The Natural Environment and Historical Water Management of Angkor, Cambodia Matti Kummu Department of Water Resources, Helsinki University of Technology, Espoo, Finland e-mail: [email protected] Abstract: This paper gives an overall view to the present natural environment and historical water management of Angkor; the ancient capital of the Khmer Empire, between the 9th and 15th centuries. The water system of the region may have played a key role in the city operation but is not clearly understood how it worked and how and when it collapsed. The natural environment and hydrology of the area and the key elements of the historical water management will be introduced. The research methods and mathematical models applied to the area are also briefly presented. Key words: history of water, Angkor, Tonle Sap Lake, Cambodia, hydrological modelling, hydrodynamic modelling 1. INTRODUCTION The World Heritage site of Angkor, in Cambodia, is famous for its monumental religious constructions, which have been studied in great detail over the last century. Recent research has uncovered an equally impressive feature of this medieval capital: an extensive hydraulic network stretching across a thousand square kilometres. Although Angkor’s hydraulic nor hydrology is not nearly as well understood as its religious architecture, this system may have played a key role in the city operation and may also be implicated the demise of the urban complex at the beginning of the 15th century. Prior to an analysis of the historical water management, the present natural environment and hydrology should be well understood. The second and third chapters give an introduction to present hydrometeorological, geographical and hydrological conditions of the area. The study area has been divided to three hydraulic zones by elevation to understand better the hydraulics and hydrology of the area. Also, the Angkor’s relation to Tonle Sap Lake and the history of the lake based on the current knowledge, the key features of the historical water management and the methods that have been applied to the research will be presented. The historical water system of the area is investigated with the help of radar images, digital elevation model, aerial photos, field studies and coring. That data will be used to create a three-dimensional (3D) hydrodynamic model for investigating water levels and currents, inundation of the flood plains and suspended sediments in the systems. This study is a part of the Greater Angkor Project, a collaborative project between the University of Sydney, the Ecolé française d’extreme-Orient (EFEO) in Siem Reap, and APSARA, the Cambodian body responsible for overseeing the monuments at Angkor. The study will be conducted in cooperation with MRCS/WUP-FIN Tonle Sap Modelling Project (Mekong River Commission Secretariat/Water Utilization Program) and is part of the Global Changes and Water Resources Project in Helsinki University of Technology.

2. NATURAL ENVIRONMENT OF ANGKOR REGION Angkor is situated north of the Tonle Sap Lake (also known as Great Lake), in the northwest of Cambodia in Southeast Asia (Figure 1). In the north the Kulen Mountain region sets a boundary to the area and watersheds while from the south area is bounded by the Tonle Sap Lake, the biggest fresh water lake in Southeast Asia. The plain terrain between Lake Tonle Sap and Kulen mountains is very shallow with the average slope of 0.1 % while the elevation varies from 3 to 60 meters above the mean sea level (a.m.s.l.). The Kulen Mountains rise to heights between 300 m and 400 m from the mean sea level having the highest point is at the elevation of 487 m (or 494 m, depends the source).

∴ Angkor

Figure 1

Map of Cambodia (Modified from Encarta, 2001: ref Keskinen, 2003)

The main temple area around Angkor Wat and Angkor Thom is situated approximately 20 km north of the Tonle Sap Lake (Figure 2) and the elevation varies between 20 and 30 m a.m.s.l. To the south of the temples, the area is bounded by the lake and the 8-12 km wide floodplain. The floodplain reached up to some 11 m a.m.s.l. The level of the lake ranges between 1 m at the end of the dry season and 6 – 10 m a.m.s.l. at the peak of the rainy season. The area of study region is around 2885 km2 and it is situated between latitudes 13˚04 - 13˚44 and between longitudes 103˚36 - 104˚13.

2

Figure 2 The situation of Angkor area related to the Tonle Sap catchment area (left) and the study area in more detailed (JICA, 1999b; Evans, 2002; DMA, 1963; and Certeza Surveying, 1964).

The climate in the study area is tropical, being dominated by seasonal winds or monsoons. The wet southwest monsoon arrives around May with heavy clouds and thundershowers, and usually continues until November, with rain occurring almost daily during this season. The dry northeast monsoon normally starts from November and continues until April (JICA, 1999a). 2.1. Hydrometeorology of the Angkor The lack of data is the main problem for the hydrometeorological analysis. During the Pol Pot and Khmer Rouge period no data were collected neither on the Angkor area nor any other part of Cambodia. The systematic measuring work for e.g. temperature, evaporation, humidity, and wind began as late as 1998. Prior to this only precipitation was measured more or less systematically. Average annual rainfall in the Tonle Sap basin ranges between 1,050 and 1,850 mm/year and can exceed 2,000 mm/year within the higher catchment elevations (MRCS/UNDP, 1998). The mean value of annual station-averages has been 1370 mm/year for the period 1950-2000 and 1425 mm/year for 1980-2000. Data from a total of five stations were used for the calculations (MRCS/WUP-FIN, 2002a). Annual rainfall in Siem Reap town varies between 900 and 1,800 mm/year with an average of 1,425 mm/year (data from 35 years, 1922-2002; MRCS, 2003). The wet season (mid April to October) brings on average some 88 % of the annual rainfall in the Siem Reap region. Occasionally annual exceptions occur particularly bordering the traditional wet season in the months March and November, in which heavy storms occur and large rainfall is recorded.

3

Frequency analysis was conducted to the precipitation rates in Siem Reap weather station. The Chi-Square and Kolmogorov-Smirnov tests showed that the annual mean precipitation is normally distributed (Figure 3). Hence, the method for normally distributed data can be used for the frequency analysis (Chin, 2000). Estimated magnitudes for 50; 100; 1,000; and 10,000-year annual rainfalls are presented in Figure 3.

Number of outcomes

8 7 6 5 4 3 2 1 >1800

17001800

16001700

15001600

14001500

13001400

12001300

11001200

0

Range (mm/a)

Year 50 100 1000 10000

Precipitation (mm/a) Min Max 1035 1816 983 1868 838 2013 718 2133

Figure 3 Precipitation distribution for Siem Reap station. Analysis data from 35 years were used. Right: estimated magnitudes of the 50; 100; 1,000; and 10,000-year minimum and maximum annual rainfalls.

The precipitation varies a lot inside the Angkor region. In total, eight measurement stations are situated in the study area and four stations around the region like presented in Figure 4 (one station not in the map, situates north from Angkor Chum). Measurements in many stations began as late as 2001. Thus, only results from years 2001 and 2002 have been used for the analysis of the precipitation variation inside the area. Highest rainfall occurs at the Kulen Mountain with an average of 1854 mm/year. The lowest rainfalls occur on the Tonle Sap floodplain at Phnom Krom and on the high plain at Banteay Srei with an average rainfall of 1183 mm/year and 932 mm/year, respectively. Thus, the precipitations at the lake’s shore line and on high plain are only 64 % and 50 % (respectively) of the precipitation in the mountain region. The results (Figure 4) indicate large variations in rainfall within the study area and show that the rain falls down locally. Presently only data from two years are available for the analysis. Also, the measurement facilities and conditions vary between the stations. Therefore, the analysis requires further work as more data collected. The present data give only an approximation of the precipitation variation in the region.

4

Figure 4 Precipitation isolines for the study area (Yearly average precipitation data from years 2001-2002 have been used. The measurement stations presented). Hatched area indicates the peak of the year 2000 flood. (DMA, 1963; Certeza, 1964; MRCS, 2003; and Evans, 2002).

The annual average evaporation rate in the region is 1693 mm. The rate is highest in March and April when the relative humidity is lowest, temperature highest, and dry NE wind is blowing. The lowest rates of evaporation occur in August and September when the temperature is around average but the relative humidity is very high. Evaporation rates range between 119.3 mm (3.86 mm/d) and 186.7 mm (6.02 mm/d) (Figure 5). The measured yearly evaporation rate is higher than the rainfall (1425 mm) in Angkor area. But for getting the real evaporated rate, the evapotranspiration should be calculated. Theory and equations can be found in any hydrological handbook (e.g. Chow, 1963). For open water, the evapotranspiration can be calculated with pan coefficient (the measured evaporation should be multiply this value) varying from 0.6 to 0.8. Dec

Month Prec (mm) Jan 4.5 Feb 3.8 Mar 26.8 53.5 Apr May 134.1 Jun 199.3 Jul 197.3 Aug 212.5 Sep 283.2 Oct 226.7 Nov 79.1 Dec 4.6 Total 1425.5

Evap (mm) 128.8 129.5 180.0 161.9 156.8 136.6 141.5 137.1 115.1 111.1 112.4 114.8 1625.6

Jan mm 300

Feb

200 Nov

Mar 100

Oct

0

Apr

Sep

May Precipitation

(mm/month)

Aug

Jun Jul

Evaporation

(mm/month)

Figure 5 Monthly average precipitation (data from 35 years, 1922-2002) and evaporation (1998-2000) in Siem Reap. Data from Siem Reap Weather Station and MRCS.

5

Average temperature for the region is 28.2 ˚C and it varies from the coolest December to the hottest April while the average temperature ranges between 25.3 and 30.6 ˚C, respectively. Daily maximum and minimum temperatures vary from 30.5 to 35.7 ˚C and from 20.1 to 25.3 ˚C, respectively. Relative humidity varies in Siem Reap region from 63.24 % in Mars to 81.29 % in September. The annual average humidity for years 1998-2002 was 73.16 %. From December to Mars the polar winds prevail in Cambodia from NE - E. In April the eastwards shift of the anticyclone causes the wind to suddenly change the direction the east. During the wet season the wind blows from SW. From beginning of October the polar NE – N winds start to dominate again in the area. The average monthly speed of the wind varies from 1.93 m/s in January and March to 2.58 m/s in August. The yearly average speed in the Angkor region is 2.52 m/s for years 1998-2002 (source: Siem Reap Weather Station). 2.2. Geology, soils, and ecology of the area In the northern mountain area, sandstone and conglomerate strata form the main mass. These were deposited in the Jurassic and Cretaceous periods, with the upper horizons extending up to the early Tertiary Age (Garami & Kertai, 1993). The most widespread geologic unit is situated on the high plain (between 25 and 70 m a.m.s.l.), is the Old Alluvium composed mostly of sand, silt, clay, and laterite with a few layers of gravel. On the shores of the Tonle Sap, and reaching elevation of 20 to 25 m above sea level, the deposits are sandy silt and clay of Young Alluvium (Rasmussen and Bradford, 1977; and Friedli, 2003) (Appendix A: Figure 13). Various soil types occur in the Angkor region. The most common soil type is red-yellow podzols. The other soil types in the area are acid lithosoils, planosols, plinthite podzols, cultural hydromorphics, grey hydromorphics, acid lithosoils, alluvial lithosoils, and lacustine alluvial soils (Appendix A: Figure 14). Presently the ecology of the Angkor area can be divided into three groups. Firstly, the flood plain consisting of occurs rice fields and flood forest. Secondly, the plain which consists of mostly shallow water rice cultivation and bush land and finally, the hills and forest area which starts north of the gentle plain area and extends up to Kulen Mountains waterfalls and thick forest (Appendix A: Figure 15). 3. HYDROLOGY OF THE ANGKOR AREA To complete a comprehensive hydrological analysis of the area, much more data (e.g. flow, water stage, sediment concentration, etc.) is needed. There is presently only hydrological data for the very last years and in some areas, e.g. Puok River no data have been collected. This chapter introduces the watersheds of the area and the changes of the river lines due to the anthropogenic modifications. Ground water characteristics will also be presented. 3.1. Watersheds At the present time the Angkor hydrological area consists of three watersheds: Puok,

6

Siem Reap and Roluos. Before the Angkor period, the region was divided into only to two main watersheds, which were Puok (including most of the present Siem Reap watershed) and Roluos. The total area of the hydrological region is 2885 km2. Angkorean changed the whole natural water way by making offtake channels to the Puok River in 10th and/or 11th century. One of the offtakes was the present Siem Reap River (point F in Figure 11 shows the place of the offtake). The most probable reason for this channel was to take more water down to East Baray. Before the Siem Reap offtake existed, the baray was fed by one of the tributaries of Roluos River. Other of the major offtakes was the Great North Channel which most probably was built to get water to West Baray. The areas of watersheds have been calculated using a combination of digital elevation model (DEM) produced by DMA (1963), and cartographic and field surveys. Due to the very shallow slope of the plain area (order of 1/1000), inaccuracies in the used DEM, and the modifications to the natural water ways made by the Angkoreans the calculated watershed borders may include some inaccuracies especially in the main temple area. More detail study will be performed with the help of new DEM created by NASA/JPL (JPL, 2002; and Evans & Kummu, 2003) in the near future.

Figure 6

Ancient (left) and present watersheds (right) of the Angkor area (DMA, 1963; Certeza, 1964; JICA, 1999b; and Evans, 2002)

Puok At the present time the catchment area of Puok is 961 km2. The Puok River is modified in many places by anthropogenic activity. The watershed used to extend all over the Kulen Mountains, which belongs to Siem Reap catchment now as presented in Figure 6. The ancient Puok watershed covered an area of 1652 km2. Angkoreans changed the whole structure of the watersheds constructing the Siem Reap Channel. It was first of the many offtakes of Puok River but over time it increased in size to the main stream due to erosion. Nowadays in many places the river is very disjointed and unnatural. For the Puok catchment no flow or sedimentation data are available. The watershed models calibrated for the Siem Reap catchment will be applied for the river later on.

7

Siem Reap The catchment area of Siem Reap River at Tonle Sap Lake is 697 km2. For Prasat Keo Bridge, where the water stages and flows have been measured, the area of watershed is 525 km2. The Siem Reap River, originally an artificial channel now captures most of the flow of the former Puok River. It has been hypothesised that the heavy bed erosion of the river made it increasingly difficult to divert water to barays and possibly other storages. This may have dramatically affected the whole water management in Angkor area. The mean annual runoff to Siem Reap River is 435 mm with a standard deviation of 112 mm (Table 1 and Appendix B: Figure 16). Data from years 1999-2001 were used in the calculations. Table 1

Average monthly flow (m3/s), runoff (mm) and percentage distribution of runoff within the year (%) (years 1999-2001) for Siem Reap River. Jan Feb Flow (m3/s) 2.40 1.60 Runoff (mm) 12.14 7.30 Runoff Distribution 1-4% 1-3%

Mar 1.27 6.42

Apr May Jun 1.57 2.40 6.62 7.67 12.15 32.39

Jul 7.84 39.60

Aug 10.38 52.45

Sep 11.48 56.13

Oct 23.88 120.70

Nov Dec 12.94 4.90 63.29 24.76

1-3% 1-2% 1-4% 3-11% 8-11% 11-13% 10-18% 22-34% 9-22% 5-7%

Roluos The catchment area of Roluos River at the Tonle Sap is 801 km2. For Kompong Thkau, where the water stages and flows were measured in the early 1960’s (Carbonnel & Guiscafré, 1963) the watershed area is 424 km2. Roluos River has had approximately the same watershed throughout its history from pre-angkorian times until today. Some modifications for the river line were done by Angkoreans especially around the Phnom Bok, eastern part of the watershed, where some embankments and channels were made to lead the water to East Baray. How the water was controlled and how it moved in the system is still unclear but will be declared during the project. Naturally, the Siem Reap Channel affected also to Roluos Catchment by taking part of its water (Figure 6). The average flow and sedimentation concentrations of the river are presented in Table 2. Table 2

Sedimentation and flow rates in Roluos watershed (Carbonnel & Guiscafré, 1963). Sed (g/s) Sed (t/month) Flow (m3/s) Flow (1000 m3/month)

Oct-62 Nov-62 Dec-62 Jan-63 Jul-63 Aug-63 Sep-63 Total 127.45 49.27 10.52 7.31 - - - - 1.87 304.48 227.20 10.01 341.37 127.70 28.17 19.58 - - - - 5.02 815.53 588.90 1926.26 4.28 1.86 0.31 0.21 - - - - 1.57 4.28 5.69 2.60 11461

4975

827

560 - - - - -

4200

11472

15227

6960

3.2. Ground water Ground water is a very important hydrological feature in the study area. The Angkor area has very good sources of ground water. It is easily accessible as the water table during the wet season and dry season lies between depths of 0 and 5 m below the ground level as presented in Figure 7 (JICA, 2000; JSA, 1996; and JSA 2002). Figure 7 shows the ground water level at Bayon, Angkor Thom. The seasonal variation is clearly visible in this data and it represents the average behaviour of the

8

ground water in the region. The results are relatively similar to the measurements taken from Angkor Wat (JSA, 2002) and Banteay Srei temple (Dr. Friedli1, personal comm.) Jan-98

Jan-99

Jan-00

Jan-02 500

-0.5

450

-1.0

400

-1.5

350

-2.0

300

-2.5

250

-3.0

200

-3.5

150

-4.0

100

-4.5

50

-5.0

0 Ground water stage

Figure 7

Jan-01

Precipitation (mm/month)

Ground water stage (m)

Jan-97 0.0

Precipitation

Ground water stages (m) at Bayon temple area (JSA, 2002) plotted against the precipitation (mm/month) in Siem Reap.

The area has large ground water storage with bedrock situated about 50-60 meters below the ground surface (JICA, 2000). The upper two layers, younger and older alluvium aquifer, have very good ground water potential while the third layer (30-50 m below the ground surface), called Pleistocene formation, seems to have a very poor ground water potential. Thus, the aquifer is some 30 – 50 m deep in the lower plain (JICA, 2000). The average water content of sandy-soils of the upper most recent deposits is about 11 %. The average water content of clayey soils varies from 14.2 % in the Uppermost Recent Deposits Layer to 16.8% (CL) and 25.7 % in the Natural Deposits Layer. (JSA, 1996). Based on the results of the drilling program carried out by Department of Hydrology of the Siem Reap Province the ground water table is shown to vary a lot depending on the area. In Kralanh District, west from the Siem Reap, for example the aquifer was penetrated at depths greater than 40 m. Compared to the east of Siem Reap Province, in Sautre Nikum District where the groundwater lies at a depth of between 11 and 12 m. (Garami & Kertai, 1993). Around the Tonle Sap Lake, the ground water stages vary dramatically (order of tens of meters) even in very short distances depending on the depth of the bedrock and the characteristics of the soil layers. 4. HYDRAULIC ZONES The Greater Angkor area can be divided hydrologically into watersheds and sub-watersheds as described above. However, in the case of Angkor it is essential to

1

Dr. Roland E. Friedli did a comprehensive study of geophysics in the beginning of the year 2003 for the Banteay Srei Conservation project.

9

divide the study area into hydraulic zones based on elevation to understand better the hydraulics and hydrology of the area. There are three main hydraulic zones: upper, middle and lower. Two of the zones can be divided into two sub-zones: the upper comprising of upper plain and mountain areas and the lower comprising of floodplain and upper drainage zone. Each zone has its typical natural hydrological and cultural hydraulic characters which are irrespective of the watershed borders. The characteristics for each zone will be described briefly later on in this chapter. In the upper zone (also known as Collector zone), the water was taken from natural rivers which ran from the Kulen Mountains and then spread to the North-South aligned channels. In the middle zone (also known as Temple zone), the water was collected mainly in large barays, water reservoirs, which were most probably built for multifunctional purposes. The lower zone (also known as Drainage zone) operated as a drainage system for the temple area and dispersed the water down into the Tonle Sap Lake. The total area of the study region is 2885 km2 and it is presently made up of three watersheds (previously described in Chapter 3). The elevation data used for the zoning of the floodplain were taken from Certeza (1963). That data give the most accurate elevation for the lake and its floodplain (MRCS/WUP-FIN, 2002a). For the other areas the elevation model made by DMA (1963), digitalized by MRCS was used.

Figure 8

Hydraulic zones of Angkor region. From north to south: Collector zone (Mountain area and upper plain), Temple zone, and Drainage zone (Upper drainage zone and floodplain). (DMA, 1963; Certeza, 1964; JICA, 1999b; and Evans, 2002).

4.1. Collector zone The collector (upper) zone includes the Kulen Mountains and the upper plain. The elevation varies from 28 to 487 m a.m.s.l. and the total area is 1218 km2, making it largest zone containing some 42.2 % of the total study area. The area is divided into two sub-zones: upper plain (elevations between 28 and 60 m a.m.s.l.) and mountain areas (elevation up from 60 m a.m.s.l.).

10

The most dominant part of the upper zone is the upper plain. The main water source is the Kulen Mountain region (mountain area) from where all three rivers of the study area (look Chapter 3) are sourced. The precipitation rates are almost double in the mountain region compared to the upper plain (Figure 4). From the 10th or 11th centuries the water was taken from the natural rivers running from the mountains and spread out into N-S aligned channels, such as the Great North Channel and Siem Reap Channel (present River) (Figure 11). 4.2. Temple zone The temple (middle) zone includes all the main temple area excluding Roluos Group. It is situated between the contours of 18 m and 28 m a.m.s.l. The Temple zone is 465 km2 which is 16.1 % of the total area. The water was directed from the Collector zone’s N-S channels to the Barays using a wide channel network (Figure 11, page 17) constructed using ground embankments. Barays were the main storages but also large amount of water was collected to the temple moats and trapeangs (ground water and rain fed reservoirs), mainly from ground and rain water sources. The total volume of the barays was approximately 80,000,000 – 120,000,000 m3. The total calculated volume varies a lot between the references (e.g. Lustig, 2001 – 78.4 million m3; Acker, 1998 – 227.2 million m3; and Garami and Kertai, 1993 – 155 million m3). It is also very highly unlikely that all the reservoirs were working at the same time. The purposes of the barays are still unclear. A lot of discussion has been written and many theories have been advanced (e.g. Groslier1974, and 1979; Stott, 1992; van Liere, 1980, and 1982; Pottier, 2000; Bernon, 1997; and Goodman, 2000). Comprehensive studies will be carried out during this project but it is still too early to convict or accept any of the presented theories. 4.3. Drainage zone The drainage (lower) zone includes the area lower than 18 m a.m.s.l. The area can be divided into two areas: Floodplain (1-10 m a.m.s.l.) and Upper drainage zone (10-18 m a.m.s.l.). The total area of the lower hydraulic zone is 1202 km2 and the areas of sub-zones are for floodplain and upper drainage zone 717 and 485 km2, respectively. The floodplain comprises some 24.9 % of the total study area and upper drainage zone 16.8 %. Hence, the drainage zone is 41.7 % of the total area. Some main components of the area are e.g. SE channel and SW channel running from the SW corner of the West Baray to SE and SW, respectively. The ancient Siem Reap Channel (present river) and Angkor Wat Channel are the main structures for transporting the water to the lake from the Temple Area. 5. HYDRAULIC FEATURES IN ANGKOR PERIOD This chapter briefly describes the hydraulic features of Angkorean water management. Additional information about the features can be found in studies by e.g. Groslier (1974 and 1979), Acker (1998), Garami and Kertai (1993), Goodman (2000), Pottier

11

(1999 and 2000), van Liere (1980 and 1982), Fletcher (2001) and Evans (2002). 5.1. Water reservoirs Due to the long dry season, the collection and storage of the water is very important for the people of the area. At the household level this is done by building small ponds adjacent to the house where then rainwater and groundwater can collect. For village use, a bigger pond, trapeang was built. The sizes of trapeangs vary from 2 to 20 ha. Normally trapeangs were related closely also to the temples. In the urban area, in Temple zone, the water was collected in the big barays by channels. Angkor area has four barays: Indratataka (Baray of Loley), East Baray, West Baray, and Jayatataka (North Baray). The vast majority of these features, barays and trapeangs, have a length width ration of 2:1 and are generally aligned E-W (Evans, 2002). Trapeangs were dug into the ground, fed by rain and ground water, and are unrelated to the channel network, while the barays were basically not dug and fed by channels and rainfall (Fletcher, 2001). Almost every temple has its own moat into which water was normally collected from rain and ground water sources. The moats are square or rectangular and are generally oriented E-W and N-S (Garami and Kertai, 1993). The reservoirs, barays, trapeangs and moats probably had multifunctional purposes for Angkorean. The uses could have included ritual uses, aesthetics, water supply for humans and animals, transport, irrigation, drainage, defence, health, fishery, and washing. 5.2. Channels, roads and other structures Channels were crossing over the whole landscape. The channels were generally very shallow and wide, some 1-2 m deep and 30 – 40 m wide. Normally the road(s) were lined the channel on an embankment(s) about 1 - 2 m above natural land surface. The sides of the embankments were probably used for living as well as it is possible to get the house above the floods in rainy season. Some channels had only one embankment on the side of downward slope and those probably have the main purpose as a road. Thus, the linear hydraulic features were normally for multipurpose uses as well. Normally the water was controlled in Angkor area by earth embankments. Not many evidences have been found that Angkorians used laterite constructions or sluice gates for controlling flows. However, there are couple of laterite constructions which probably has served as a water control feature. One is situated at the Ban Penh Reach channel, which has a low-level offtake from the Siem Reap just downstream from the junction of the Siem Reap with the Puok. Other one is Krol Romeas (misleading having a same name as the animal compound next to Preah Kanh temple) which is situated at the east end of East Baray and probably functioned as an inlet or outlet for the baray. 6. TONLE SAP LAKE Tonle Sap Lake is the largest freshwater lake in Southeast Asia and an important

12

component in the Mekong water system. Its size varies from approximately 160 km long and 35 km wide during the dry season up to 250 km long and almost 100 km wide at the height of the flooding (e.g. Keskinen, 2003). The Tonle Sap River connects the lake to the Mekong River and joins it at Chaktomuk junction near Phnom Penh (Figure 1, page 2), after which the river immediately splits into the smaller Bassac River and the larger Mekong River. (Lamberts, 2001). The area is globally unique and the lake has an extraordinary hydrological system. In the wet season, the Tonle Sap River reverses its direction of flow to the Tonle Sap Lake, instead from the lake due to the heavy flooding of the Mekong River. The lake functions as a natural overflow basin for the Mekong system.

Figure 9

Tonle Sap catchment plotted with the coverage of year 2000 flood. (DMA, 1963; JICA, 1999b; and Evans, 2002).

6.1. Water stage The area of the lake varies between the dry and wet season from 3500 km2 up to 14 500 km2 and the depth of the lake increases from 0.5 m up to 6-9 m. The bottom of the lake lies approximately 0.5-0.7 m a.m.s.l. During the year the surface of the lake varies between 1 m and 9 m a.m.s.l, respectively. Figure 9 shows the peak of the flood in year 2000 (possibly the biggest flood during the last 50 years) when the water level reached 9.8 m and 10.3 m a.m.s.l at the Kampong Lung station and Prek Dam station, respectively. During the wet season, the volume of the lake increases from about 1.3 km3 to 50-80 km3 depending on the flood intensity. (The area and volume of the lake during the floods vary in different sources. The reason is that the area calculated as a lake and its floodplain varies between the researches). 6.2. Sedimentation The first theory postulated was that the lake was been infilled with sediment at the rate

13

of 2-4 mm/year (e.g. Carbonnel & Guiscafré, 1963; and Garami & Kertai, 1993); but the resent studies by Mildenhall (1996), Tsukawaki et al. (1997), and Penny (2002) show that the sedimentation rate during the last 5000 years has been approximately 0.1 mm/year. Thus, no significant sedimentation is occurring in the dry season lake area. Sedimentation is heavier on the flooded forest and flood plains in the vicinity of the lake proper and the rivers (MRCS/WUP-FIN, 2003a). Hence, during the Angkorean time the bottom of the dry season lake was only 10 cm below the present bottom level. Thus, no significant changes have occurred in the lake’s morphology during the last thousand years. Therefore, it can be assumed that the present conditions of the lake are similar to present during the Angkorian time.

Figure 10

Estimated sedimentation rates of the northern part of the Tonle Sap Lake. (“This study” means here the results of Penny, 2002). (Penny, 2002).

Sediment flux to Tonle Sap Lake during the wet season is manifold compared to the outflow flux during the dry season (MRCS/WUP-FIN, 2003a). The average annual input to the lake is about 5.7 million tons of sediment (TSS) while the outflow is about 1.2 million TSS. Thus, Tonle Sap Lake is retaining more than 80 % of the annual amount of sediments it receives from Mekong River. The TSS data, secchi depth measurements, and the model calculations indicate that efficient sedimentation takes place in the flood season in the vicinity of Tonle Sap River and the tributaries, in the delta area and in the flooded forest around the lake proper (MRCS/WUP-FIN, 2003a). 6.3. History of the Lake Analysis of the sediment cores extracted from the lake bed suggests a dramatic change in the rate of sedimentation in the lake around 5100 - 5600 B.P. It has been suggested by Tsukawaki (2002) that the lake was not connected with the Mekong River prior to this time, and the lake (or series of lakes) received sediments and water from its own catchment area only and the Tonle Sap River flowed directly to the South China Sea. If this is the case, then the current ecosystem of the lake only started to develop some 5500 years ago.

14

The probable cause for the environmental change described above is a rapid rise of sea-level after the Last Glacial Maximum (LGM: about 18 000 years B.P.). This was followed by a high stand period at the Holocene Climatic Optimum (about 6000 years B.P.) around the Gulf of Thailand (Tsukawaki, 2002; and Nguyen et al., 2000). The risen back water effect, because of the higher sea level, affected to the Mekong floods, shifted to upstream direction, and finally cased the diversion of Mekong flood water to the Tonle Sap area about 5500 years ago. This change started to create the present hydrological and ecological conditions in the Tonle Sap Lake and its floodplains (MRCS/WUP-FIN, 2003a). 6.4. How the Angkor was related to the Tonle Sap Lake The Tonle Sap Lake had a very important role to play in the history of Angkor. Initially it was an important source of nourishment. The Tonle Sap ecosystem is believed to be one of the most productive inland waters and one of the most fish-abundant lakes in the world (Bonheur, 2001). At the present time the lake provides about 60% of the annual commercial fisheries production of Cambodia and its annual fish catch is recently estimated to be approximately 235 000 tons (Van Zalinge et al., 2001) being 40% of Cambodia’s total supply of protein. The lake was also a very important part of the transportation network for Angkor, e.g. goods, food and people were easy to move from point A to B by boats. The location of Angkor is optimum for the city. It is safe from the flood but at the same time very close to the lake even during the dry season (Figure 9, page 13). The floodplain is relatively narrow in the Angkor region (averagely some 12 to 14 km) compared to the other parts of the lake where it can be quadruple (e.g. Battambang region, west side of the lake). Hence, it was possible to come easily and fast by boat up to the city by the present Siem Reap River (Angkor period, channel) probably during the dry season as well. The annual flood creates an optimum condition to cultivate floating rice (called also deepwater rice) during the rainy season and flood recession rice during the dry season when water starts to recede. Floating rice is grown in low-lying areas and depressions that accumulate floodwater at a depth of 0.50 m or more, up to 3-4 m. According to Chou (2001), the floating rice was growth naturally without sowing during the Angkor period. The flood recession rice is grown in the fertile soils on the floodplains. The sowing is conducted as water recedes, starting from the upper fields (Javier, 1997; ref. Lustig, 2001). During the pre and early Angkor period, the floods might have been lower than they are nowadays. The facts which indicate to that are the early temple areas (e.g. Roluos group, Pr. Chedei and Pr. He Phka) which are much lower (between 8 to 12 m a.m.s.l.) than the main Angkor temple area (Evans, 2002). Hence, the higher floods may have forced to move the capital further from the lake. The most probable reason for that would be the different climate conditions on the Mekong, including Tonle Sap watershed area, or different morphological conditions between the Mekong River and Tonle Sap Lake. Clear evidences have not yet been

15

found to support either of the theories. More investigation will be made concerning the issue of flood levels. 7. METHODS FOR STUDYING THE WATER SYSTEM The methods used for the Angkor water system studies are field studies, coring, Ultra-Light plane, aerial photos, AIRSAR-TOPSAR data, two hydrological models and 3D hydrodynamic model. In this chapter, those methods are briefly presented. 7.1. Field methods Field studies are an essential part of the study. They offered a lot of information that is not possible to get from anywhere else. To get a good view of the hydraulic network, individual structures and features, as well as to understand better the local nature, hydrology, and habits of the people the comprehensive field studies are necessary to do. The net sedimentation in the ancient channels can be determined with the help of coring (Player, 2002) but the exact sedimentation rates can be defined only from the features which have been wet since they were constructed using radiocarbon dating (Penny, 2002). Palaeo-environmental data have been collected from features, such as moats and trapeangs, which have been wet since they were constructed. The work has been conducted by Dr. Penny. With the radiocarbon method it is possible to date the sediment layers and thus calculate the net sedimentation for each periods. The Ultralight plane is used to carry out precise checks on problem locations quickly and more safely than is always possible on the ground. The size of the area and areas to where it is difficult or even impossible to go because of the land mines and lack of paths make the use of ultra-light plane very useful. With the plane it is possible to get a view over the landscape from the height of 1 000 m or then a very close look to the details from the height of couple of tens of meters. The potentials of the Ultralight are incredible for the areas like Angkor where the hydrological and hydraulic features are spread over 1000 km2 and the access to the places is difficult or even impossible from the ground. 7.2. AIRSAR-TOPSAR data and aerial photos Aerial photos have been used for identifying the hydraulic features during the field work. Pottier (1999) produced the archaeological maps of southern Angkor (Figure 11) with the help of aerial photos taken by FINNMAP (1992). Airborne synthetic aperture radar (AIRSAR-TOPSAR) data have been used to create a GIS database of the network of canals, reservoirs and embankments used to manage water in Angkorean times in the Northern Angkor (Evans, 2002; and Fletcher et al, 2002). Together with Pottier’s (1999) mapping an exhaustive hydraulic map over the Greater Angkor area have been created (Figure 11). The TOPSAR digital elevation model (DEM) is critical for understanding the

16

dynamics of the fluvial environment. The data have a 5x5m ground resolution and can resolve sub-metre differences in elevation in areas of low topographic relief. The data, radar images and DEM have proven particularly useful in the discovery and analysis of Angkorean hydraulics and hydrology.

Study Area Radar Strips 1-4

Kulen Hills

Foothills (C) Great North Channel (Ends)

Northern Angkor (Mapped in Current Study)

Foothills

Kulen Hills Foothills

Banteay Srei

Great North Channel

(F)

Siem Reap River

Puok River

Route 6 to Sisophon

(B) Baray (Jayatataka)

Banteay Sra

Bayon

Prei Khmeng

West Baray (E) (D)

East Baray

Angkor Thom

Angkor Wat Roluos River

Kuk Svay Thom Area (A)

Chau Srei Vibol

Comparative Test Area

(D)

Baray (Indratataka) Siem Reap Southeast Channel (D) Prasat He Phka

Phnom Krom

High Water Mark (Dashed Line)

(D)

Roluos Prasat Trapeang Phong (D)

Southern Angkor (Mapped by Pottier (1999))

Damdek Canal

Low Water Mark

±

Tonle Sap (Great Lake)

0

2.5

5

Study Area

10 Kilometers

Water

Figure 11. Structure of Angkorean archaeological and hydraulic features showing the locations mentioned in this paper (Evans, 2002 – the current study in the map means the study made by Evans).

7.3. Mathematical models applied to Angkor area The hydrology and hydrodynamic of the Angkor area are investigated with the help of mathematic modelling. Based on a GIS database and TOPSAR elevation model, a three-dimensional (3D) EIA Flow Model (Virtanen et al., 1998) has been set up for simulating water levels and currents, inundation of the flood plains and suspended sediments. Distributed hydrological model, VMOD (Lauri & Virtanen, 2002), applied to the watersheds have been used for the boundary flows of 3D Flow Model (MRCS/WUP-FIN, 2003a). The models have been calibrated for the present hydrological and meteorological conditions of the study region and have been applied for the ancient land use, meteorological, and morphological conditions of the watershed as accurately as possible within the bounds of current knowledge. Using

17

the model, sedimentation and erosion rates can be defined for the area using different historical scenarios for landuse and climate, allowing archaeologists to test hypotheses about the decline of Angkor against rich and complex datasets. Some of the most valuable uses of the model in the study area are (Evans & Kummu, 2003): • • • • • • •

flows and currents in complex channel network reconstruction of the water system influence of the land use changes scenario analysis sediment accumulation and filling of the channels, both in short term and centuries timescale erosion rates in the channels and canals, especially in Siem Reap River and Great North Channel sedimentation rates in the Barays the river floodplain vegetation effects to the currents and sediment concentration

These results will be used for the archaeological, hydraulic and hydrological analysis of the area. With the analysis, it is possible to get more information about the water system, an estimation of how significant that system was in the city’s operation, and to indicate whether or not the demise of the hydraulic network implicated its urban environment in the middle of the second millennium CE.

Figure 12.

Surface (left) and bottom (right) sediment concentrations calculated with 3D Flow Model, West Baray.

8. DISCUSSION In this chapter, the benefits of the Angkor area from water management point of view have been discussed. Also, the possible problems which might have been occurred in the area are briefly presented. 8.1. Advantages of the area from water management point of view The situation of Angkor is optimal from a hydrological point of view. The Tonle Sap Lake was very close to the city but still it was safe from its floods. The lake offered an

18

excellent source of nourishment and potential for transportation. The floodplain was also very fertile for growing rice. Ground water in the area is close to the surface even during the dry season, unlike in many other places around the lake. Thus, the Khmers had easy access to water all year round, guaranteeing water for drinking and bathing even during the driest dry seasons. Also the significant public health benefits of high water tables cannot be ignored (Himel, 2001; ref Goodman, 2001). Regular bathing in the family ponds was noted to have taken place by Chou Ta Kuan (Chou, 2001) at least twice a week. The very gentle slope of the plain created an excellent terrain for managing the water with artificial channels and reservoirs. The Khmers were exceptionally good at managing the water and moving it from one place to another. The channel network was built with care and spread over a 1000 km2 area (Evans, 2002) – some of the channels and reservoirs (e.g. West Baray, thought the water is not flowing to the Baray its original route) are still in use today. 8.2. Disadvantages and problems in the Angkor area from hydrological point of view Although, the closeness of the Great Lake had its advantages it also had disadvantages, for example, it decreased the area of flooded fertile fields. Hence, the potential of the rice cultivation using the flood water was limited in relation to the closeness of the main temple area. The floating and flood recession rice form nowadays only some 4 % of the total rice production in the Siem Reap Province (see Appendix C: Table 3 and Figure 17) which might have been the case also during the Angkor period. Even though the Khmers were excellent water engineers, they could not avoid normal problems relating to erosion and sedimentation. In Barays and some channels, sedimentation was a major problem. Over time, the settled sediments decreased the capacity of the channels and barays and made the supplying water to the city more difficult. The possible climate changes may have affected a lot to the Angkor area. Goodman (2000) proposed that climate played a role in the emergence of Angkor. The possible changes in temperature, precipitation and evaporation may have affected to the local hydrology in many ways. Too much water may caused floods in the city area. On the other hand, it advanced rice growing as a lack of water may have decreased the rice crops, making the water supply more difficult, and lower the ground water level dramatically. Thung (1999) proposed that the tectonic shifts have affected on local hydrology and are responsible for down cutting in the various watercourses on the Angkor plain. However, no clear geophysical evidences have been found to support this theory. Thus, more information about the tectonic shifts in the area is needed for further discussion about this argument. Also, the changes in land use may have affected the hydrology of the area. The more people the more rice is needed. Thus, probably large areas of forest had been turned to rice fields to feed the growing population in Angkor. This affected to the local

19

hydrological cycle in many ways. The surface flow increased with the sediment flux to the rivers; causing several problems on downstream channels and reservoirs. 9. CONCLUSION This paper gives an overview of the present natural environment and hydrology of the Angkor region. The precipitation varies significantly inside the Angkor region being at the lake’s shore line and on high plain only 64 % and 50 % (respectively) of the precipitation in the mountain region where the highest rainfall occurs. The changes of the river lines due to the anthropogenic modifications have been presented with the current and ancient watershed borders. The area is also divided to three hydraulic zones to understand better the hydraulics and hydrology of the area. Each zone has its typical hydrological and hydraulic characteristics: in the upper zone the water was taken from natural rivers and then spread to the North-South aligned channels, in the middle zone the water was collected mainly in large barays, and the lower zone operated as a drainage system. The Tonle Sap Lake is not filling up with sediment against the earlier belief. Angkor was related very closely to the lake. It was an important source of nourishment and one of the key features of the transportation system. The annual flood creates an optimum condition to cultivate floating and recession rice. The Angkor area was perfectly situated from hydrological point of view: the lake was close but the city was safe from the flood, the high ground water table offered a secured water supply also during the dry season, and the shallow slope of the terrain offered good possibilities of manage the water. Even thought the situation was optimal Khmers had problems with sedimentation and erosion in their hydraulic network. It is likely that the climate changes during the periods played a role in the history of Angkor.

20

REFERENCES Acker, R. 1998. New Geographic Tests of the Hydraulic Thesis at Angkor, South East Asia Research, 6, 1, pp. 5-47. Bernon, O. 1997. Note sur l’hydraulique théocratique angkorienne. Bulletin de l’École francaise d’Extrême-Orient. Paris. Tome 84, pp. 340-348. Bonheur, Neou (2001). Tonle Sap Ecosystem and Value, Technical Coordination Unit for Tonle Sap, Ministry of Environment, Phnom Penh, Cambodia. Available online at www.mekoninfo.org. Carbonnel, J-P. and J. Guiscafré. 1963. Grand Lac du Cambodge, Sedimentologie et Hydrologie 1962-1962. Paris. Certeza Surveying, 1964. Final Report. Certeza, Quezon City. Chin, D. A. 2000. Water-Resources Engineering. Prentice Hall, New Jersey. ISBN 0-201-35091-2. Chou, Ta Kuan (Zhou Daguan). 2001. The Customs of Cambodia. Edited and newly translated from the French by Michael Smithies. The Siam Society, Under Royal Patronage. Bangkok. Chow, V. T. 1964. Handbook of Applied Hydrology, A Compendium of Water Resources Technology. McGraw-Hill. DMA, 1963. Digital Elevation Model for Tonle Sap Basin. Based on US Defence Mapping Agency, Digitized by MRCS/TSU. Evans, D. 2002. Pixels, Ponds and People: Urban Form at Angkor from Radar Imaging. Honours Thesis for the Department of Archaeology at the University of Sydney, Australia. 107 pages. Evans, D. and M. Kummu. 2003. Modelling Cultural and Natural Hydrology Using Radar Imaging at Angkor, Cambodia. Will be published in ICGRHWE Conference, Three Gorges Dam, China. FINNMAP. 1992. Aerial photos. Fletcher, R, D.H. Evans, and I. J. Tapley. 2002. "AIRSAR's contribution to understanding the Angkor World Heritage Site, Cambodia - Objectives and preliminary findings from an examination of PACRIM2 datasets." Proceedings of the 11th Australasian Remote Sensing and Photogrammetry Conference, Brisbane, Australia September 2-6 2002. Fletcher, R. 2001. "A.R. Davis Memorial Lecture. Seeing Angkor: New views of an old city." Journal of the Oriental Society of Australia 32-33:1-25. Friedli, R. 2003. Report on Geophysical Survey in Banteay Srei area. Banteay Srei Conservation project, Siem Reap. Garami, F. and I. KERTAI. 1993. Water Management in the Angkor Area. Budapest: Angkor Foundation. Goodman, J. 2000. Reinterpreting Angkor: The Water, Environment and Engineering Context. The Journal of Sophia Asian Studies No. 18 (2000), pp. 131-164. Tokyo.

21

Goodman, J. 2001. The Hydraulic Thesis at Angkor – beyond the geographical tests. A commentary and response to the paper “New Geographical Tests of the hydraulic thesis at Angkor”, by Robert Ackers (published in South East Asia Research, Volume 6, Number 1m March 1998. (unpublished). Groslier, Bernard Philippe 1974. "Agriculture and Religion in the Angkorean Empire." Etudes-rurales 53-56:95-117. Groslier, Bernard Philippe 1979. "La Cite Hydraulique Angkorienne." Bulletin de l'Ecole Francaise d'Extreme Orient 66:161-202. JICA, 1999a. The Study on Groundwater Development in Southern Cambodia. Draft final report (II). Ministry of Rural Development and Kokusai Kogyo Co., Ltd. JICA, 1999b. Cambodia Reconnaissance Survey Digital Data. Ministry of Public Works and Transportation (MPWT), Kingdom of Cambodia. Japan International Cooperation Agency (JICA), Japan, March 1999. JICA, 2000. The Study on Water Supply System for Siem Reap Region in Cambodia. Final Report. The Ministry of Industry, mines and energy – The Royal Government of Cambodia. JPL. 2002. AIRSAR: http://airsar.jpl.nasa.gov/.

Airborne

Synthetic

Aperture

Radar.

URL:

JSA, 1996. Annual Report on the Technical Survey of Angkor Monument 1996. Japanese Government Team For Safeguarding Angkor, UNESCO / Japanese Trust Fund for the Preservation of the World Cultural Heritage. JSA, 2002. Annual Report on the Technical Survey of Angkor Monument 2002. Japanese Government Team For Safeguarding Angkor, UNESCO / Japanese Trust Fund for the Preservation of the World Cultural Heritage. Keskinen, M. 2003. The Great Diversity of Livelihoods – Socio-economic survey of the Tonle Sap Lake. WUP-FIN Socio-economic Studies on Tonle Sap 8, MRCS/WUP-FIN, Phnom Penh. Koponen, J., M. Virtanen, H. Lauri, N. van Zallinge, and J. Sarkkula. 2003. Modelling Tonle Sap Basin for Sustainable Resources Management. (will be published in, The 1st Yellow River Conference, China) Lamberts, D. 2001. Tonle Sap Fisheries: A case study on floodplain gillnet fisheries, Asia-Pacific. Fishery Commission, FAO, Bangkok, Thailand. Lauri, H. and M. Virtanen. 2002. Distributed modelling of lake Tonle Sap catchment. Environmental Software 2002, Eds. C.A.Brebbia et al., WIT Press, Swansea, 7 pages. Lustig, E. 2001. Water and the Transformation of Power at Angkor, 10th to 13th Centuries A.D. Unpublished BA (Hons) Thesis, Department of Prehistoric and Historic Archaeology: University of Sydney. Mildenhall, D.C. 1996. Palynology of Holocene and Last Glaciation Samples from Lake Tonle Sap, Cambodia.

22

MRCS, 2003. Database for precipitation and water stage data. Mekong River Commission Secretariat. MRCS/UNDP, 1998. Natural Resources-Based Development Strategy for the Tonle Sap Area, Cambodia. (CMB/95/003). Final Report, volume 2, part B. Sectorial Studies. Phnom Penh. MRCS/WUP-FIN, 2002a. Data Report. Water Utilization Program – Modelling of the Flow Regime and Water Quality of the Tonle Sap. MRCS/WUP-FIN, Phnom Penh, Cambodia. Draft, August 2002. MRCS/WUP-FIN, 2002b. Data Collection Report. Water Utilization Program – Modelling of the Flow Regime and Water Quality of the Tonle Sap. MRCS/WUP-FIN, Phnom Penh, Cambodia. Draft, August 2002. MRSC/WUP-FIN, 2003a. Modelling Tonle Sap Watershed and Lake Processes for Environmental Change Assessment. Model Report. MRCS/WUP-FIN, Phnom Penh, Cambodia. Draft, January 2003. MRSC/WUP-FIN, 2003b. Tonle Sap Development Scenario Impacts and Guidelines. MRCS/WUP-FIN, Phnom Penh, Cambodia. Draft, April 2003. Nguyen, V. L., T. K. Oanh Ta, and M. Tateishi. 2000. Late Holocene depositional environments and coastal evolution of the Mekong River Delta, Southern Vietnam. Journal of Asian Earth Sciences 18, pp. 427-439. Penny, D. 2002. Sedimentation rates in the Tonle Sap, Cambodia. Report to the Mekong River Commission. Player, S. 2002. Summary Report of Exploratory Coring on the South East Canal of the Angkor Plain, Cambodia. A field study undertaken in January 2002 by the Greater Angkor Project. The University of Sydney. Pottier, C. 1999. Carte Archéologique de la Région d’Angkor. Zone Sud. Ph.D thesis, 3 vols. Universite Paris III - Sorbonne Nouvelle (UFR Orient et Monde Arabe). Pottier, C. 2000. Some evidence of an inter-relationship between hydraulic features and rice field patterns at Angkor during ancient times. The Journal of Sophia Asian Studies, No 18, pp. 99-119. Rasmussen, W. C. and G. M. Bradford. 1977. Ground-Water Resources of Cambodia. Geological Survey Water-Supply Paper 1608-P. United States Government Printing Office, Washington. Stott, P. 1992. Angkor: Shifting the Hydraulic Paradigm, in J. Rigg (ed.). The Gift of Water: 47-58. London: School of Oriental and African Studies. Thung, Heng. 1999. Did the earth move to fall Angkor? SPAFA Journal, 9, pp. 5-14. Tsukawaki et al. 1997. Sedimentation rates in the northern part of lake Tonle Sap, Cambodia, during the last 6 000 years. Summaries of Researchers Using AMS at Nagoya University 8. 125-133.

23

Tsukawaki, S. et al. 2002. Environmental Changes of Lake Tonle Sap and the Lower Course of the Mekong River System in Cambodia During the Last 6,500 year – Results of the Tonlesap 96 Project. Symposium on Environmental Changes of the Lake Tonle Sap, Phnom Penh, Cambodia. van Liere, W. J. 1980. Traditional Water Management in the Lower Mekong Basin. World Archaeology, 11, 3, pp. 265-280. van Liere, W. J. 1982. Was Angkor a Hydraulic Society? Ruam Botkwan Prawarisat (Recueil d’articles d’histoire), Silpakorn University, Bangkok, 4, pp. 36-48. van Zalinge, N., N. Thuok, and S. Nuov. 2001. Status of the Cambodian Inland Capture Fisheries with Special Reference to the Tonle Sap Great Lake, in Cambodia Fisheries Technical Paper Series, Volume III, Inland Fisheries Research and Development Institute of Cambodia (IFReDI), Phnom Penh, Cambodia. Virtanen, M., J. Koponen, and O. Nenonen. 1998. Modelling the systems of three reservoirs, rivers, lakes, coastal area and the sea in northern Finland. International Review of Hydrobiology 83, pp. 705 - 712.

24

APPENDIX A GEOLOGY, SOIL AND ECOLOGY MAPS FOR STUDY AREA

Figure 13

Geological features of the Angkor area

Figure 14

Soil types for the Angkor area

Figure 15

Ecological map (present) from Angkor area

25

APPENDIX B SIEM REAP FLOW

160.00 140.00

Aver Min Max

Runoff (mm)

120.00 100.00 80.00 60.00 40.00 20.00

Figure 16

DEC

NOV

OCT

SEP

AUG

JUL

JUN

MAY

APR

MAR

FEB

Nov-01

Sep-01

Jul-01

Mar-01

May-01

Jan-01

Nov-00

Sep-00

Jul-00

Mar-00

May-00

Jan-00

Nov-99

Sep-99

Jul-99

May-99

Jan-99

100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 Mar-99

Flow (m3/s)

JAN

0.00

Average monthly runoff (mm) and daily flow (m3/s) for Siem Reap River at Prasat Keo.

26

APPENDIX C RICE PRODUCTION IN SIEM REAP PROVINCE Table 3

Average rice productions and cultivated areas for Siem Reap Province, years 1997-2001 (MRCS/WUP-FIN, 2003b: source Ministry of Agriculture, Mr. Rathana). Siem Reap Province (1997-2001) Wet season rice Dry season rice Floating rice Recession rice

Cultivated Harvested Area (ha) Area (ha) 178,395 173,033 10,038 9,804 6,446 4,457 2,749 2,732

Yield Production (T/ha) (T) 1.37 233,520 2.43 23,806 1.19 5,250 1.95 5,316

2000-2001 1999-2000 1998-1999 1997-1998 1996-1997

Recession rice

Floating rice

Dry season rice

Wet season rice

0

50,000

100,000

150,000

200,000

250,000

300,000

Production (T)

Figure 17

Distribution of rice production among different varieties at the area of Siem Reap Province (MRCS/WUP-FIN, 2003b: source Ministry of Agriculture, Mr. Rathana).

27