII, nr. 1/2013 (June), p. 91.

Smith G. (2003), Western region flash flood project, AMS Conference, Session 6 GIS Applications. Teodor, S, Mătreaţă, Simona, (2011), A way of determining how small river basins of Someș River are susceptible to flash-

floods, Carpathian Journal of Earth and Environmental Sciences, 6, No. 1, p. 89 -98. Villarin G., Krajewski W.F., Ntelekos A., Georgakakos K.P., Smith J.A., (2010), Towards probabilistic forecasting of flash

floods: The combined effects of uncertainty in radar – rainfall and flash floods guidance, Journal of Hydrology, 394, 275-284.

Zaharia Liliana, Minea G., Toroimac Ioana Gabriela, Barbu Ruth, Sârbu I., (2013), Estimation of the Areas with Accelerated Surface Runoff in the Upper Prahova Watershed (Romanian Carpathians), Conference on Water Observation and Information System for Decision Support, BALWOIS 2012 - Ohrid, Republic of Macedonia.

*** (2006), Corine Land Cover Data Set. http://www.eea.europa.eu/data-and maps/data#c11=&c17=&c5=all&c0=5&b_start=0

*** Digital Terrain Model (DTM) http://www.geo-spatial.org/download/datele-landsat-etm-in-stereo701 *** The National Institute of Hydrology and Water Management Database. ***The Romanian soils map in electronic format, 1:200,000, I.C.P.A., Bucureşti. *** The Romanian geological map in electronic format, 1:200 000 (geo-spatial.org.).

Submitted: Revised: Accepted and published online September 16, 2015 December 12, 2015 January 27, 2016

Documents

assessment of the flood occurrence potential in the upper teleajen