JOINT OBSERVATION AND EVALUATION OF THE EMERGENCY WATER SUPPLEMENT FROM CHINA TO THE MEKONG RIVER

The study analyses the causes of the drought in the Lancang-Mekong Basin, influence of Lancang cascade reservoirs operation on dry season volume of the Mekong River, hydrological influence of the emergency water supplement in 2016 on water level, and discharge and volume of the Mekong mainstream.


List of figures
List of tables

Executive Summary
Recent meteorological and agricultural drought conditions over the Mekong Basin have worsened and triggered China to implement its emergency water supplement from its cascades dams in the Lancang River to the Mekong River by increasing the water discharge from Yunnan's Jinghong Reservoir. China decided to implement its emergency water supplement in a 'threephase plan': (1) from 9 March to 10 April 2016, with an average daily discharge of no less than 2,000 m 3 /s; (2) from 11 April to 20 April 2016 with the discharge of no less than 1,200 m 3 /s; and (3) from 21 April to 31 May 2016 with the discharge of no less than 1,500 m 3 /s. The Mekong River Commission acknowledges this action by China, in which China has also stated that it implemented the water supplement at a challenging time, especially within the context where China itself was also suffering from drought, which has affected its household water supply and agricultural production.
The China's Ministry of Water Resources and Mekong River Commission Secretariat then coorganised experts from both sides to conduct a Joint Observation and Evaluation of the Emergency Water Supplement from China and its effect of easing the drought situation in the Mekong Basin.
The scope of the Joint Observation and Evaluation covers: (1) Temporal Scope -dry season of 2016, which runs from 1 December 2015 to 31 May 2016 and especially during the emergency water supplement period from 15 March to 15 May 2016; and (2) Spatial Scope -from Jinghong hydrological station on the Lancang River to the Mekong Delta.
In this study, an agreement was reached on exchange and sharing of 21 hydrological stations for water level and 7 stations for discharge on the Mekong mainstream from the Mekong River Commission Secretariat; and 1 station for water level and discharge on the Lancang mainstream and 1 station for water level and discharge on the main tributary of Man An, from China.
The analyses cover (1) Cause of the drought in the Lancang-Mekong Basin considering temperature, rainfall, flows, soil moisture and water stress; (2) Overall influence of Lancang cascade reservoirs operation on dry season volume of the Mekong River; (3) Hydrological influence of the emergency water supplement in 2016 on water level, discharge and volume of the Mekong mainstream; (4) Net contribution of the water supplement to discharge of the Mekong River; (5) Variation of water level and discharge of the Mekong mainstream during the water supplement; (6) Flow propagation along the mainstream; and (7) Salinity variation in the Mekong Delta during the period of the emergency water supplement.
The Joint Observation and Evaluation of the Emergency Water Supplement from China to the Mekong River were jointly and objectively conducted with the Mekong River Commission Secretariat and Ministry of Water Resources of China. In the course of this study, besides regular hydrological data sharing in the flood season from China, additional daily water level and discharge for the dry season of 2016 and its long term average of 1960-2009 and 2010-2015, from both sides were exchanged and used in the analyses of this report. Similarly, methodology of the analyses were jointly developed and adopted. Analyses were carried out and the results  Monitoring at Chiang Khan suggests that additional water of 300 m 3 /s for one day on top of the emergency water supplement from China was detected on 27 March 2016. This additional water arrived at Nong Khai on 28 March, at Nakhon Phanom on 31 March, at Mukdahan on 1 April, at Pakse on 3 April and at Stung Treng on 4 April 2016. Immediately after the peak of the additional water, a drop in discharge of 300 m 3 /s was recorded on 31 March 2016.
 Total volume in the dry season of 2016 (December 2015 to May 2016) at Jinghong presented huge portion (40%-89%) of the total volume at different stations along the Mekong mainstream. Additionally, the volume from 10 March to 10 April 2016, which was first period of the emergency water supplement, claimed significant portion, specifically 99% at Chiang Saen, 92% at Nong Khai and 58% at Stung Treng. Similarly, net contribution of the water supplement in term of discharge to total discharge was 47% at Jinghong, 44% at Chiang Saen, 38% at Nong Khai and 22% at Stung Treng. This contribution also alleviated salinity intrusion in the Mekong Delta.

Recommendation
During conduct of the Joint Observation and Evaluation, discussion and exchange between the Mekong River Commission Secretariat and China were sincere with warmth and friendliness. Both parties respected each other views with mutual understanding. It is therefore recommended this kind of study and working attitude should continue to boost strong foundation for further cooperation between China, Mekong River Commission Secretariat and its Member Countries.
This good spirit of cooperation should keep its momentum and be extended to further study on Hydrological Impact of the Lancang Hydropower Cascade on Downstream Floods and Droughts. Likewise, future direction of the study should also focus on positive and negative impact of water resources and hydropower development in the tributaries of the Mekong mainstream.
Limitation of the study Due to limited data and time constraints at the time of the study, the detailed calculations could not be performed; only monthly average computations were normally conducted. Additionally, processes, impacts and linkages of (meteorological, agricultural, hydrological and socioeconomic) droughts and relationship between the drought and global extreme events, namely the El Niño or La Niña were not thoroughly performed. Moreover, detailed evaluation was hampered by limited data from the release of reservoirs on the tributaries of the Mekong River and good quality of hydrological data including rating curves before 2009. Similarly, flow contribution from the Tonle Sap Lake and flow distribution in the Mekong and Bassac Rivers downstream Phnom Penh were not included. Furthermore, several assumptions, including travelling time in the section between Kratie and the Mekong Delta, were used in the analyses. It is also recognised that fully comprehensive salinity analysis in the Mekong Delta could not be performed without additional effort on salinity modelling. In sum, the analyses in this report focused mainly on general hydrological situation and average condition of flows on the Lancang-Mekong mainstream.

Background
Observation of global land and ocean temperature reveals that Years 2015-2016 are the warmest years of record. The El Niño 1 2015-2016 is recorded to be the strongest and has already created weather chaos around the world including the Lancang-Mekong Basin, which have been hit by abnormally dry conditions. Consequently, the countries in the Lancang-Mekong Basin have all suffered in various degrees from the drought caused by the effects of the super El Niño since the end of 2015. Equally, the Mekong Delta is particularly subjected to the most severe drought over the past century, the water level of Mekong River has dropped to the lowest level, which brings considerable damage to the agricultural production of the region and affect living conditions and livelihood of the riparian residents.
The recent drought conditions over the Mekong Basin have worsened and triggered China to implement its emergency water supplement from its cascades dams in the Lancang River to the Mekong River by increasing the water discharge from Yunnan's Jinghong Reservoir. China decided to implement its emergency water supplement in a 'three-phase plan': (1) from 9 March to 10 April 2016, with an average daily discharge of no less than 2,000 m 3 /s; (2)  About half of the total length of the Lancang-Mekong River of about 4,900 km is located in the territory of China. This section of the river flows through narrow areas of high mountains and deep valleys, thus, the volume of the Lancang River accounts only for 13.5% of the annual total volume. The flow regime of the river is mainly influenced by the Monsoon rains that occur every year in the downstream Southeast Asia, especially in Lao PDR, where the basin area covers mainly tropical rainforest and farmland.
The last 170 km of the Lancang River within the territory of China and the section flowing through Myanmar to the border of Thailand, the river transits from section with steep to mild slope, and flows through broad fertile valley. After leaving the territory of China, the river flows along the border of Lao PDR and Myanmar, passes through the border of Lao PDR and Thailand in the downstream of Chiang Saen.
The construction of cascade reservoirs in the mainstream in the middle and lower reaches of the Lancang River has been completed. The Xiaowan Reservoir and Nuozhadu Reservoir have especially the multi-year regulating capacity, with regulating storage of 21.2 billion m 3 in total. By scientifically operating and regulating, the Lancang River cascade reservoirs are capable to balance the water discharge/volume between the wet season and dry season, benefiting the Mekong River on the aspects of flood control, irrigation, navigation and so on.
The Lancang-Mekong Basin can be generally divided into 6 major zones: zone one represents the Lancang Basin in China, and five zones are in the Mekong Basin, coincident with the five fluvial geomorphological reaches along the mainstream. The rationale behind the number and extent of these six reaches of the Lancang-Mekong mainstream encompasses a range of considerations, which include hydrological regime, physiography, landuse, existing, planned and potential resource developments as well as the perceived nodes along the mainstream at which there exist discernable transformations in hydrological response and where the impacts of existing and potential resource developments are likely to be detectable.
Zone 1 -Lancang River in China. The Lancang Basin is mainly characterized by steep alpine valley, located in the under-developed region with extremely inconvenient transportation and deficient natural resources, except extraordinary rich hydropower resources. Water use rate is about 3% in this area, and the water consumed is less than 1% of the total volume of the Lancang-Mekong Basin. This zone contributes about 13.5% volume of the Lancang-Mekong River. The runoff normally comes from rainfall, snowmelt and groundwater. This zone has distinguishing wet season and dry season. The dry season lasts from November to April, during which the volume mainly depends on the snowmelt and groundwater. Additionally, there are currently six hydropower projects on the mainstream of the Lancang River, which could generally increase the volume of the Lancang River by 70% in the dry season and reduce it by 30% in the rainy season. This helps flood mitigating and drought relieving with a proper regulation.
Zone 2 -Chiang Saen to Vientiane/Nong Khai. The additional hydrological contributions to it are generated almost entirely in Lao PDR. This reach is well defined physiographic sub-region of the lower basin being almost entirely mountainous and covered with natural and mostly undisturbed land cover. There is little scope for extensive agricultural development comparative in scale to that further downstream nor are there any plans for any significant water resources developments. Pre-feasibility and feasibility studies of the hydropower potential here, for example, has centred upon small run of river schemes (no regulation beyond diurnal pondage). Although this zone could hardly be described as pristine, the hydrological response from it is certainly the most natural and undisturbed within the basin. In addition, however, it is at the downstream boundary of this zone that virtually every relevant facet of the basin starts to undergo rapid transition.
Zone 3 -Vientiane/Nong Khai to Pakse. The upstream boundary of Zone 3 is the point at which the broader picture of Mekong hydrology changes from one dominated in both wet and dry seasons by the Zone 1 to one increasingly influenced by the contributions from the large left bank tributaries in Lao PDR, namely the Nam Ngum, Nam Theun, Nam Hinboun, Se Bang Fai, Se Bang Hieng and Se Done rivers. Also entering the mainstream within this zone extending to Pakse, is the Mun/Chi system from the right bank and Thailand. The Mun and Chi Rivers are highly developed low relief, agricultural basins with comparatively low runoff potential and significant reservoir storage for dry season irrigation. The left bank Lao tributaries are under steady development in terms of agricultural water demand and hydropower development.
Zone 4 -Pakse to Kratie. The major hydrological contributions to the mainstream in this reach coming from the Sekong, Sesan and Srepok catchments, jointly the largest hydrological subcomponent of the basin. Over 25% of the mean annual flow volume on the Mekong mainstream at Kratie originates from these three river basins, which are therefore a crucial element in the hydrological dynamics of this part of the system, not least with respect to the Tonle Sap Lake flow reversal. Zone 5 -Kratie to Phnom Penh. This reach encompasses the hydraulic complexities of the Cambodian floodplain, the Tonle Sap Lake and River. By this stage over 95% of the total flow has already entered the Mekong system and the balance of emphasis moves from hydrology and water discharge to the critical assessment of water level, overbank storage and flooding and the hydrodynamics that determine the timing, duration and volume of the seasonal flow reversal into and out of the Tonle Sap Lake.
Zone 6 -Phnom Penh to the sea. This stretch defines lower Cambodia, the flow bifurcations and the delta region in Viet Nam, with the total volumes of flow entering the latter observed as the sum of those recorded at Tan Chau and Chau Doc.

Data and methodology
Due to time constraint and resources limitation, only water level and discharge of the key hydrological stations along the Lancang-Mekong mainstream before and after the emergency water supplement are analysed to evaluate the effect of the emergency water supplement. The evaluation covers the generic analysis of the drought in the Lancang-Mekong Basin, analysis of influential factor contribution to flows of the Mekong River, hydrological influence analysis and descriptive benefit analysis of the emergency water supplement.

Scope
The scope of the Joint Observation and Evaluation covers:

Data exchange and sharing
The Joint Observation and Evaluation of Emergency Water Supplement from China to the Mekong River boost the cooperation of the MRC and China to the next level by enhancing hydrological data exchange and sharing from both parties 7 . In this study, an agreement of hydrological data exchange and sharing was reached for 21 hydrological stations for water level and 7 stations for discharge on the Mekong mainstream from the MRCS; and 1 station for water level and discharge on the Lancang mainstream and 1 station for water level and discharge on the main tributary of Man An, from China. Location of the hydrological stations on the Lancang-Mekong mainstream is illustrated in Figure 2.
Data exchange and sharing from the MRCS  Daily water level (21 hydrological stations) and discharge data (

Methodology
The effect of the emergency water supplement from China was evaluated by analysing daily water level and discharge in the dry season 2016, which runs from 1 December 2015  The evaluation focused on the generic analyses of the drought in the Lancang-Mekong Basin, influential hydrological factors of Mekong water flow/volume, and socio-economic benefits of the emergency water supplement. More specifically, the evaluation covered the following:  Cause of the drought in the Lancang-Mekong Basin was assessed by considering monitoring data of temperature, rainfall, flows, soil moisture, water stress and status of the El Niño 2015-2016.
 Overall influence of Lancang cascade dams operation on dry season volume of the Mekong River was analysed by comparing long term average of dry season discharge, then converted to volume of 1960-2009 and 2010-2015.  Hydrological influence of the emergency water supplement in 2016 on water level, discharge and volume of the Mekong mainstream was investigated using monthly average of water level, discharge and volume of the dry season of 1960-2009, 2010-2015 and 2016. Additionally, contribution of volume at Jinghong and volume from stretch along the Mekong mainstream was also studied.
 Net contribution of the emergency water supplement to discharge of the Mekong River was performed using technique of discharge hydrograph separation and adjustment.
 Analysis of variation of water level and discharge of the Mekong mainstream during the water supplement was carried out by placing daily observed water level and derived discharge in the dry season of 2016 with its daily long term average, minimum and maximum of the dry season of 1962-2009 and comparing to individual dry season of 2010-2015.
 Flow propagation along the Mekong mainstream was conducted using variation of daily water level and discharge, and sequence of its events, including the emergency water supplement.
 Salinity variation in the Mekong Delta during the period of the emergency water supplement was analysed using daily maximum and minimum salinity concentration at seven monitoring sites in the Mekong Delta. The discharge of Jinghong Reservoir was then controlled to no less than 1,500 m 3 /s from 21 April to 31 May 2016. The accumulated volume of this period was 5.48 billion m 3 .
From 9 March to 31 May 2016, the total released volume at Jinghong was found to be 12.65 billion m 3 .

Analysis of cause of the drought in the Lancang-Mekong Basin
Cause of the drought in the Lancang-Mekong Basin was assessed by considering status of the El Niño 2015-2016 and monitoring data of temperature, rainfall, flows, soil moisture, and water stress.

Rainfall and inflow discharge to the Lancang Basin
From November 2015 to April 2016, the average rainfall in the upstream catchment of Jinghong was 166.9 mm by statistical analysis according to the measured rainfall in the Lancang Basin, which was decreased by 19% comparing with an average rainfall of 206.4 mm of the same period.
Moreover, inflow discharge to Xiaowan Reservoir and Nuozhadu Reservoir from November 2015 to March 2016 was calculated and then compared to the long term average values, the results are presented in Table 1. The inflow discharges to Xiaowan Reservoir and Nuozhadu Reservoir were found to be reduced by 14%-38% and 10%-38% respectively, comparing to the long term average values of the same period.
In short, from the aspects of measured rainfall and inflow discharge to Xiaowan Reservoir and Nuozhadu Reservoir, it generally suggests that the Langcang Basin was experienced shortage of inflows from November 2015 to March 2016.

Drought in the Mekong Basin
The drought phenomenon is usually grouped into four types 10 :  Meteorological or climatological drought, which focuses on the degree of 'dryness' in terms of an accumulated rainfall deficit.
 Agricultural drought, which expresses the rainfall shortfall primarily in terms of its impact upon crop production through insufficient soil moisture. It generally applies to rainfed agriculture, though irrigated crops can be affected when the water resources themselves become restricted or too expensive.
 Hydrological drought refers to shortages in both surface water and groundwater. This can take the form of critically low river flow, drawn-down reservoir storage and deeper groundwater levels, which make pumped abstraction too expensive or mechanically impossible.
 Socio-economic drought associates the supply and demand consequences for economic goods. Energy outputs from hydropower schemes can be curtailed due to low stream flow and low levels of reservoir storage. There are industrial, agricultural, environmental and social consequences from any curtailment of water supply and water use during droughts.
Meteorological drought is the prime mover in the sequence. The first consequence of an accumulated rainfall deficit is a reduction in soil moisture storage, which once it reaches a critical level, will have impacts upon crops and animal grazing. Agricultural impacts are therefore the first to appear and in most cases provide the first confirmation that there is in fact a drought of any sort at all. These impacts can vary from crop to crop, from farm to farm, from region to region and they depend upon the crop and its resistance to moisture stress, the stage in its growth, whether there are alternative water supplies other than rainfall, and whether livestock can be provided with alternative grazing.
As the rainfall and moisture deficit continues to accumulate, hydrological drought begins to manifest itself. Firstly natural stream flows decrease and fall below normal 11 , ultimately causing a water resources shortfall as reservoirs and other sources of water supply become drawn down. If the event has a long duration and particularly in the case of multi-year droughts, groundwater levels fall and abstraction can become too expensive, too damaging or even mechanically impossible. In this case, a Lognormal distribution applied the General Extreme Value (GEV) is used in the calculation of the ARI. For example, ARI 1:5 (equivalent to 20 times in 100 years, meaning 80% probability of exceedance) daily flow is the mean daily flow that is not equalled or exceeded one in five years in the record under consideration. In the context of the ARI, the mean daily flow is equal to the flow of 50% probability of exceedance.
Finally, drought becomes apparent as a socio-economic process of water shortage and its impacts. There may be food price increases due to reduced domestic agricultural output and (possibly) their replacement with more expensive imports. There may be power rationing due to reduced generating capacity and some industries that are high consumers of water (petrochemicals, metallurgical, bottling plants) might have to reduce production, with secondary consequences for employment, prices, the availability of goods and national economic growth.
As adequate data and information on this domain were not available in the Mekong Basin countries, this section could not be fully developed; however, a general observation of the drought situation is compiled using various sources as presented in Annex A.
The cause of drought in this study was investigated using the status of the El Niño 2015-2016, temperature, rainfall, soil moisture and water index as described in below section.
El Niño 2015-2016and El Niño 1997-1998 The 13 Height of the sea surface is caused by both gravity (which doesn't change much over 100's of years), and the active (always changing) ocean circulation. The normal slow, regular circulation (ocean current) patterns of sea-surface height move up and down (warming and cooling and wind forcing) with the normal progression of the seasons: winter to spring to summer to fall. The differences between what is normal for different times and regions are called anomalies or residuals. The year-to-year and, even, decade-to-decade changes in the ocean that indicate climate events such as the El Niño, La Niña and Pacific Decadal Oscillation are dramatically visualized by these data. Sea surface height is the most modern and powerful tool for taking the 'pulse' of the global oceans (NASA: https://sealevel.jpl.nasa.gov/science/elninopdo/latestdata/, accessed on 7 June 2016).
14 United States National Aeronautics and Space Administration (NASA), http://www.nasa.gov/feature/goddard/nasastudying-2015-el-nino-event-as-never-before, accessed on 8 June 2016. between 3-5 °C above the normal average 15 . However, the condition lasted for only around two weeks. Northeast Thailand, Lao PDR and North Viet Nam, nevertheless, experienced lower temperature than the average in February 2016. The temperature started rising up again in early March across the region and intensifying in some areas with severe condition in April 2016. It is considered that the region received highest temperature at national records.  15 Average daily air temperature is calculated for each grid cell by averaging the twenty-four 1-hourly air temperatures. The dekadal average air temperature is then estimated by averaging the ten daily air temperatures for each grid cell. The temperature data is derived from satellite weather data from the Air Force Weather Agency (AFWA).

Tan Chau
Subsurface soil moisture Subsurface soil moisture 16 levels are best used to monitor an established crop. The subsurface soil moisture is assumed to hold 0-400 mm/m of water depending on the soil's water-holding capacity (based on soil texture and soil depth).
Subsurface soil moisture started getting worse in March 2016 in Thailand, Cambodia and Mekong Delta (Figure 7). The moisture content remained less than 25 mm making unfavourable condition for the crops. The dry condition intensified in the following months of April. Only some small part of the east Thailand received some moisture in fourth week of April as the rain pours down ( Figure 5). Western part of Lao PDR had a better soil moisture condition throughout the dry season 2016.

Normalised Difference Water Index
The Normalized Difference Water Index 17 (NDWI) or water stress for agriculture is a satellitederived index from the Near-Infrared (NIR) and Short Wave Infrared (SWIR) channels. Map of the NDWI depicted in Figure 8 shows that, starting from fourth week of January 2016, the water stress value was already at moderate level in northeast Thailand and around floodplain of the Tonle Sap Lake of Cambodia. The condition became worse in February to end of April, which would damage a large area of agricultural production in northeast Thailand and Cambodia. The water stress conditions became less serious towards the end of April in these two countries, thanks to rainfall over the Mekong Basin. However, it looks relatively good for Lao PDR and Mekong Delta during January-April 2016. Likewise, overall increase in dry season volume were observable between 4% and 12% at hydrological stations along the Mekong mainstream, as presented in Table 2. However, it is important to note that the increase was also partly attributed to regional climate condition (rainfall) and contribution from tributaries.   Meanwhile, discharges at key stations along the Mekong mainstream were also increased to a different extent, as shown in Table 3. Therefore, with a proper operation of the Lancang cascade dams, the discharge along the Mekong mainstream increased considerably in these two months of March-April, which were the period of minimum discharge for 1960-2009. More specifically, monitoring records in 2016 reveal a further increase in discharge even higher than the average of 2010-2015. This implies the emergency water supplement undoubtedly helps mitigate the prolonged meteorological and agricultural droughts in the Mekong Basin.   Table 4.
In December 2015, the water levels at most stations along the Lancang-Mekong River were generally lower than the average value of 1960-2009. However, from January to May 2016, the water levels at all stations were typically higher than the average of 1960-2009. As shown in Table 9, water level in March 2016 at the hydrological stations rose to an overall extent of 0.18-1.53 m.

Volume of the dry season (billion m 3 ) Deviation of volume between (billion m 3 )
1960     During the emergency water supplement from 10 March to 10 April 2016 (32 days), discharges at Jinghong stayed at about 2,000 m 3 /s, with an accumulated volume of 6.00 billion m 3 . Taking the travelling time into consideration, ratio at which the volume at Jinghong contributes to the total accumulated volume of the hydrological stations was calculated. The results are shown in Table 7. The total accumulated volume at Stung Treng is found to be 10.30 billion m 3 for the period between 27 March and 27 April (moving band of 32 days). Thus, the volume of the emergency water supplement in 2016 at Jinghong claims 58% of that at Stung Treng.  Overall monthly satellite rainfall over the Mekong Basin is presented in Figure 5 and ground rainfall observation at representative stations is depicted in Figure 6. It is understood that only small amount of rainfall was observed over the Mekong Basin. Additionally, since data and information of water releases from water infrastructures in the Mekong Basin and water withdrawal along the Mekong mainstream were not available at the time of this analysis, it is considered that the water supplement was a lumped sum of the emergency water supplement from China, lateral inflow and outflow of the Lancang-Mekong mainstream.
Analysis of the influential factors of flows of the Mekong mainstream was performed using hydrograph separation and hydrograph adjustment during the period of the emergency water supplement of March-May 2016. A simple hydrograph separation method 18 was applied by drawing a horizontal line between the beginning of rising limb of the hydrograph, which marked the arrival of the water supplement, and the end of falling limb of the hydrograph. This method was used to separate discharge of the water supplement from 'regular discharges'. On the other hand, the hydrograph at Jinghong was adjusted using discharge offset and travelling time to the hydrograph at Chiang Sean, Nong Khai and Stung Treng. These two methods were used for a cross-check in this analysis. It is found that discharge difference between these methods at all selected stations was relatively small and within the error margin of the accuracy of its rating curves 19 . Results of the analysis are illustrated in Figure 13 and summarised in Table 8.
Examining the hydrograph at Jinghong for March-May 2016 reveals that there were two distinct bands of the emergency water supplement from China: (1) steady flows of 2,200 m 3 /s from 10 March to 10 April 2016 and (2) steady flows of 1,500 m 3 /s from 21 April to 31 May 2016. These bands propagated along the Mekong mainstream as seen at Chiang Saen, Nong Khai and Stung Treng (Figure 13). The first band of 32 days was particularly investigated. Net contribution of the emergency water supplement at a given station was evaluated as a difference between average discharges of the moving band and the 'regular discharges' at the station ( Table 8). The net contribution of the emergency water supplement is found to be 1,024 m 3 /s (or 47% of total discharges during the water supplement) at Jinghong, 962 m 3 /s (or 44%) at Chiang Saen, 906 m 3 /s (or 38%) at Nong Khai, and 818 m 3 /s (or 22%) at Stung Treng.         Flow propagation along the Mekong mainstream was conducted using variation of daily water level and discharge, and sequence of its events, including the emergency water supplement. Variation of water level and discharge at Jinghong during the emergency water supplement was generally planned as follows: As shown in Figure 16, the discharge at Jinghong from March 9 to March 11 was gradually increased to 2,160 m 3 /s, with an obvious rise of water level from 535.76 m to 537.05 m. For an analysis on the travelling time, the beginning time of the emergency water supplement was thus deemed as 9 March 2016. For general flow conditions, characteristics of rapid fluctuation of daily observed water level and rated discharge of the Mekong mainstream for the dry season between Chiang Saen to Pakse follows the flow pattern observed in Chiang Saen. This is because the flow pattern is not typically perturbed by runoff generated from intense rainfall, which does not usually occur in Apr the basin during the dry season. The pattern becomes smoother and less variable as the Mekong River entering Cambodia, at Stung Treng since the flows from the Tonle Sap Lake dominated the flows in the Mekong River during this period. For particular dry season flow conditions of 2016, where flow volume stored in the Tonle Sap Lake was relatively low, patterns of variation of daily water level and discharge observed at Chiang Saen could be still seen at Tan Chau and Chau Doc (Figure 17 and Figure 18).

Variation of daily discharge at Chiang Saen
The emergency water supplement arrived at Chiang Saen on 11 March and started increasing till 14 March (3 days). As presented in Table 9 and depicted in Figure 17, Figure 18, Figure 19 and Figure  Due to flow conditions downstream Kratie are normally influenced by the outflow of the Tonle Sap Lake and tide of the sea, using variation of water level to mark arrival time of the emergency water supplement in this area is not obvious. It took 18 days for the emergency supplement water to travel a total length of 2,147 km from Jinghong to Kratie. Thus, this suggested a moving velocity of 1.4 m/s (or 5 km/h). It is assumed that the moving velocity was slowed down to 1 m/s in floodplain area. It would take around 4 days to travel 324 km between Kratie and Tan Chau. This is therefore believed that the emergency water supplement arrived to Tan

Salinity variation in the Mekong Delta
As adequate data and information on benefits of the emergency water supplement on reducing the meteorological agricultural drought affected area were not available at the time of this study, general observation of the benefits of easing the drought was compiled using various sources as presented in Annex A. Thus, analysis in this section was limited to salinity variation at in Soc Trang Province.
Soc Trang Province locates 231 km from Ho Chi Minh city, 60 km from Can Tho, close to Tra Vinh, Vinh Long, Hau Giang, Bac Lieu, with coastline of 72 km coastline and alluvial flat of 30,000 ha. It has an ocean climate and two seasons, rainy season from May to November, and dry season from December to May. The average temperature is between 26°C and 28°C. The economy is agriculture dominated, with cropland of 259,799 ha, among which 94% is rice field. The other cropland is covered by maize, Mung beans, jackfruit, coconut trees, green onion and garlic etc.  Figure 22 shows that there was a 4-day low salinity at early April at all the stations, though it was in rising tide period. The maximum salinity in April was between 2.2‰ and 6.4‰ less than that in March. The most prominent reduction occurred at An Lac Tay, from 8.0‰ in March to 2.1‰ in April (Table 10). The maximum salinity at Dai Ngai decreased from 13.8‰ in March to 7.4‰ in April. The maximum salinity decreased by 15% and 74%, and the minimum salinity decreased by 9% and 78% according to observation stations. Hence, the emergency water supplement from China played an important role in controlling seawater intrusion and reducing salinity, which would help protect ecosystem and environment in the Mekong Delta.

Conclusions
Recent meteorological and agricultural drought conditions over the Mekong Basin have worsened and triggered China to implement its emergency water supplement from its cascades dams in the Lancang River to the Mekong River by increasing the water discharge from Yunnan's Jinghong Reservoir. The emergency water supplement was implemented with a 'three-phase plan': (1) from 9 March to 10 April 2016, with an average daily discharge of no less than 2,000 m 3 /s; (2) from 11 April to 20 April 2016 with discharge of no less than 1,200 m 3 /s; and (3) from 21 April to 31 May 2016 with discharge of no less than 1,500 m 3 /s. The Mekong River Commission acknowledges this action by China, in which China has also stated that it implemented the water supplement at a challenging time, especially within the context where China itself was also suffering from drought, which has affected its household water supply and agricultural production.
The Joint Observation and Evaluation of the Emergency Water Supplement from China to the Mekong River were jointly and objectively conducted with the Mekong River Commission Secretariat and China. In the course of this study, besides regular hydrological data sharing in the flood season from China, additional hydrological data and information (daily and long term average of water level and discharge) for the dry season of 2016 from both sides were exchanged and used in the analyses of the Joint Report of this study. Equally, methodology of the analyses were jointly developed and adopted. Analyses were carried out and the results were exchanged, discussed and agreed. The contents of the Joint Report were also jointly developed. It is found that the emergency water supplement from China increased water level and discharge along the Mekong mainstream and decreased salinity intrusion in the Mekong Delta. The following are the key findings from this study:

Annex A -Observation of drought situation in the Mekong Countries from various sources
It is critically important to note that data and information in this annex is generally compiled from various sources including local and regional newspapers and governmental websites.
The meteorological and agricultural drought in 2016 is the direct result of the disruption of normal global weather patterns by one of the strongest El Niño events ever recorded. However, the longer than normal duration of this event had caused large rainfall deficits to build up in many places. Besides the lack of rain, the current El Niño conditions had also increased average daily temperatures across the region. The exceptionally hot conditions also caused high evapotranspiration and agricultural stress in crops already affected by low rainfall, resulting in further crop losses and poor yields, especially in rain-dependent agricultural areas. Overall conditions of drought in the Mekong countries are presented in Figure A1.

Cambodia
The increase of air temperature was causing animal lives and disease. The prolonged El Niño event had caused a significant increase of air temperature. It was found that 42.6°C set in Preah Vihea on 15 April 2016 has broken the national record. Such high heat with serious drought condition has caused many lives of cattle and animals including fish and put rural villagers at risk. In April 2016, more than 300 cows and buffaloes were found dead due to disease in Stung Treng Province and 70 tons of fish in Kampong Thom were lost due to high temperature with too low water level in the protected Chhmar River. Likewise, in early May, 50 black monkeys died from lack of water in a protected area in Battambang province 22 . . Moreover, the Ministry of Education of Cambodia also estimated that 2,500 schools out of a total of 10,000 were lack of water supply 24 , as depicted in Figure A2.

Lao PDR
Observation of overall rainfall, temperature and water stress indices presented in the main report reveals that Lao PDR started dry out in Southern part including Savannakhet, Saravane, Sekong, Champassak, and Attapeu provinces starting from January 2016. However, the situation became better in February. Lao PDR received most rain in the North than other parts of the LMB in April making the vegetation condition wetter.

Thailand
Following section are presented by the information from the webpage of the Royal Thai Government 25 .
The Royal Thai Government held a press briefing on 'Public-Private Alliance on Combating of Drought', and called on all sectors to conserve water as much as possible. The Royal Thai Government continues to undertake measures in preventing and tackling drought in a bid to ensure sufficient water supply throughout the dry season. The Thai Government also urged people to consume water wisely, and called on farmers to cooperate with the Government. The following are highlight of the press briefing: The Royal Irrigation Department stated that 2016 was another year in which drought situation was severe with very little amount of water budget. As a consequence of the great flood in The Geo-Informatics and Space Technology Development Agency (GISTDA) revealed that satellite images had been used for the analysis of Government agencies' water allocation plan. The GISTDA had also constantly monitored off-season rice growing situation, and found that in 2013-2014, up to 15 million rai of land were used for off-season rice cropping. The Thai Government, therefore, urged farmers to reduce growing off-season rice, and replace with drought-resistant crops, as well as take alternate jobs. This resulted in major reduction of offseason rice cropping, especially in the Chao Phraya River Basin where it went down to only around 3 million rai.
The Department of Water Resources added that compared to the previous years, drought situation in 2015 was not as critical as that in 2013, which was the most severe. Drought situation nationwide has not reached the critical stage except Nakhon Ratchasima, Buriram, and Surin which were just near the critical state. Nevertheless, the Thai Government had carried out operations under 2015-2026 Water Resources Management Strategy which comprises (1) Strategy on water consumption management; (2) Strategy on promotion of water security in production sector (agriculture/industry); (3) Strategy on flood management; (4) Strategy on water quality management; (5) Strategy on restoration and conservation of denuded watershed forests and prevention of land erosion; and (6) Strategy on general management. The Thai Government had also implemented district-based water consumption management in 928 districts across the country to determine exact amount and demand for water consumption in a bid to ensure efficiency in water management which led to national security, prosperity, and sustainability.
The Department of Agricultural Extension stated that the Cabinet had ordered Ministry of Agriculture and Cooperatives to provide assistance to farmers in the repetitive drought-affected areas, and to implement agricultural development and revenue generation schemes in a bid to alleviate adverse effects of the drought during 2014/2015 in 3,051 sub-districts (Tambons) of 58 provinces. Approximately 2.87 million farming households had been benefited with sufficient water sources during the dry season, more employment opportunities as labours, more agricultural facilities, reduction of production cost, and higher quality crops.
The Ministry of Agriculture and Cooperatives revealed that volume of total usable water in 481 large and medium-scale reservoirs was 16,870 million m 3 or 33%. Usable water volume in 4 major dams in the Chao Phraya River basin (Bhumibhol, Sirikit, Kwae Noi Bumrung Dan, and Pa Sak Jolasid) standed at 3,489 million m 3 or 19% in total; in water sources outside the irrigated areas nationwide 182.10 million m 3 or 52% of total capacity (as of 20 January 2016); and in 4,789 smaller reservoirs across the country 1,072.55 million m 3 , 59% of the total capacity (as of 21 January 2016).
The Thai Government is putting effort in implementing water management measures, which may result in insufficient water for agricultural purpose, especially for water-consuming crop growing and off-season rice cropping. The Government called on fellow farmers to understand and cooperate by following its advice, and put priority in public interests. The public were also urged to consume water wisely. Drought disaster this time was a challenge that has impacted all.
In addition to the above, Figure A3 illustrates recent situation of the drought in Thailand 26 .

Viet Nam
According to the Ministry of Agriculture and Rural Development, in the Mekong Delta saltwater had made its way to paddy fields two months sooner than in previous years as the rainy season began late last year but ended earlier than usual 27 .
Saltwater had travelled 90 kilometres inland in many parts of the Mekong Delta. Eleven out of 13 provinces in the Mekong Delta confirmed saltwater had impacted on agriculture and caused fresh water shortages 28 . In addition, Salinity measured in local shrimp farms, covering 130,000 ha of water surface, surpassed 35 ppt. Shrimp cannot live in water where salinity is higher than 40 ppt 29 .
Furthermore, reports from the provinces in the central highlands, south-central and Mekong Delta regions showed that more than 390,000 households had run short of fresh water as of 13 April 2016 and water shortages had wreaked havoc on over 240,000 ha of rice, more than 18,000 ha of other crops, 55,600 ha of orchards, and 100,000 ha of industrial trees. Around 4,600 ha of fisheries had been damaged 30 .
The Ministry of Agriculture and Rural Development said at a meeting on drought in Soc Trang Province (4 May 2016) that around 225,800 households in the Mekong Delta had run short of water for daily use by the end of April 2016, 70,800 households higher than in early March 2016. Additionally, as indicated in Figure A4,