Use of Seasonal Streamflow Forecasts for Flood Mitigation With

Case Report Use of Seasonal Streamﬂow Forecasts for Flood Mitigation with Adaptive Reservoir Operation: A Case Study of the Chao Phraya River Basin, Thailand, in 2011

Wongnarin Kompor 1,*, Sayaka Yoshikawa 1,2 and Shinjiro Kanae 1

1 Department of Civil and Environmental Engineering, School of Environment and Society, Tokyo Institute of Technology, 2-12-1-M1-6 Ookayama, Meguro-ku, Tokyo 152-8552, Japan; [email protected] (S.Y.); [email protected] (S.K.) 2 Global and Local Environment Co-creation Institute (GLEC), Ibaraki University, 2-2-35 Sakuragawa, Mito, Ibaraki 310-0801, Japan; [email protected] * Correspondence: [email protected]; Tel.: +66-92-940-9891

Received: 2 October 2020; Accepted: 12 November 2020; Published: 16 November 2020

Abstract: Predicting streamflow can help water managers make policy decisions for individual river basins. In 2011, heavy rainfall from May until October resulted in the largest flood event in the history of Thailand. This event created difficulty for water managers, who lacked information to make predictions. Studies on the 2011 Thai flood have proposed alternative reservoir operations for flood mitigation. However, no study to date has used predictive information to determine how to control reservoirs and mitigate such extreme floods. Thus, the objective of this study is to update and develop a method for using streamflow predictive data to support adaptive reservoir operation with the aim of mitigating the 2011 flood. The study area was the Chao Phraya River Basin, one of the most important basins in Thailand. We obtained predictive information from a hydrological model with a reservoir operation module using an ensemble of seasonal precipitation data from the European Centre for Medium–Range Weather Forecasts (ECMWF). The six-month ECMWF prediction period was used to support the operation plan for mitigating flooding in 2011 around each reservoir during the wet season. Decision-making for reservoir operation based on seasonal predictions was conducted on a monthly time scale. The results showed that peak river discharge decreased slightly, by around 4%, when seasonal predictive data were used. Moreover, changing the reservoir operation plan and using seasonal predictions decreased the peak river discharge by around 20%.

Keywords: adaptive reservoir operation; ECMWF predictive data; H08 model; reservoir operation; river management; streamﬂow prediction; Thailand

1. Introduction Natural disasters have occurred throughout human history. In particular, catastrophic floods and droughts have had a significant impact on society. However, humans can reduce the damage caused by disasters by developing advanced technologies such as dams [1–3]. The purpose of dam construction is to control water so that it provides the greatest possible benefit to society [4,5]. Flood and drought mitigation can be achieved through well-organized reservoir operation [6–8]. However, a report by the Intergovernmental Panel on Climate Change [9] noted that the effects of climate change make extreme weather and climate events, such as extreme precipitation events and droughts, more frequent and severe globally. Thus, water managers face significant challenges in terms of controlling water in a way that benefits society.

Water 2020, 12, 3210; doi:10.3390/w12113210 www.mdpi.com/journal/water Water 2020, 12, 3210 2 of 19

In 2011, catastrophic flooding occurred in Thailand. From May until October 2011, the country experienced heavy rainfall as a result of five typhoons. This rainfall was estimated to total 1439 mm or about 143% of the average rainy season rainfall from 1982 to 2002 [10,11]. This flood caused substantial damage to the Thai economy and society. The capital city, Bangkok, and industrial areas were inundated from May until December, which caused the Thai economy to lose about USD 45 billion [12]. Because of a lack of information on future rainfall, there was no plan for this unexpectedly severe rainfall event, which affected the Chao Phraya River Basin (CPRB). Thus, two large reservoirs located behind the Bhumibol and Sirikit dams were unable to mitigate the flooding. Several studies in Thailand have focused on optimizing reservoir operation and analyzing the regulated flow of the 2011 floodwaters in the CPRB [13–16]. Mateo et al. [6], which aimed to mitigate flooding in the CPRB in 2011 through alternative dam operation strategies, proposed four operation schemes for controlling target dam storage to mitigate flooding. In that study, researchers proposed four operation schemes for controlling target dam storage to mitigate flooding. The results showed that peak streamflow could be reduced using the proposed schemes. However, that investigation did not use seasonal predictive information. Predictions at the seasonal scale are useful for water management and can support timely reservoir operation. To make decisions related to reservoir operation, water managers must act on predictive information and water demand at the time [17,18]. In Thailand, most reservoirs are operated using a rule curve. The rule curve, which guides water managers to properly control reservoirs, is created from historical streamflow data in the target basin area [19]. It is analyzed using past streamflow data obtained since the recording of streamflow began. Two rule curves are used: the upper rule curve and the lower rule curve. The upper rule curve is defined as the standard water level in a reservoir each month. It is essential to maintain a water level that does not exceed the upper control level. Water is maintained between the upper water level and the maximum water level to prevent flooding. The lower rule curve indicates the minimum water level in the reservoir each month, setting a standard level that is not below the control level. The aim is to reserve sufficient water between the minimum and the lowest water levels that will meet the needs for cultivation and prevent water shortages during the dry season [20]. In 2011, dam storage at both reservoirs was lower than the rule curve at the beginning of the wet season, and no information was available on whether streamflow would increase or decrease during the subsequent months. In addition, the reservoirs might not have been appropriately managed to mitigate flooding in 2011 because of a lack of predictive information. Moreover, the substantial rainfall in 2011 was an unexpected event, and such events make it more difficult to operate reservoirs to mitigate flooding. Therefore, a method of predicting streamflow is needed for effective water management in Thailand. Aiming to support water management, previous streamflow prediction studies have simulated predictions using hydrological models [21]. The hydrological ensemble prediction system (HEPS) uses a hydrological model forced with a range of probabilities in a seasonal climate forecast environment based on general circulation models (GCMs) [22]. HEPS has been widely used in numerous studies around the world. However, GCM ensemble climate forecasting is generally biased, which can increase the uncertainty of streamflow forecasts. In addition, the spread of the GCM ensemble climate forecast may be too wide for a small region such as the CPRB [23]. These limitations must be addressed before GCM ensemble climate forecasting can be used for water management [24]. Previous studies of seasonal streamflow prediction have proposed various methods to improve the accuracy of seasonal streamflow prediction [25,26]. However, little attention has been paid to using seasonal streamflow forecasts to support water management. Although Mateo et al. [6] proposed a useful operation scheme, they did not consider seasonal streamflow prediction. Moreover, their research mitigated the 2011 flood by controlling storage of the target dams. In reality, water managers cannot set target storage levels at the beginning of the wet season without predictive information and water demand; in addition, climatic factors are uncertain. Therefore, to predict seasonal streamflow while accounting for reservoir operation, we updated and developed a method for using seasonal streamflow predictive data to support adaptive reservoir operation. Thus, the objective Water 2020, 12, 3210 3 of 19

ofWater this study2020, 12,is x FOR to demonstrate PEER REVIEW the use of streamflow predictive information for adaptive reservoir3 of 19 operation with the aim of mitigating the 2011 flood in the CPRB. This paper presents a comparison ofreservoir river discharge operation with with and the without aim ofthe mitigating use of predictive the 2011 flood information in the CPRB. to support This paper adaptive presents reservoir a operation.comparison In of addition, river discharge the results with and of river without discharge the use of with predictive the 2011 information reservoir to operation support adapt andive river dischargereservoir with operation. the alternative In addition, reservoir the results operation of river were discharge compared with inthe this 2011 study. reservoir operation and river discharge with the alternative reservoir operation were compared in this study. 2. Study Area, Data and Methods 2. Study Area, Data and Methods 2.1. Study Area 2.1. Study Area The CPRB, which is located in central Thailand, is one of the largest river basins in the country, with anThe area CPRB, of about which 158,000 is located km2 (30%in central of Thailand’s Thailand, area; is one see of Figurethe largest1). The river CPRB basins has in long the country, supported 2 thewith local an community area of about and 158,000 economy. km (30% The Thai of Thailand’s people have area; made see Figure use of 1). the The CPRB CPRB for has transport, long supported the local community and economy. The Thai people have made use of the CPRB for drainage, recreation, fishing, agriculture (in particular, rice), and as a source of water for centuries. transport, drainage, recreation, fishing, agriculture (in particular, rice), and as a source of water for Moreover, Bangkok is located downstream of the CPRB. The total number of households in the CPRB is centuries. Moreover, Bangkok is located downstream of the CPRB. The total number of households 23.0 million (1996), and it accounts for 66% of Thailand’s gross domestic product (GDP) [27]. Given its in the CPRB is 23.0 million (1996), and it accounts for 66% of Thailand’s gross domestic product (GDP) importance, the impact of natural disasters on the Thai economy and society, in particular downstream [27]. Given its importance, the impact of natural disasters on the Thai economy and society, in ofparticular the CPRB, downstream is significant. of Figurethe CPRB,1 shows is significant an overview. Figure of river 1 shows operations an overview for two of major river reservoirsoperations in thefor CPRB. two major The maximum reservoirs damin the storage CPRB. The capacity maximum of the dam Bhumibol storage reservoir capacity of (13,500 the Bhumibol million centimeters;reservoir MCM)(13,500 is greatermillion thancentimeters; that of the MCM) Sirikit isreservoir greater than (9700 that MCM). of the The Sirikit representative reservoir (9700discharge MCM). station The for therepresentative CPRB is station discharge C2. The station river storagefor the capacity,CPRB is station which isC2. the The lowest river flowstorage rate capacity, that indicates which flooding,is the 3 is 3590lowest m flow/s. rate that indicates flooding, is 3590 m3/s.

FigureFigure 1. Overview1. Overview of flowof flow direction, direction, locations locations of reservoirs, of reservoirs, gauging gauging station, station, rainfall station,rainfall and station, hydrograph and in thehydrogra Chao Phrayaph in the River Chao Basin Phraya at station River C2; Basin 30 year at average station (1981–2010) C2; 30 year rainfall average observations (1981–2010) at the rainfall rainfall stationobservations (black bar), at the observed rainfall dischargestation (black (square bar), box observed with blue discharge line), and (square naturalized box with observed blue line), discharge and (dischargenaturalized without observed dam; discharge square box (discharge with black without dotted dam; line). square box with black dotted line).

Water 2020, 12, 3210 4 of 19

Ping, Wang, Yom, and Nan, which are sub-basins of the CPRB, are located upstream. The Bhumibol and Sirikit reservoirs are located along the Ping and Nan Rivers, respectively. Water flows from the north of Thailand before reaching station C2, which is located at Nakorn Sawan. Downstream of the CPRB is the Gulf of Thailand. The dry season in the CPRB is from November to April, and the wet season is from May to October. To supply water in the dry season and mitigate flooding in the wet season, two large reservoirs were built with a combined storage capacity of 23 billion m3. The Bhumibol and Sirikit reservoirs play an essential role in water management in the CPRB. These two large reservoirs ensure that river discharge increases during the dry season and decreases in the wet season relative to the river discharge pattern without reservoirs (Figure1). The period of peak river discharge in the CPRB is from September to October, because heavy rainfall usually occurs from August to September. Thus, the outflow from reservoirs is usually lower during the wet season from August to October than from May to July.

2.2. Data Seasonal precipitation data generated with the European Centre for Medium-Range Weather Forecasts (ECMWF) System 5 [28] were used as input data. The resolution of these precipitation data is 1◦ in both latitude and longitude. The temporal resolution of seasonal precipitation is daily from 1993 to 2016. ECMWF System 5 predicts precipitation with a forecast range of six months and releases ensemble predictions on the ﬁrst of each month. For each ECMWF System 5 prediction, 25 ensemble members are used. In this study, we selected a starting prediction month of June (the start of the wet season) and conducted seasonal precipitation prediction for July, August, and September. Geographic data and meteorological forcing data, such as daily observed temperature, relative humidity, wind speed, wave radiation, and surface air pressure, were obtained from previous research [29,30]. Table1 summarizes the input data used in this study. Precipitation data were reanalyzed from measurements from rain gauge stations. The spatial resolution of the reanalyzed precipitation data was 5 min. The temporal coverage of the observed input data was from 2010 to 2011.

Table 1. Summary table of data in this study.

Data Type Name Source Resolution Data Length Surface air pressure Wind speed 1/12 Speciﬁc humidity [29] ◦ 2010–2011 Every 6 h Shortwave and Meteorological Longwave radiation forcing data Temperature 1/12 Rainfall Re-analysis 1 ◦ 2010–2011 Daily ECMWF System 5 1 [28] ◦ June–October 2011 rainfall Daily

Geographic data Geographic map [30] 1/12◦ - Royal Irrigation Discharge at station C2 Department Daily 2011 Observed data Outﬂow and at the Bhumibol and Sirikit Electricity Generating reservoir Authority of Thailand Dam storage (EGAT) 1 Data were reanalyzed by Prof. Kenji Tanaka et al. using observed precipitation data from the Royal Irrigation Department (RID) and Thai Meteorological Department (TMD). Water 2020, 12, 3210 5 of 19

2.3. Method

2.3.1. Hydrological Model We used the H08 model [31] to simulate hydrological processes. The H08 model is a distributed global hydrological model that has been adapted and calibrated for application in the CPRB [6]. It consists of six sub-models, namely, land surface hydrology, river routing, reservoir operation, crop growth, environmental flow, and water abstraction. These six models can run separately, or coupled models can run all processes in an integrated manner. Because this study was focused on the use of seasonal predictive data for water management through reservoir operation, the land surface hydrology, river routing, and reservoir operation models were used. The land surface hydrology model in H08 is based on the leaky bucket model [32,33]. It considers a single-layer bucket for all vegetation and soil types, predicting soil moisture to a depth of 15 cm. In the leaky bucket model, soil moisture drains continuously from the bucket, whereas in the original bucket model, runoff is generated only when the bucket is full. The river routing model in the H08 model, known as total runoff integrated pathways (TRIP) [34], is an idealized global river channel network model. In this model, streamflow is calculated by summing the runoff from the land surface model while considering hydrologic routing. The river is depicted as a straight line with no cross-sectional component that flows at a constant velocity toward a downstream cell. Flow directions were determined from digital elevation models manually corrected using atlases [6]. The TRIP model does not consider lakes or swamps, reservoir operation, withdrawal, or losses to evaporation from water surfaces. The reservoir operation model in the H08 model uses operating rules for individual reservoirs. In general, reservoirs can be classified as irrigation or non-irrigation reservoirs. For non-irrigation reservoirs, reservoir operation is planned to minimize variation in sub-annual and interannual streamflow. For irrigation reservoirs, the aim is to release water daily in proportion to the irrigation water requirements of the lower reaches. To simulate reservoir operation in the CPRB, we based the reservoir operation schedule on historical outflow data from the Bhumibol and Sirikit reservoirs. During the dry season, the Bhumibol and Sirikit reservoirs release water, whereas they store water during the wet season to mitigate flooding. The dry season in the CPRB runs from the beginning of January until May, whereas July to December is considered the wet season. Important parameters, such as the bulk transfer coefficient and shape parameter in the land surface model, followed Mateo et al. [6]. We conducted our simulation on a daily time scale, with 5 min longitudinal and latitudinal grids. The domain of the CPRB ranged from 97◦ E to 102◦ E longitude and from 13◦ N to 20◦ N latitude.

2.3.2. Bias Correction Prior to using the seasonal predictive data, we used a popular bias correction method, namely, quantile mapping, to increase the accuracy of rainfall prediction. The outputs from the climate model are usually produced at a coarse grid scale, which may lead to errors in capturing forecast uncertainty and introduce biases. Quantile mapping [35] was used to correct the cumulative distribution function between the observed rainfall data from 1987 to 2010 and the seasonal rainfall forecast data in the CPRB. The bias correction technique was applied to data in a grid-to-grid manner. Figures A4 and A5 (in AppendixC) shows the comparison between observed rainfall data and the seasonal rainfall forecast after bias correction.

2.3.3. Adaptation of Reservoir Operation based on Predictive Data Four parameters are essential to simulating reservoir operation in the reservoir operation sub-model of the H08 model: the rate of release during the dry season, the rate of release during the wet season, target storage, and the target storage date. In a previous study [6], target storage and the target storage date were altered to create an alternative reservoir operation schedule. However, when using Water 2020, 12, x FOR PEER REVIEW 6 of 19

Moreover, reservoir operation, particularly during the wet season, is crucial for decision-making to mitigate flooding and storing the appropriate amount of water to support society. Thus, the difference between this study and previous research is that we changed the rates of release from the Bhumibol and Sirikit reservoirs for each month during the wet season based on the monthly predictive data. Reservoir operation in this study occurred from June to October, because June and October are the first and last months of the wet season, respectively. Two alternative reservoir operations were adaptively applied, as shown in Figure 2. The first alternative operation was simply to follow the rule curve of each reservoir. We refer to this strategy as “PlanU” operation, because reservoir storage is limited by the upper rule curve. The second alternative operation was to follow a new rule curve proposed after the 2011 flood until the end of October. In addition, we set the target storage for each reservoir, which resembled the reservoir operation proposed in a previous study [6]. The second alternative operation is designated “PlanM” operation here. The main difference between the two plans is the amount of reservoir storage. After the flooding of 2011, the rule curves of the reservoirs were changed (this is usually done once per decade) from the dashed line with squares to the dashed line with triangles shown in Figure 2. If the water manager controls the water level using the rule curve, the maximum reservoir storage available during the wet season to support the dry season is about 15,350 MCM: about 8750 MCM in the Bhumibol reservoir and about 6600 MCM in the Sirikit reservoir. In addition, the water demand in the CPRB [36] is about 11,377 MCM per year. Thus, the target storage level to meet water demand can be attained if the Bhumibol and Sirikit reservoirs meet about 78% and 87% of their total storage capacities, respectively. Therefore, the purpose of PlanM is to prevent overflow at both reservoirs by setting the target storage level, thereby mitigating flooding. Figure 3 shows a sample study operation plan based on seasonal predictions, which is applied

Waterat the2020 monthly, 12, 3210 scale. First, we used predictive data to simulate reservoir storage and river discharge6 of 19 as shown in the prediction box of Figure 3. Then, we determined the reservoir operation as follows: (1) No change in reservoir operation if most reservoir storage predictive data (more than 50% of predictionsthe total ensemble for adaptive predictions) reservoir fall between operation the each upper month, rule curve one can and only lower control rule curve. the rate of release. Moreover,(2) Increase reservoir the operation,rate of outflow particularly until most during reservoir the wetstorage season, is below is crucial the upper for decision-making rule curve if most to mitigatepredicted flooding reservoir and storage storing is theabove appropriate the upper amount rule curve. of water to support society. Thus, the difference between(3) Decrease this study the and rate previous of outflow research until most is that reservoir we changed storage the is rates above of the release lower from rule the curve Bhumibol if most andpredicted Sirikit reservoir reservoirs storage for each is monthbelow the during lower the rule wet curve. season based on the monthly predictive data. Reservoir(4) Stop increasing operation the in thisrate study of outflow occurred if most from river June discharge to October, predictions because June are andabove October the river are thestorage first capacity and last at months station of C2 the in wet that season, month. respectively. Two alternative reservoir operations were adaptivelyAs shown applied, in Figure as shown 3, most in Figure predicted2. The firstreservoir alternative storage operation in Bhumibol was simply reservoir to follow (left figure the rule in curveprediction of each box reservoir. of Figure We3) fell refer above to this the strategy upper rule as “PlanU” curve. Thus, operation, we changed because the reservoir operation storage plan by is limitedincreasing by thethe upperrate of rule outflow curve. from The the second Bhumibol alternative reservoir. operation Meanwhile, was to most follow predicted a new rule reservoir curve proposedstorage in afterSirikit the (right 2011 figure flood untilin prediction the end ofbox October. of Figure In 3) addition, fell below we the set lower the target rule storagecurve. Thus, for each we reservoir,changed the which operation resembled plan by the reducing reservoir the operation rate of outflow proposed from in the a previousSirikit reservoir. study [The6]. Theresults second after alternativewe changed operation the reservoir is designated operation “PlanM” are represented operation by here. the The red mainline in di fftheerence operation between box the of two Figure plans 3. isThen the amountwe used ofthe reservoir new rates storage. of outflow from both reservoirs to simulate the next prediction month.

Figure 2. Reservoir operation in this study; PlanU (red line), and PlanM (yellow line). After the flooding of 2011, the rule curves of the reservoirs were changed (this is usually done once per decade) from the dashed line with squares to the dashed line with triangles shown in Figure2. If the water manager controls the water level using the rule curve, the maximum reservoir storage available during the wet season to support the dry season is about 15,350 MCM: about 8750 MCM in the Bhumibol reservoir and about 6600 MCM in the Sirikit reservoir. In addition, the water demand in the CPRB [36] is about 11,377 MCM per year. Thus, the target storage level to meet water demand can be attained if the Bhumibol and Sirikit reservoirs meet about 78% and 87% of their total storage capacities, respectively. Therefore, the purpose of PlanM is to prevent overflow at both reservoirs by setting the target storage level, thereby mitigating flooding. Figure3 shows a sample study operation plan based on seasonal predictions, which is applied at the monthly scale. First, we used predictive data to simulate reservoir storage and river discharge as shown in the prediction box of Figure3. Then, we determined the reservoir operation as follows: Water 2020, 12, x FOR PEER REVIEW 7 of 19 Water 2020, 12, 3210 7 of 19 Figure 2. Reservoir operation in this study; PlanU (red line), and PlanM (yellow line).

Figure 3. Methodology of this study. Figure 3. Methodology of this study.

3. Results(1) No change in reservoir operation if most reservoir storage predictive data (more than 50% of the total ensemble predictions) fall between the upper rule curve and lower rule curve. 3.1. Reliability(2) Increase of the the P rateredictive of outflow Data until most reservoir storage is below the upper rule curve if most predictedSimulations reservoir from storage 25 is ensembles above the for upper river rule discharge curve. using the original ECMWF System 5 predictive(3) Decrease data and the bias rate- ofcorrected outflow predictive until most data reservoir are shown storage in is Figure above the4, with lower a rulezero- curvemonth if lead most- predictedtime from reservoirJune to October. storage We is below selected the results lower rulewith curve. a zero-month lead-time because we operated the reservoir(4) Stop at monthly increasing intervals. the rate of The outflow initial if conditions most river for discharge simulating predictions the zero are-month above lead the river time storagepredated capacity the prediction at station mo C2nth. in that For month. example, we used the initial prediction in May for the June prediction.As shown A comparison in Figure3 of, mostdaily predicteddischarge between reservoir the storage observed in Bhumibol data and predictive reservoir (leftdata figureusing the in predictionoriginal ensemble box of Figure is shown3) fell in above Figure the 4a. upper The ensemblerule curve. simulation Thus, we of changed river discharge the operation using plan the byoriginal increasing ECMWF the rate System of outflow 5 predictions from the (Figure Bhumibol 4a) reservoir. is generally Meanwhile, higher than most the predicted observed reservoir river storagedischarges. in Sirikit During (right the figure peak in period prediction in October, box of Figure simulations3) fell below using the the lower original rule curve. ECMWF Thus, weprediction changed thewere operation higher p lan than by the reducing observed the rate data, of outflow including from the the Thai Sirikit operati reservoir.on simulation. The results A after comparison we changed of thepredictive reservoir data operation using the are bias represented-corrected by ensemble the red lineis shown in the in operation Figure 4b. box The of Figureensemble3. Then simulation we used of theriver new discharge rates of outflowusing bias from-corrected both reservoirs predictive to simulate data showed the next improved prediction accuracy month. compared with ensemble simulation of river discharge using the original ECMWF prediction. 3. Results The ensemble simulation results for river discharge using the bias correction technique were 3.1.lower Reliability than those of the for Predictive Thai operation Data (red line in Figure 4) but still higher than the observed data. The reason for these differences is that 2011 was an abnormal year, and Thailand had not previously experiencedSimulations an extreme from 25 ensembles event like for the river 2011 discharge flooding. using Thus, the when original we ECMWF apply a System bias correction 5 predictive to datahistorical and bias-corrected data, the predicted predictive results data are are underestimates. shown in Figure Overall,4, with a the zero-month accuracy lead-time of the ensemble from June river to October.discharge We prediction selected resultswas improved with a zero-month with the quantile lead-time mapping because webias operated correction the technique. reservoir at We monthly used intervals.the root mean The initialsquare conditions error (RMSE), for simulating relative mean the zero-month error (RME), lead coefficient time predated of variation the prediction of the month.RMSE For(CV), example, Nash–Sutcliffe we used efficiency the initial (NSE), prediction and coefficient in May forof determination the June prediction. (R2) to dete A comparisonrmine the accuracy. of daily dischargeThe results between for each the accuracy observed index data are and shown predictive in Figure data using 5. The the zero original-month ensemble lead-time is prediction shown in Figureimproved4a. Theafter ensemble bias correction; simulation in particular, of river discharge the NSE using showed the originala significant ECMWF improvement. System 5 predictions However,

Water 2020, 12, 3210 8 of 19

(Figure4a) is generally higher than the observed river discharges. During the peak period in October, simulations using the original ECMWF prediction were higher than the observed data, including the Thai operation simulation. A comparison of predictive data using the bias-corrected ensemble is Water 2020, 12, x FOR PEER REVIEW 8 of 19 shown in Figure4b. The ensemble simulation of river discharge using bias-corrected predictive data showedsome months, improved such accuracy as August, compared had lower with accu ensembleracy after simulation bias correction, of river dischargebecause the using simulation the original used ECMWForiginal predictive prediction. data similar to the observed data, in particular in 2010.

(a) (b)

Figure 4. Comparison of daily discharge at station C2 between observations (black line), simulation Figure 4. Comparison of daily discharge at station C2 between observations (black line), simulation results (red line), simulation results using ensemble predicted rainfall data (light blue dotted line), results (red line), simulation results using ensemble predicted rainfall data (light blue dotted line), and the median of simulation data using ensemble predicted rainfall data (blue line). Simulation and the median of simulation data using ensemble predicted rainfall data (blue line). Simulation results using (a) the original European Centre for Medium-Range Weather Forecasts (ECMWF) System results using (a) the original European Centre for Medium-Range Weather Forecasts (ECMWF) 5 ensemble predicted data; and (b) bias-corrected ensemble predictions. System 5 ensemble predicted data; and (b) bias-corrected ensemble predictions. The ensemble simulation results for river discharge using the bias correction technique were lower than those for Thai operation (red line in Figure4) but still higher than the observed data. The reason for these differences is that 2011 was an abnormal year, and Thailand had not previously experienced an extreme event like the 2011 flooding. Thus, when we apply a bias correction to historical data, the predicted results are underestimates. Overall, the accuracy of the ensemble river discharge prediction was improved with the quantile mapping bias correction technique. We used the root mean square error (RMSE), relative mean error (RME), coefficient of variation of the RMSE (CV), Nash–Sutcliffe efficiency (NSE), and coefficient of determination (R2) to determine the accuracy. The results for each accuracy index are shown in Figure5. The zero-month lead-time prediction improved after bias correction; in particular, the NSE showed a significant improvement. However, some months, such as August, had lower accuracy after bias correction, because the simulation used original predictive data similar to the observed data, in particular in 2010. (a) (b) (c) The simulation results using the bias-corrected predictive data were underestimates, in particular for extreme events such as the 2011 flood. The underestimation of simulated discharge after bias correction has been reported in previous studies [37]. This problem is caused by the bias correction itself, which uses previous observations to correct the predictive data. Thus, bias correction does not always improve the accuracy of predictive data, particularly for extreme events.

(d) (e) (f)

Figure 5. Scatter plot of accuracy indices between simulated daily discharge using the original predicted data (x-axis) and simulated daily discharge using quantile mapping (y-axis). (a) Root mean square error (RMSE); (b) coefficient of variation of the RMSE (CV); (c) relative mean error (RME); (d) Nash–Sutcliffe model efficiency coefficient (NSE); (e) coefficient of determination (R2); (f) correlation coefficient (CC).

Water 2020, 12, x FOR PEER REVIEW 8 of 19

Figure 4. Comparison of daily discharge at station C2 between observations (black line), simulation results (red line), simulation results using ensemble predicted rainfall data (light blue dotted line), and the median of simulation data using ensemble predicted rainfall data (blue line). Simulation results using (a) the original European Centre for Medium-Range Weather Forecasts (ECMWF) Water System2020, 12, 5 3210 ensemble predicted data; and (b) bias-corrected ensemble predictions. 9 of 19

(a) (b) (c)

(d) (e) (f)

Figure 5. 5. ScatterScatter plot plot of ofaccuracy accuracy indices indices between between simulated simulated daily discharge daily discharge using the using original the predicted original data (x-axis) and simulated daily discharge using quantile mapping (y-axis). (a) Root mean square error predicted data (x-axis) and simulated daily discharge using quantile mapping (y-axis). (a) Root mean (RMSE); (b) coefficient of variation of the RMSE (CV); (c) relative mean error (RME); (d) Nash–Sutcliffe square error (RMSE); (b) coefficient of variation of the RMSE (CV); (c) relative mean error (RME); (d) model efficiency coefficient (NSE); (e) coefficient of determination (R2); (f) correlation coefficient (CC). Nash–Sutcliffe model efficiency coefficient (NSE); (e) coefficient of determination (R2); (f) correlation 3.2. Adaptivecoefficient Operation (CC). Using Predictive Data for 2011

TheReservoir simulation storage results predictions using from the June bias are-corrected shown in predictive Figure6. No data change were in reservoirunderestimates, operation in particularoccurred infor June, extreme because events most such reservoir as the storage2011 flood. predictions The underestimation (more than 50% of ofsimulated the total discharge ensemble afterpredictions) bias correction fell between has thebeen rule reported curves in of bothprevious the Bhumibol studies [37]. and This Sirikit problem reservoirs. is caused Although by the the bias first correctionprediction itself, month which for the uses Sirikit previous reservoir observations fell below to the correct rule curve,the predictive most reservoir data. Thus, storage bias predictions correction doeswere not within always that improve range at the the accuracy end of December. of predictive Prediction data, particularly of river discharge for extreme for July events. at station C2 is shown in Figure7a. Figure7b,c shows predictions of reservoir storage for July at the Bhumibol and 3.2.Sirikit Adaptive reservoir. Operation The tendency using Predictive for the D riskata offor reservoir 2011 storage to be above the maximum capacity is apparent. However, most ensemble predictions remained under the upper rule curve in both reservoirs. Reservoir storage predictions from June are shown in Figure 6. No change in reservoir operation Therefore, no changes in the rate of outflow from either reservoir were made in July. Thus, no difference occurred in June, because most reservoir storage predictions (more than 50% of the total ensemble was observed between adaptive operation using predictive data and adaptive operation without predictions) fell between the rule curves of both the Bhumibol and Sirikit reservoirs. Although the predictive data, as shown in Figure7. first prediction month for the Sirikit reservoir fell below the rule curve, most reservoir storage predictions were within that range at the end of December. Prediction of river discharge for July at station C2 is shown in Figure 7a. Figure 7b,c shows predictions of reservoir storage for July at the Bhumibol and Sirikit reservoir. The tendency for the risk of reservoir storage to be above the maximum capacity is apparent. However, most ensemble predictions remained under the upper rule curve in both reservoirs. Therefore, no changes in the rate of outflow from either reservoir were made

Water 2020, 12, x FOR PEER REVIEW 9 of 19

inWater July. 2020 Thus,, 12, x FOR no differencePEER REVIEW was observed between adaptive operation using predictive data 9 and of 19 adaptive operation without predictive data, as shown in Figure 7. in July.Predictive Thus, nodata difference without adaptive was observed operation, between shown adaptive in Figure operation 8, indicated using that predictive most ensemble data and predictionsadaptive operation were higher without than predictive the upper data, rule as curveshown and in Figure reached 7. the maximum capacity of both reservoirsPredictive (Figure data 8b, c).without Based onadaptive this result, operation, we increased shown thein Figurerate of outflow8, indicated from that the twomost reservoirs ensemble untilpredictions most ensemble were higher predictions than the for upper reservoir rule curve storage and (more reached than the 50% maximum of the capacity total ensemble of both predictions)reservoirs (Figure fell between 8b,c). Basedthe rule on curves. this result, The weresults increased after we the increased rate of outflow the rate from of outflow the two from reservoirs both reservoirsuntil most are ensemble shown as a predictions red dashed forline. reservoir As a result storage of the increased (more than rate 50% of outflow of the in total August, ensemble river dischargepredictions) increased fell between after adaptivethe rule curves. operation The using results predictive after we increased data, as indicat the rateed of by outflow the pink from line both in Figurereservoirs 8b,c, are which shown shows as a thatred dashed most ensemble line. As a predictions result of the after increased adaptive rate operation of outflow fell in August,between river the ruledischarge curves. increased after adaptive operation using predictive data, as indicated by the pink line in FigureThe 8b ensemble,c, which predictive shows that data most for ensemble September, predictions in which afterwe used adaptive the increased operation rate fell of between outflow thein Augustrule curves. for simulation, are shown in Figure 9a–c. No adaptive operation was conducted in the BhumibolThe ensembleand Sirikit predictive reservoirs, data because for September, most ensemble in which predictions we used for the reservoir increased storage rate ofwere outflow within in theAugust range forof the simulation, rule curves. are The shown results in of Figure ensemble 9a–c. prediction No adaptive without operation adaptive was operation conducted and inwith the adaptiveBhumibol operation and Sirikit using reservoirs, predictive because data most are ensemble presented predictions in Figure for 9a reservoir–c. Because storage no were change within in operationthe range occurred of the rule in curves. September, The resultsthere was of ensemble no difference prediction in September. without adaptive operation and with Wateradaptive2020, 12 operation, 3210 using predictive data are presented in Figure 9a–c. Because no change10 of 19 in operation occurred in September, there was no difference in September.

(a) (b) (c)

Figure 6. River(a) discharge at station C2 (a) and reservoir(b) storage predictions (green line)(c) from June Figureuntil December 6. River dischargeat the Bhumibol at station (b) C2and (a )Sirikit and reservoir(c) reservoir storage. Blue predictions line represents (green simulation line) from using June Figure 6. River discharge at station C2 (a) and reservoir storage predictions (green line) from June untilobserved December meteorological at the Bhumibol data (hindcast). (b) and SirikitBlack dash (c) reservoir.ed line represents Blue line rule represents curve. simulationDash-point usingblack until December at the Bhumibol (b) and Sirikit (c) reservoir. Blue line represents simulation using observedline represents meteorological maximumdata storage (hindcast). capacity. Black dashed line represents rule curve. Dash-point black observed meteorological data (hindcast). Black dashed line represents rule curve. Dash-point black line represents maximum storage capacity. line represents maximum storage capacity.

(a) (b) (c)

Figure 7. River discharge predictions at station C2 (a) and storage predictions from July until December Figure 7. River(a) discharge predictions at station(b ) C2 (a) and storage predictions from(c) July until at the Bhumibol (b) and Sirikit (c) reservoir, shown with the green line, and river discharge and storage December at the Bhumibol (b) and Sirikit (c) reservoir, shown with the green line, and river discharge prediction after adaptive operation with predictive data from July until December (red dashed line). andFigure storage 7. River prediction discharge after adaptive predictions operation at station with C2 predictive (a) and data storage from predictions July until December from July (reduntil December at the Bhumibol (b) and Sirikit (c) reservoir, shown with the green line, and river discharge Predictivedashed line). data without adaptive operation, shown in Figure8, indicated that most ensemble and storage prediction after adaptive operation with predictive data from July until December (red predictions were higher than the upper rule curve and reached the maximum capacity of both reservoirs dashed line). (Figure8b,c). Based on this result, we increased the rate of outflow from the two reservoirs until most ensemble predictions for reservoir storage (more than 50% of the total ensemble predictions) fell between the rule curves. The results after we increased the rate of outflow from both reservoirs are shown as a red dashed line. As a result of the increased rate of outflow in August, river discharge

Waterincreased 2020, 12 after, x FOR adaptive PEER REVIEW operation using predictive data, as indicated by the pink line in Figure10 8ofb,c, 19 which shows that most ensemble predictions after adaptive operation fell between the rule curves.

(a) (b) (c)

FigureFigure 8. 8. RiverRiver discharge discharge predictions predictions at at C2 C2 station station (a (a) ) and and stora storagege predictions predictions from from August August until until DecemberDecember at at the the Bhumibol Bhumibol (b (b) )and and Sirikit Sirikit (c ()c )reservoir reservoir,, shown shown with with the the green green line, line, and and river river discharge discharge andand storage storage predictions predictions after after adaptive adaptive operation operation with with predictive predictive data data from from August August until until December December (red dashed line). (red dashed line). The ensemble predictive data for September, in which we used the increased rate of outﬂow in August for simulation, are shown in Figure9a–c. No adaptive operation was conducted in the

(a) (b) (c)

Figure 9. River discharge predictions at C2 station (a) and storage predictions from September until December at the Bhumibol (b) and Sirikit (c) reservoir, shown with the green line, and river discharge and storage predictions after adaptive operation with predictive data from September until December (red dashed line).

The discharge predictions for September tended to decrease from the beginning to the middle of the month. The reason for this could related to the discharge from the upstream area, which was controlled by the two reservoirs. In addition, the amount of rainfall began to decrease in September relative to August. For this reason, the discharge predictions increased rapidly after the middle of the month because dam storage reached its maximum capacity in both reservoirs; that is, the sudden increase in predicted discharge in September was due to concurrent changes in the rates of release from both reservoirs. Because peak discharge is in October, the rate of outflow cannot change in October. Thus, September was the last month in which reservoir operations could be conducted.

3.3. Adaptive Operation Using Predictive Data Based on PlanM Operation The operating conditions for this simulation were the same as those for the previous simulation, which are explained in Section 2.3.3. The upper rule curve of PlanM is represented with an orange line as shown in Figure 2. Thus, most ensemble predictions for reservoir storage had to fall under the orange line and above the lower rule curve. The prediction period began in June. The results of adaptive operation with predictive data beginning in July are shown in Figure 10. As shown in Figure 10b,c, most ensemble predictions for reservoir storage were higher than the target storage (orange dashed line). Therefore, we increased the rate of outflow from both reservoirs in July until the ensemble prediction for reservoir storage was under the target storage level (orange dashed line) and above the lower rule curve shown in Figure 10a–c. The difference relative to before the adaptive operation is shown with a pink line. We used the rate of release after adaptive operation using predictive data to obtain predictions for August, as shown in Figure 11a–c. Most ensemble predictions of reservoir storage were higher

Water 2020, 12, x FOR PEER REVIEW 10 of 19

Water 2020, 12, 3210 11 of 19 (a) (b) (c)

BhumibolFigure and 8.Sirikit River reservoirs, discharge predictions because at most C2 station ensemble (a) and predictions storage predictions for reservoir from August storage until were within the rangeDecember of the rule at the curves. Bhumibol The (b) resultsand Sirikit of (c ensemble) reservoir, predictionshown with the without green line, adaptive and river operation discharge and with adaptiveand operation storage predictions using predictive after adaptive data areoperation presented with predictive in Figure data9a–c. from Because August nountil change December in operation (red dashed line). occurred in September, there was no diﬀerence in September.

(a) (b) (c)

FigureFigure 9. River 9. River discharge discharge predictions predictions at C2 station station (a ()a )and and storage storage predictions predictions from from September September until until DecemberDecember at theat the Bhumibol Bhumibol (b (b)) and and SirikitSirikit ( (cc)) reservoir reservoir,, shown shown with with the thegreen green line, line, and river and riverdischarge discharge andand storage storage predictions predictions after after adaptive adaptive operation with with predictive predictive data data from from September September until untilDecember December (red(red dashed dash line).ed line).

TheThe discharge discharge predictions predictions for for SeptemberSeptember tended tended to to decrease decrease from from the thebeginning beginning to the to middle the middle of theof month.the month. The The reason reason for for this this couldcould related to to the the discharge discharge from from the theupstream upstream area, area,which which was was controlledcontrolled by by the the two two reservoirs. reservoirs. InIn addition,addition, the the amount amount of ofrainfall rainfall began began to decrease to decrease in September in September relative to August. For this reason, the discharge predictions increased rapidly after the middle of the relative to August. For this reason, the discharge predictions increased rapidly after the middle of month because dam storage reached its maximum capacity in both reservoirs; that is, the sudden the month because dam storage reached its maximum capacity in both reservoirs; that is, the sudden increase in predicted discharge in September was due to concurrent changes in the rates of release increasefrom in both predicted reservoirs. discharge Because in peak September discharge was is duein October, to concurrent the rate changes of outflow in the cannot rates change of release in from bothOctober. reservoirs. Thus, Because September peak was discharge the last month is in in October, which reservoir the rate operations of outflow could cannot be conducted. change in October. Thus, September was the last month in which reservoir operations could be conducted. 3.3. Adaptive Operation Using Predictive Data Based on PlanM Operation 3.3. Adaptive Operation Using Predictive Data Based on PlanM Operation The operating conditions for this simulation were the same as those for the previous simulation, whichThe operating are explained conditions in Section for 2.3.3. this The simulation upper rule were curve the of same PlanM as thoseis represented for the previouswith an orange simulation, whichline are as shown explained in Figure in Section 2. Thus, 2.3.3 most. Theensemble upper predictions rule curve for ofreservoir PlanM storage is represented had to fall with under an the orange lineorange as shown line inand Figure above2 .the Thus, lower most rule ensemblecurve. The predictions prediction period for reservoir began in storage June. had to fall under the orange lineThe andresults above of adaptive the lower operation rule curve. with predictive The prediction data beginning period beganin July inare June. shown in Figure 10. As shown in Figure 10b,c, most ensemble predictions for reservoir storage were higher than the target The results of adaptive operation with predictive data beginning in July are shown in Figure 10. storage (orange dashed line). Therefore, we increased the rate of outflow from both reservoirs in July As shownuntil the in ensemble Figure 10 predictionb,c, most for ensemble reservoir predictions storage was for under reservoir the target storage storage were level higher (orange than dashed the target storageline) (orange and above dashed the lower line). rule Therefore, curve shown we increased in Figure the 10a rate–c. The of outflowdifference from relative both to reservoirs before the in July untiladaptive the ensemble operation prediction is shown forwith reservoir a pink line. storage was under the target storage level (orange dashed line) andWe above used the rate lower of release rule curve after adaptive shown inoperation Figure using10a–c. predictive The diff dataerence to obtain relative predictions to before the adaptivefor August, operation as shown is shown in Figure with 11a a pink–c. Most line. ensemble predictions of reservoir storage were higher We used the rate of release after adaptive operation using predictive data to obtain predictions for August, as shown in Figure 11a–c. Most ensemble predictions of reservoir storage were higher than the target storage level (orange dashed line) in both reservoirs. Thus, we increased the rate of outflow in August until most ensemble predictions for reservoir storage (more than 50% of the total ensemble predictions) were under the target storage level (orange dashed line) and above the lower rule curve. The results after adaptive operation using predictive data from PlanM for August are represented by the orange dashed lines in Figure 11a–c. We increased the rate of outflow from the two reservoirs until the ensemble prediction for reservoir storage was near the lower rule curve and near the river storage capacity at station C2. However, most ensemble predictions remained higher than the target storage of the Bhumibol and Sirikit reservoirs (orange dashed line). Nonetheless, most ensemble predictive data fell between the rule curves for both reservoirs. Water 2020, 12, x FOR PEER REVIEW 11 of 19 Water 2020, 12, x FOR PEER REVIEW 11 of 19 than the target storage level (orange dashed line) in both reservoirs. Thus, we increased the rate of outflowthan the intarget August storage until level most (orange ensemble dashed predictions line) in for both reservoir reservoirs. storage Thus, (more we thanincreased 50% ofthe the rate total of ensembleoutflow in predictions) August until were most under ensemble the target predictions storage for level reservoir (orange storage dashed (more line) andthan above 50% of the the lower total ruleensemble curve. predictions) The results were after under adaptive the targetoperation storage using level predictive (orange dasheddata from line) PlanM and above for August the lower are representedrule curve. Theby theresults orange after dashed adaptive line operations in Figure using 11a– c.predictive We increased data thefrom rate PlanM of outflow for August from theare tworepresented reservoirs by until the orangethe ensemble dashed prediction lines in Figure for reservoir 11a–c. Westorage increased was near the therate lower of outflow rule curve from and the neartwo reservoirsthe river storage until the capacity ensemble at stationprediction C2. forHowever, reservoir most storage ensemble was near predictions the lower remained rule curve hig andher thannear the targetriver storage storage capacityof the Bhumibol at station and C2. Sirikit However, reservoirs most (orange ensemble dashed predictions line). Nonetheless remained hig, mosther ensemblethan the target predictive storage data of thefell Bhumibolbetween the and rule Sirikit curves reservoirs for both (orange reservoirs. dashed line). Nonetheless, most ensembleThe predictionpredictive data results fell forbetween September the rule are curves shown for inboth Figure reservoirs. 12. No adaptive operation was conductedThe prediction based on predictive results for data, September although are most shown ensemble in Figure predictions 12. No for adaptive reservoir o perationstorage were was higherconducted than based the target on predictive storage leveldata, (orangealthough dashed most ensemble line). River predictions discharge for in reservoir September storage tended were to decreasehigher than from the the target beginning storage until level the (orange middle ofdashed the month. line). TheRiver reason discharge for this in could September be related tended to the to dischargedecrease from from the upstream, beginning which until wasthe middle controlled of the by month. the two The reservoirs. reason for In this addition, could be the related amount to the of rainfalldischarge in Septemberfrom upstream, began which to decrease was controlled relative to by August. the two Thus, reservoirs. the discharge In addition, predictions the amount after th ofe mirainfallddle ofin theSeptember month increased began to rapidlydecrease because relative the to damAugust. storage Thus, of theboth discharge reservoirs predictions reached maximum after the capacity.middle of No the adaptive month increased operation rapidly was undertaken because the in dam October, storage because of both October reservoirs is thereached month maximum of peak Water 2020discharge.capacity., 12, 3210 No Thus, adaptive September operation was thewas last undertaken month of inadaptive October, operation. because October is the month of peak12 of 19 discharge. Thus, September was the last month of adaptive operation.

(a) (b) (c) (a) (b) (c) Figure 10. River discharge predictions at station C2 (a) and storage predictions from July until FigureDecemberFigure 10. River 10. atRiver dischargethe Bhumibol discharge predictions (b predictions) and Sirikit at station at(c ) stationreservoir, C2 ( a C2) and shown (a) storage and with storage predictionsthe green predictions line, from and Julyfrom river until Julydischarge December until at theandDecember Bhumibol storage at (the predictionsb) andBhumibol Sirikit after (b (c) )and adaptive reservoir, Sirikit operation(c shown) reservoir, with with shown the predictive green with line,the data green and from line, river July and discharge until river Decemberdischarge and storage predictions(orangeand storage after dashed adaptive predictions line). operation after adaptive with predictive operation data with from predictive July until data December from July (orange until December dashed line). (orange dashed line).

(a) (b) (c) (a) (b) (c) FigureFigure 11. River11. River discharge discharge predictions predictions atat stationstation C2 C2 ( (aa) )and and storage storage predictions predictions from from August August until until DecemberDecemberFigure at 11. the atRiver the Bhumibol Bhumiboldischarge (b )( bpredictions and) and Sirikit Sirikit at ( (c c)station) reservoir,reservoir, C2 ( showna shown) and withstorage with the the predictionsgreen green line, line, andfrom and river August river discharge dischargeuntil and storageandDecember storage prediction at predictionthe Bhumibol after after adaptive(b adaptive) and Sirikit operation operation (c) reservoir, with shownprediction prediction with data the data greenfrom from line,August August and until river until December discharge December (orange(orangeand dashed storage dashed line).prediction line). after adaptive operation with prediction data from August until December (orange dashed line). The prediction results for September are shown in Figure 12. No adaptive operation was conducted based on predictive data, although most ensemble predictions for reservoir storage were higher than the target storage level (orange dashed line). River discharge in September tended to decrease from the beginning until the middle of the month. The reason for this could be related to the discharge from upstream, which was controlled by the two reservoirs. In addition, the amount of rainfall in September began to decrease relative to August. Thus, the discharge predictions after the middle of the month increased rapidly because the dam storage of both reservoirs reached maximum capacity. No adaptive operation was undertaken in October, because October is the month of peak discharge. Water 2020, 12, x FOR PEER REVIEW 12 of 19 Thus, September was the last month of adaptive operation.

(a) (b) (c)

FigureFigure 12. River12. River discharge discharge prediction prediction atat stationstation C2 C2 ( (aa) )and and storage storage prediction prediction from from September September until until DecemberDecember at the at the Bhumibol Bhumibol (b ()b and) and Sirikit Sirikit ((cc)) reservoir, shown shown with with the the green green line, line, and andriver river discharge discharge and storageand storage prediction prediction after after adaptive adaptive operation operation with predictive predictive data data from from September September until until December December (orange(orange dashed dashed line). line).

4. Discussion This study updated the previous work of Mateo et al. [6] to develop a method for using seasonal streamflow predictive information to support adaptive reservoir operation. This previous study [6] showed promising results for successful mitigation of the 2011 flood without using predictive information. The authors showed the 2011 flood could have been mitigated by releasing stored water during the early wet season, thus giving reservoirs sufficient storage capacity to retain a greater volume of water during the peak period. However, more benefit could be gained by using seasonal predictive information to support reservoir operations. In reality, reservoirs, particularly in Thailand, cannot always release water during the early wet season because water stored during the wet season is held to support society during the dry season. Thus, the significant difference between models with and without predictive information is that the use of predictive information allows water managers to make more suitable decisions to derive the best benefits for society. Previous studies [38–40] used ensemble streamflow prediction to design reservoir operations in snow-dominated river basins. These studies used long-term streamflow prediction. The major difference between these previous studies [38–40] and the current study is the river basin characteristics, because the CPRB, which is located in Thailand, is not affected by snow. In addition, the rule curve is commonly used by Thai water managers to operate reservoirs. Thus, our study used the benefits of the rule curve and seasonal predictive information to support reservoir operation. An annual comparison of reservoir operation with and without the use of predictive information is presented in Figure 13. The green line represents simulated river discharge in 2011 when predictive data were not used to manage operations. The red and orange lines represent simulated river discharge using predictive data based on 2011 operation and PlanM. When predictive information was used, the peak river discharge was significantly lower because the rate of outflow was increased to prevent overflow from both reservoirs as shown in Figure A1 of Appendix A. However, reservoir storage was above the maximum level at the end of October as shown in Figure A2 of Appendix A. The increased rate of outflow could also delay the overflow: Because river discharge in the CPRB peaks at the beginning of October, peak river discharge was lower with the use of predictive information than without it. Furthermore, the river discharge based on PlanM tended to be lower than that based on 2011 operation. By applying seasonal predictions from June to September based on 2011 operation, we mitigated the 2011 flood by around 4%. When we changed the reservoir operation strategy from PlanU to PlanM, we mitigated the 2011 flood by around 8%. When reservoir operation was based on PlanM with predictive data, about 20% of the 2011 flood was mitigated. The reason why reservoir operation under PlanM mitigated the 2011 flood better than under PlanU was that we reduced the maximum target storage of both reservoirs to prevent overflow. In addition, we used predictive data representing river discharge and storage results from the start of the prediction period until the end of 2011. Thus, we mitigated flooding by applying seasonal predictions. Nonetheless, overflow from

Water 2020, 12, 3210 13 of 19

4. Discussion This study updated the previous work of Mateo et al. [6] to develop a method for using seasonal streamflow predictive information to support adaptive reservoir operation. This previous study [6] showed promising results for successful mitigation of the 2011 flood without using predictive information. The authors showed the 2011 flood could have been mitigated by releasing stored water during the early wet season, thus giving reservoirs sufficient storage capacity to retain a greater volume Water 2020, 12, x FOR PEER REVIEW 13 of 19 of water during the peak period. However, more benefit could be gained by using seasonal predictive wasinformation used, the topeak support river reservoirdischarge operations.was significantly In reality, lower reservoirs, because the particularly rate of outflow in Thailand, was increased cannot toalways prevent release overflow water from during both the reservoirs early wet as season shown because in Figure water A1 of stored Appendix during A the. However, wet season reservoir is held storageto support was societyabove the during maximum the dry level season. at the Thus, end theof October significant as shown difference in Figure between A2 models of Appendix with and A. Twithouthe increased predictive rate of information outflow could is that also the delay use ofthe predictive overflow: information Because river allows discharge water in managers the CPRB to peaksmake more at the suitable beginning decisions of October, to derive peak the bestriver benefits discharge for society. was lower with the use of predictive informationPrevious than studies without [38 –it.40 ]Furthermore, used ensemble the streamflowriver discharge prediction based toon designPlanM reservoirtended to operations be lower thanin snow-dominated that based on 2011 river operation. basins. These studies used long-term streamflow prediction. The major differenceBy applying between seasonal these previous predictions studies from [38 June–40] to and September the current based study on is 2011 the riveroperation, basin characteristics,we mitigated thebecause 2011 the flood CPRB, by around which is 4%. located When in weThailand, changed is not the a ff reservoirected by operation snow. Inaddition, strategy the from rule PlanU curve to is PlanM,commonly we usedmitigated by Thai the water 2011 managersflood by around to operate 8%. reservoirs. When reservoir Thus, ouroperation study usedwas thebased benefits on PlanM of the withrule curvepredictive and seasonaldata, about predictive 20% of the information 2011 flood to was support mitigated. reservoir The operation.reason why reservoir operation underAn PlanM annual mitigated comparison the 2011 of reservoir flood better operation than withunder and PlanU without was thethat use we of reduced predictive theinformation maximum targetis presented storage in Figure of both 13 . reservoirs The green lineto prevent represents overflow. simulated In river addition, discharge we in used 2011 predictive when predictive data representingdata were not river used discharge to manage and operations. storage results The red from and orangethe start lines of the represent prediction simulated period river until discharge the end ofusing 2011. predictive Thus, we datamitigated based flooding on 2011 by operation applying and seasonal PlanM. predictions. When predictive Nonetheless, information overflow was from used, boththe peak reservoirs river discharge occurred was with significantly every reservoir lower operation because the at rate the of end outflow of September. was increased However, to prevent the overflowoverflow time from and both length reservoirs were as delayed shown inand Figure shifted A1 withof Appendix changesA in. However, target storage. reservoir The storage amount was of waterabove wa thes depleted maximum in level both atreservoirs the end ofprior October to the as huge shown increase in Figure in supply A2 of around Appendix AugustA. The to increased October. Thus,rate of decreasing outflow could the amount also delay of damthe overflow: storage to Because prevent river overflow discharge using in the the predictive CPRB peaks method at the reportedbeginning in ofthis October, study can peak mitigate river discharge flooding, wasparticularly lower with if the the upper use ofrule predictive curve is lowered information to reach than thewithout target it. storage Furthermore, level. Figure the river A3 discharge(in Appendix based B) on shown PlanM the tended comparison to be lower of peak than discharge that based and on percentage2011 operation. of peak discharge different from simulations.

Figure 13. Comparison of river discharge, showing simulated discharge with PlanU operation in 2011 Figure 13. Comparison of river discharge, showing simulated discharge with PlanU operation in 2011 (green line), simulated discharge with PlanM operation in 2011 (blue line), and simulated discharge: (green line), simulated discharge with PlanM operation in 2011 (blue line), and simulated discharge: (1) after adaptive operation based on 2011 operations using seasonal predictive data (red line) and (2) (1) after adaptive operation based on 2011 operations using seasonal predictive data (red line) and (2) after adaptive operation based on PlanM using predictive data (orange line). after adaptive operation based on PlanM using predictive data (orange line). By applying seasonal predictions from June to September based on 2011 operation, we mitigated 5.the Conclusions 2011 ﬂood by around 4%. When we changed the reservoir operation strategy from PlanU to PlanM, This study updates previous studies, in which various reservoir operations for flood mitigation were proposed, using ECMWF predictive information to support adaptive reservoir operation. We developed a method for applying seasonal predictions from the ECMWF, demonstrating the use of predictive information for adaptive reservoir operation in the CPRB in Thailand. We used model H08, a distributed hydrological model with a reservoir module, to obtain predictions of river discharge and reservoir storage with a six-month prediction period. In addition, we used two different operating plans to compare river discharge results. The first (PlanU) was simply to follow the 2011 operation schedule, which followed the 2011 rule curve. The second (PlanM) resembled reservoir operations described in a previous study, in which the maximum target storage in 2011 was reduced. The results for river discharge based on seasonal predictions obtained using PlanU showed that

Water 2020, 12, 3210 14 of 19 we mitigated the 2011 flood by around 8%. When reservoir operation was based on PlanM with predictive data, about 20% of the 2011 flood was mitigated. The reason why reservoir operation under PlanM mitigated the 2011 flood better than under PlanU was that we reduced the maximum target storage of both reservoirs to prevent overflow. In addition, we used predictive data representing river discharge and storage results from the start of the prediction period until the end of 2011. Thus, we mitigated flooding by applying seasonal predictions. Nonetheless, overflow from both reservoirs occurred with every reservoir operation at the end of September. However, the overflow time and length were delayed and shifted with changes in target storage. The amount of water was depleted in both reservoirs prior to the huge increase in supply around August to October. Thus, decreasing the amount of dam storage to prevent overflow using the predictive method reported in this study can mitigate flooding, particularly if the upper rule curve is lowered to reach the target storage level. Figure A3 (in AppendixB) shown the comparison of peak discharge and percentage of peak discharge different from simulations.

5. Conclusions This study updates previous studies, in which various reservoir operations for flood mitigation were proposed, using ECMWF predictive information to support adaptive reservoir operation. We developed a method for applying seasonal predictions from the ECMWF, demonstrating the use of predictive information for adaptive reservoir operation in the CPRB in Thailand. We used model H08, a distributed hydrological model with a reservoir module, to obtain predictions of river discharge and reservoir storage with a six-month prediction period. In addition, we used two different operating plans to compare river discharge results. The first (PlanU) was simply to follow the 2011 operation schedule, which followed the 2011 rule curve. The second (PlanM) resembled reservoir operations described in a previous study, in which the maximum target storage in 2011 was reduced. The results for river discharge based on seasonal predictions obtained using PlanU showed that seasonal predictive information could be useful for flood mitigation and slightly decreased the peak river discharge, by about 4%, for 2011. In contrast, the results for PlanM showed that peak river discharge was reduced by about 20% relative to the observed 2011 river discharge. The reservoir operation under PlanM, which set target storage levels for each reservoir, tended to make each reservoir release its storage earlier than those under PlanU. That is, a combination of seasonal prediction and PlanM best mitigated flooding.

Author Contributions: Conceptualization, W.K. and S.K.; methodology, W.K.; validation, W.K., S.Y. and S.K.; software, W.K.; formal analysis, W.K.; data curation, W.K.; writing—original draft preparation, W.K.; writing—review and editing, W.K., S.Y. and S.K.; supervision, S.K.; project administration, S.K.; funding acquisition, S.K. All authors have read and agreed to the published version of the manuscript. Funding: This research was supported by Japan Society for Promotion of Science (JSPS) Grant-in-Aid for Scientific Research (JSPS KAKENHI Grant Numbers 16H06291). Acknowledgments: The first author is very grateful to Dhanapala Arachindra Sachindra and Natsuki Yshida. This research was supported partially by the Japan Science and Technology Agency/Japan International Cooperation Agency, and the Science and Technology Research Partnership for Sustainable Development (JST/JICA, SATREPS). Conflicts of Interest: The authors declare no conflict of interest.

Appendix A Figure A1 shows simulated outflow from the Bhumibol and Sirikit reservoirs in 2011 under PlanU and PlanM operations. The results for outflow using seasonal predictions are indicated by the red and blue lines for PlanU and PlanM, respectively. Because it used seasonal predictions, the rate of outflow under PlanU was elevated only in August, whereas that under PlanM was elevated in both July and August. Because the increased rate of outflow was higher in July and August when seasonal predictive data were used, as shown in Figure A1 for both reservoirs, reservoir storage in July and August was lower under PlanM, which used seasonal predictive data, than under PlanU, as shown Water 2020, 12, x FOR PEER REVIEW 14 of 19

WaterCooperation 2020, 12, x Agency, FOR PEER and REVIEW the Science and Technology Research Partnership for Sustainable Development14 of 19 (JST/JICA, SATREPS). Cooperation Agency, and the Science and Technology Research Partnership for Sustainable Development (JST/JICA,Conflicts of SATREPS). Interest: The authors declare no conflicts of interest.

ConflictsAppendix of AInterest: The authors declare no conflicts of interest.

AppendixFigure A A1 shows simulated outflow from the Bhumibol and Sirikit reservoirs in 2011 under PlanU and PlanM operations. The results for outflow using seasonal predictions are indicated by the red andFigure blue A1 lines shows for PlanU simulated and PlanM, outflow respectively. from the Bhumibol Because andit used Sirikit seasonal reservoirs predictions, in 2011 the under rate PlanUof outflow and PlanMunder operationPlanU wass. Theelevated results only for inoutflow August, using whereas seasonal that predictions under PlanM are indicatedwas elevated by the in redboth and July blue and lines August. for PlanU Because and the PlanM, increased respectively. rate of outflow Because was it used higher seasonal in July predictions, and August the when rate ofseasonal outflow predictive under PlanU data waswere elevated used, as onlyshown in inAugust, Figure whereas A1 for both that reservoirs,under PlanM reservoir was elevated storage in both July and August. Because the increased rate of outflow was higher in July and August when JulyWater and2020 ,August12, 3210 was lower under PlanM, which used seasonal predictive data, than under PlanU,15 of as 19 seasonalshown in predictive Figure A2. data In were addition, used, adaptive as shown ope inration Figure following A1 for both PlanM reservoirs, slightly reservoir lowered storage reservoir in Julystorage and compared August was to lowerPlanU. under However, PlanM, adaptive which usedoperation seasonal under predictive PlanM usingdata, than seasonal under predictions PlanU, as shownloweredin Figure in reservoir A2 Figure. In addition,A2. storage, In addition, which adaptive in adaptive turn operation mitigated ope followingration flooding following PlanM during slightly PlanM the peak slightly lowered flow loweredperiod reservoir in reservoirOctober, storage storagemorecompared than compared toad PlanU.aptive to operation However,PlanU. However, under adaptive PlanU. adaptive operation operation under PlanM under using PlanM seasonal using seasonal predictions predictions lowered loweredreservoir reservoir storage, whichstorage, in which turn mitigated in turn mitigated ﬂooding duringflooding the during peak ﬂowthe peak period flow in October,period in more October, than moreadaptive than operation adaptive underoperation PlanU. under PlanU.

(a) (b)

Figure A1. Comparison(a) of outflow at (a) Bhumibol dam and (b) Sirikit dam(b) , showing simulated outflowFigure A1. usingComparison PlanU operation of outﬂow in 2011 at (green (a) Bhumibol line) and dam simulated and (b )discharge Sirikit dam, using showing PlanM simulatedoperation Figure A1. Comparison of outflow at (a) Bhumibol dam and (b) Sirikit dam, showing simulated inoutﬂow 2011 (blue using line) PlanU: (1) operation after adaptive in 2011 operation (green line) using and predictive simulated data discharge based using on 2011 PlanM operations operation (red in outflow using PlanU operation in 2011 (green line) and simulated discharge using PlanM operation line)2011 and (blue (2) line): after (1) adaptive after adaptive operation operation using predictive using predictive data based data on based PlanM on 2011(yellow operations line). (red line) in 2011 (blue line): (1) after adaptive operation using predictive data based on 2011 operations (red and (2) after adaptive operation using predictive data based on PlanM (yellow line). line) and (2) after adaptive operation using predictive data based on PlanM (yellow line).

(a) (b)

Figure A2. Comparison of dam storage at (a) Bhumibol dam and (b) Sirikit dam, showing simulated Figure A2. Comparison(a) of dam storage at (a) Bhumibol dam and (b) Sirikit dam,(b) showing simulated dam storage using PlanU operation in 2011 (green line) and simulated discharge using PlanM operation dam storage using PlanU operation in 2011 (green line) and simulated discharge using PlanM in 2011 (blue line): (1) after adaptive operation using predictive data based on 2011 operations (red line): Figureoperation A2 . inComparison 2011 (blue of line) dam: storage(1) after at adaptive(a) Bhumibol operation dam and using (b) predictiveSirikit dam, data showing based simulated on 2011 damand (2) storage after adaptiveusing PlanU operation operation using predictivein 2011 (green data based line) onand PlanM simulated (yellow discharge line). The using yellow PlanM dash

operationline shows in the 2011 target (blue storage line) of: ( PlanM1) after in adaptive this study. operation using predictive data based on 2011

Water 2020, 12, x FOR PEER REVIEW 15 of 19

(a) (b)

Figure A2. Comparison of dam storage at (a) Bhumibol dam and (b) Sirikit dam, showing simulated dam storage using PlanU operation in 2011 (green line) and simulated discharge using PlanM operation in 2011 (blue line): (1) after adaptive operation using predictive data based on 2011 operations (red line): and (2) after adaptive operation using predictive data based on PlanM (yellow line). The yellow dash line shows the target storage of PlanM in this study.

Water 2020, 12, 3210 16 of 19 Appendix B

FigureAppendix A3 compares B simulated discharge at station C2 in 2011. Overall, adaptive operation under PlanM reducedFigure peak A3 compares discharge. simulated Moreover, discharge PlanM at station operation C2 in 2011. using Overall, seasonal adaptive predictions operation under mitigated floodingPlanM in 2011. reduced peak discharge. Moreover, PlanM operation using seasonal predictions mitigated ﬂooding in 2011.

Figure A3. Comparison of peak discharge (blue bar color) and percentage of peak discharge different Figure Afrom3. Comparison simulation (red of bar peak color) discharge between peak (blue simulated bar color) discharge and usingpercentage PlanU operation of peak and discharge simulated different from simulationdischarge using (red PlanM bar color) operation between in 2011 and peak simulated simulated discharge: discharge (1) after adaptive using PlanU operation operation using and predictive data and (2) after adaptive operation using predictive data by PlanM. simulated discharge using PlanM operation in 2011 and simulated discharge: (1) after adaptive operationAppendix using C predictive data and (2) after adaptive operation using predictive data by PlanM. Figure A4 shows the results for seasonal rainfall prediction after bias correction from June, July, AppendixAugust, C and September until December 2011. The results are on a daily time scale. Figure A5 shows scatter plots of observed rainfall (y-axis) and predicted rainfall (x-axis) and the ensemble range for FigureJune A4 until shows December, the results with panels for seasonal showing di rainfallfferent prediction prediction lead after times. bias For correction example, panel from (d) June, in July, August, Figureand September A5 shows a until scatter December plot of observed 2011. and The predicted results rainfall,are on witha daily yellow, time red, scale. blue, Figure and green A5 shows scatter plotsrepresenting of observed rainfall rainfall predictions (y- foraxis) September and predicted with 0, 1, 2,rainfall and 3 month (x-axis) lead and times, the respectively. ensemble range for June until December,Based on these with results, panels rainfall showing predictions different tended prediction to be underestimated lead times. relativeFor example, to observed panel (d) in rainfall data. From these results, the adaptive reservoir operation, which increased the outflow rate from reservoirs, only occurred during August when using PlanU to operate in this study. In addition,

adaptive reservoir operation could be started prior to August to mitigate ﬂooding in 2011 if the rainfall prediction was more accurate. Water 2020, 12, 3210 17 of 19 Water 2020, 12, x FOR PEER REVIEW 16 of 19

(a) (b)

Water 2020Figure, 12, A4.xFigure FORComparison A4.PEER Comparison REVIEW of dailyof daily rainfall rainfall data, data, showing showing observed observed rainfall rainfall (blue line), (blue ensemble line), seasonal ensemble seasonal17 of 19 rainfall predictionsrainfall predictions (light (light pink pink dashed dashed line), line), and and averageaverage ensemble ensemble seasonal seasonal rainfall rainfall prediction prediction (pink (pink line). Rainfallline). predictions Rainfall predictions from: (a )from June: (a until) June until December December 2011, 2011, ( b(b)) JulyJuly until until December December 2011, 2011,(c) August (c) August until until December 2011, and (d) September until December 2011. December 2011, and (d) September until December 2011.

(a) (b) (c) (a) (b) (c)

(d) (e) (d) (e) Figure A5. Scatter plot of monthly rainfall data, showing observed rainfall (y-axis) and average rainfall Figure A5. Scatter plot of monthly rainfall data, showing observed rainfall (y-axis) and average prediction (x-axis) with ensemble seasonal rainfall predictions (colored lines). Rainfall predictions for Figure A5. Scatterrainfall plot of prediction monthly rainfall (x-axis) data, with showing ensemble observed seasonal rainfall rainfall (y - predictionsaxis) and average (colored lines). Rainfall (a) June 2011, (b) July 2011, (c) August 2011, (d) September 2011, and (e) October 2011. rainfall predictionpredictions (x-axis) with for (a ensemble) June 2011, seasonal (b) July rainfall2011, (c ) predictionsAugust 2011, (colored (d) Sep tember lines). 2011, Rainfall and (e) October 2011. predictions for (a) June 2011, (b) July 2011, (c) August 2011, (d) September 2011, and (e) October 2011. References

References 1. Mercer, J.; Kelman, I.; Taranis, L.; Suchet-Pearson, S. Framework for integrating indigenous and scientific 1. Mercer, J.; Kelman,knowledge I.; Taranis, for L.; disaster Suchet- Pearson,risk reduction. S. Framework Disasters for2009 integrating, 34, 214–239 indigenous, doi:10.1111/j.1467 and scientific-7717.2009.01126 .x. knowledge for2. disasterFadul, risk E.; reduction. Masih, I.; Disasters De Fraiture, 2009 , C. 34 Adaptation, 214–239, doi:10.1111/j.1467 strategies to cope-7717.2009.01126 with low, high.x. and untimely floods: 2. Fadul, E.; Masih, Lessons I.; De Fraiture, from theC. Adaptation Gash spate strategies irrigation to copesystem, with Sudan. low, highAgric. and Water untimely Manag. floods: 2019 , 217, 212–225, Lessons from thedoi:10.1016/j.agwat.2019.02.035. Gash spate irrigation system, Sudan. Agric. Water Manag. 2019, 217, 212–225, doi:10.1016/j.agwat.2019.02.035.3. Tasseff, B.; Bent, R.; Van Hentenryck, P. Optimization of Structural Flood Mitigation Strategies. Water 3. Tasseff, B.; Bent, R.Resour.; Van Res. Hentenryck, 2019, 55, 1490 P. Optimization–1509, doi:10.1029/2018wr024362. of Structural Flood Mitigation Strategies. Water Resour. Res. 20194. , 55Berhane,, 1490–1509 G.; Gebreyohannes,, doi:10.1029/2018wr024362. T.; Martens, K.; Walraevens, K. Overview of micro-dam reservoirs (MDR) in 4. Berhane, G.; Gebreyohannes,Tigray (nor T.;thern Martens, Ethiopia): K.; Walraevens, Challenges K. Overview and benefits. of micro -J.dam Afr. reservoirs Earth (MDR) Sci. 2016 in , 123, 210–222, Tigray (northerndoi:10.1016/j.jafrearsci.2016.07.022. Ethiopia): Challenges and benefits. J. Afr. Earth Sci. 2016, 123, 210–222, doi:10.1016/j.jafrearsci.2016.07.022.5. Ehteram, M.; Mousavi, S.-F.; Karami, H.; Farzin, S.; Emami, M.; Othman, F.; Amini, Z.; Kisi, O.; El-Shafie, 5. Ehteram, M.; Mousavi,A. Fast S. -convergenceF.; Karami, H.; optimization Farzin, S.; Emami,model for M.; single Othman, and multiF.; Amini,-purposes Z.; Kisi, reservoirs O.; El-Shafie, using hybrid algorithm. A. Fast convergenceAdv. optimization Eng. Informatics model 2017 for, single32, 287 and–298 multi, doi:10.1016/j.aei.2017.04.001.-purposes reservoirs using hybrid algorithm. Adv. Eng. Informatics6. Mateo, 2017 ,C.M.; 32, 287 Hanasaki,–298, doi :10.1016/j.aei.2017.04.001.N.; Komori, D.; Tanaka, K.; Kiguchi, M.; Champathong, A.; Sukhapunnaphan, T.; 6. Mateo, C.M.; Hanasaki,Yamazaki, N.; Komori, D.; Oki, D.; T. AssessingTanaka, K.; the Kiguchi, impacts M.; of Champathong,reservoir operation A.; Sukhapunnaphan, to floodplain inundation T.; by combining Yamazaki, D.; Oki,hydrological, T. Assessing reservoithe impactsr management, of reservoir andoperation hydrodynamic to floodplain models. inundation Water Resour.by combining Res. 2014 , 50, 7245–7266, hydrological, reservoidoi:10.1002/2013wr014845.r management, and hydrodynamic models. Water Resour. Res. 2014, 50, 7245–7266, doi:10.1002/2013wr014845.7. Ehsani, N.; Vorosmarty, C.; Fekete, B.; Stakhiv, E.Z. Reservoir operations under climate change: Storage 7. Ehsani, N.; Vorosmarty,capacity C.; options Fekete, to B.; mitigate Stakhiv, risk. E.Z. J. Hydrol.Reservoir 2017 operations, 555, 435 –under446, doi:10.1016/j.jhydrol.2017.09.008. climate change: Storage capacity options8. toLiu, mitigate H.; Deng, risk. B.;J. Hydrol. Liu, Y.; 2017 Jiang,, 555 C.;, 435Wu,–446 Z.;, Long,doi:10.1016/j.jhydrol.2017.09.008. Y. Preliminary Numerical Analysis of the Efficiency of a 8. Liu, H.; Deng, B.; Liu,Central Y.; Jiang,Lake ReservoirC.; Wu, Z.; in Long, Enhancing Y. Preliminary the Flood Numerical and Drought Analysis Resistance of the of Efficiency Dongting of Lake. a Water 2018, 10, Central Lake Reservoir225, doi:10.3390/w10020225. in Enhancing the Flood and Drought Resistance of Dongting Lake. Water 2018, 10, 225, doi:10.3390/w10020225.

Water 2020, 12, 3210 18 of 19

References

1. Mercer, J.; Kelman, I.; Taranis, L.; Suchet-Pearson, S. Framework for integrating indigenous and scientific knowledge for disaster risk reduction. Disasters 2009, 34, 214–239. [CrossRef] 2. Fadul, E.; Masih, I.; De Fraiture, C. Adaptation strategies to cope with low, high and untimely floods: Lessons from the Gash spate irrigation system, Sudan. Agric. Water Manag. 2019, 217, 212–225. [CrossRef] 3. Tasseff, B.; Bent, R.; Van Hentenryck, P.Optimization of Structural Flood Mitigation Strategies. Water Resour. Res. 2019, 55, 1490–1509. [CrossRef] 4. Berhane, G.; Gebreyohannes, T.; Martens, K.; Walraevens, K. Overview of micro-dam reservoirs (MDR) in Tigray (northern Ethiopia): Challenges and benefits. J. Afr. Earth Sci. 2016, 123, 210–222. [CrossRef] 5. Ehteram, M.; Mousavi, S.-F.; Karami, H.; Farzin, S.; Emami, M.; Othman, F.; Amini, Z.; Kisi, O.; El-Shafie, A. Fast convergence optimization model for single and multi-purposes reservoirs using hybrid algorithm. Adv. Eng. Informatics 2017, 32, 287–298. [CrossRef] 6. Mateo, C.M.; Hanasaki, N.; Komori, D.; Tanaka, K.; Kiguchi, M.; Champathong, A.; Sukhapunnaphan, T.; Yamazaki, D.; Oki, T. Assessing the impacts of reservoir operation to floodplain inundation by combining hydrological, reservoir management, and hydrodynamic models. Water Resour. Res. 2014, 50, 7245–7266. [CrossRef] 7. Ehsani, N.; Vorosmarty,C.; Fekete, B.; Stakhiv,E.Z. Reservoir operations under climate change: Storage capacity options to mitigate risk. J. Hydrol. 2017, 555, 435–446. [CrossRef] 8. Liu, H.; Deng, B.; Liu, Y.; Jiang, C.; Wu, Z.; Long, Y. Preliminary Numerical Analysis of the Efficiency of a Central Lake Reservoir in Enhancing the Flood and Drought Resistance of Dongting Lake. Water 2018, 10, 225. [CrossRef] 9. Murray, V.; Ebi, K.L. IPCC Special Report on Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (SREX). J. Epidemiol. Community Health 2012, 66, 759–760. [CrossRef] 10. Komori, D.; Nakamura, S.; Kiguchi, M.; Nishijima, A.; Yamazaki, D.; Suzuki, S.; Kawasaki, A.; Oki, K.; Oki, T. Characteristics of the 2011 Chao Phraya River flood in Central Thailand. Hydrol. Res. Lett. 2012, 6, 41–46. [CrossRef] 11. Gale, E.L.; Saunders, M.A. The 2011 Thailand flood: Climate causes and return periods. Weather 2013, 68, 233–237. [CrossRef] 12. World Bank Group. Available online: https://openknowledge.worldbank.org/handle/10986/26862 (accessed on 3 October 2020). 13. Suiadee, W.; Tingsanchali, T. A combined simulation–genetic algorithm optimization model for optimal rule curves of a reservoir: A case study of the Nam Oon Irrigation Project, Thailand. Hydrol. Process. 2007, 21, 3211–3225. [CrossRef] 14. Pinthong, P.; Das Gupta, A.; Babel, M.S.; Weesakul, S. Improved Reservoir Operation Using Hybrid Genetic Algorithm and Neurofuzzy Computing. Water Resour. Manag. 2008, 23, 697–720. [CrossRef] 15. Wichakul, S.; Tachikawa, Y.; Shiiba, M.; Yorozu, K. DEVELOPING A REGIONAL DISTRIBUTED HYDROLOGICAL MODEL FOR WATER RESOURCES ASSESSMENT AND ITS APPLICATION TO THE CHAO PHRAYA RIVER BASIN. J. Jpn. Soc. Civ. Eng. Ser. B1 (Hydraul. Eng.) 2013, 69, I_43–I_48. [CrossRef] 16. Wichakul, S.; Tachikawa, Y.; Shiiba, M.; Yorozu, K. Development of a Flow Routing Model Including Inundation Effect for the Extreme Flood in the Chao Phraya River Basin, Thailand 2011. J. Disaster Res. 2013, 8, 415–423. [CrossRef] 17. Viel, C.; Beaulant, A.-L.; Soubeyroux, J.-M.; Céron, J.-P. How seasonal forecast could help a decision maker: An example of climate service for water resource management. Adv. Sci. Res. 2016, 13, 51–55. [CrossRef] 18. Long, Y.; Wang, H.; Jiang, C.; Ling, S. Seasonal Inflow Forecasts Using Gridded Precipitation and Soil Moisture Information: Implications for Reservoir Operation. Water Resour. Manag. 2019, 33, 3743–3757. [CrossRef] 19. Prasanchum, H.; Kangrang, A. Optimal reservoir rule curves under climatic and land use changes for Lampao Dam using Genetic Algorithm. KSCE J. Civ. Eng. 2017, 29, 1247–1364. [CrossRef] 20. Kangrang, A.; Prasanchum, H.; Hormwichian, R. Development of Future Rule Curves for Multipurpose Reservoir Operation Using Conditional Genetic and Tabu Search Algorithms. Adv. Civ. Eng. 2018, 2018, 1–10. [CrossRef] 21. Harrigan, S.; Prudhomme, C.; Parry, S.; Smith, K.; Tanguy, M. Benchmarking ensemble streamflow prediction skill in the UK. Hydrol. Earth Syst. Sci. 2018, 22, 2023–2039. [CrossRef] Water 2020, 12, 3210 19 of 19

22. Schaake, J.C.; Hamill, T.M.; Buizza, R.; Clark, M. HEPEX: The Hydrological Ensemble Prediction Experiment. Bull. Am. Meteorol. Soc. 2007, 88, 1541–1548. [CrossRef] 23. Fang, G.; Yang, J.; Chen, Y.; Zammit, C. Comparing bias correction methods in downscaling meteorological variables for a hydrologic impact study in an arid area in China. Hydrol. Earth Syst. Sci. 2015, 19, 2547–2559. [CrossRef] 24. Li, Y.; Jiang, Y.; Lei, X.; Tian, F.; Duan, H.; Lu, H. Comparison of Precipitation and Streamflow Correcting for Ensemble Streamflow Forecasts. Water 2018, 10, 177. [CrossRef] 25. Lucatero, D.; Madsen, H.; Refsgaard, J.C.; Kidmose, J.; Jensen, K.H. Seasonal streamflow forecasts in the Ahlergaarde catchment, Denmark: The effect of preprocessing and post-processing on skill and statistical consistency. Hydrol. Earth Syst. Sci. 2018, 22, 3601–3617. [CrossRef] 26. Crochemore, L.; Ramos, M.-H.; Pappenberger, F. Bias correcting precipitation forecasts to improve the skill of seasonal streamflow forecasts. Hydrol. Earth Syst. Sci. 2016, 20, 3601–3618. [CrossRef] 27. Office of Natural Water Resources Committee (ONWRC) of Thailand Working Group. Available online: http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/SC/pdf/wwap_Thailand_case% 20studies1_EN.pdf (accessed on 3 October 2020). 28. Johnson, S.J.; Stockdale, T.N.; Ferranti, L.; Balmaseda, M.A.; Molteni, F.; Magnusson, L.; Tietsche, S.; Decremer, D.; Weisheimer, A.; Balsamo, G.; et al. SEAS5: The new ECMWF seasonal forecast system. Geosci. Model Dev. 2019, 12, 1087–1117. [CrossRef] 29. Yoshimura, K.; Sakimura, T.; Oki, T.; Kanae, S.; Seto, S. Toward flood risk prediction: A statistical approach using a 29-year river discharge simulation over Japan. Hydrol. Res. Lett. 2008, 2, 22–26. [CrossRef] 30. Kotsuki, S.; Tanaka, K.; Kojiri, T.; Hamaguchi, T. The Water Budget Analysis with Land Surface Model in Chao Phraya River Basin. In Proceedings of the 23rd annual conference, Japan Society of Hydrology and Water Resources, Tokyo, Japan, 8 September 2010; pp. 44–45. 31. Hanasaki, N.; Kanae, S.; Oki, T.; Masuda, K.; Motoya, K.; Shirakawa, N.; Shen, Y.; Tanaka, K. An integrated model for the assessment of global water resources – Part 1: Model description and input meteorological forcing. Hydrol. Earth Syst. Sci. 2008, 12, 1007–1025. [CrossRef] 32. Manabe, S. CLIMATE AND THE OCEAN CIRCULATION. Mon. Weather. Rev. 1969, 97, 775–805. [CrossRef] 33. Robock, A.; Vinnikov, K.Y.; Schlosser, C.A.; Speranskaya, N.A.; Xue, Y. Use of Midlatitude Soil Moisture and Meteorological Observations to Validate Soil Moisture Simulations with Biosphere and Bucket Models. J. Clim. 1995, 8, 15–35. [CrossRef] 34. Oki, T.; Sud, Y.C. Design of Total Runoff Integrating Pathways (TRIP)—A Global River Channel Network. Earth Interactions 1998, 2, 1. [CrossRef] 35. Themeßl, M.J.; Gobiet, A.; Heinrich, G. Empirical-statistical downscaling and error correction of regional climate models and its impact on the climate change signal. Clim. Chang. 2012, 112, 449–468. [CrossRef] 36. Thaiwater. Available online: https://www.thaiwater.net/web/attachments/25basins/10-chaopraya.pdf (accessed on 3 October 2020). 37. Cannon, A.J.; Sobie, S.R.; Murdock, T.Q. Bias Correction of GCM Precipitation by Quantile Mapping: How Well Do Methods Preserve Changes in Quantiles and Extremes? J. Clim. 2015, 28, 6938–6959. [CrossRef] 38. Anghileri, D.; Voisin,N.; Castelletti, A.; Pianosi, F.; Nijssen, B.; Lettenmaier, D.P.Value of long-term streamflow forecasts to reservoir operations for water supply in snow-dominated river catchments. Water Resour. Res. 2016, 52, 4209–4225. [CrossRef] 39. Alemu, E.T.; Palmer, R.N.; Polebitski, A.S.; Meaker, B. Decision Support System for Optimizing Reservoir Operations Using Ensemble Streamflow Predictions. J. Water Resour. Plan. Manag. 2011, 137, 72–82. [CrossRef] 40. Voisin, N.; Hamlet, A.F.; Graham, L.P.; Pierce, D.W.; Barnett, T.P.; Lettenmaier, D.P. The Role of Climate Forecasts in Western U.S. Power Planning. J. Appl. Meteorol. Clim. 2006, 45, 653–673. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).