Suriyan Saramul

Figure 5 - Coral reef and Seagrass Mapping in shallow waters of Ninh Hai – Ninh Thuan from VNREDSAT1 image base on BRI (leŌ ) and DII (right) methods

VNREDSAT1 is a very good resource for mapping seagrass beds and coral reefs on the Ninh Hai coast with a good overall accuracy of 72.38% and a Kappa coeffi cient of 0.56 using the DII method. This resource promises to be an effi - cient tool for the management and conservaƟ on of underwater ecosystems in Vietnam. The total area of seagrass beds along the coasts of My Hoa – Thai An Ninh Thuan waters comprises 275.22 (ha). The total area of coral reefs comprises 772.45ha and is located along the coast on the outer parts of the seagrass area. It needs more research to determine, how deep the VNREDSAT1 image could map the coral reefs on the Ninh Hai coast

We would like to thank the Vietnam NaƟ onal project of “Building oceanographic data from VNREDSAT-1 on Ninh Thuan - Binh Thuan province for sustainable marine economic development” Project code: VT/UD-07/14-15 (supervised by Dr. Nguyen Huu Huan), belonging to Vietnamese Space Technology Program, which provided the budget and the enƟ re survey data for this study.

Dai NH (2011): Seagrass in Vietnam. In: Seagrasses: resource status and trends in Indonesia, Japan, Malaysia, Thailand and Vietnam. Seizando-Shoten, Tokyo. Lyzenga DR (1981): Remote sensing of boƩ om refl ectance and water aƩ enuaƟ on parameters in shallow water using aircraŌ and Landsat data InternaƟ onal journal of remote sensing 2:71-82 doi:10.1080/01431168108948342. Nagelkerken I, van der Velde G, Gorissen MW, Meijer GJ, Van’t Hof T, den Hartog C (2000): Importance of Mangroves, Seagrass Beds and the Shallow Coral Reef as a Nursery for Important Coral Reef Fishes, Using a Visual Census Technique Estuarine, Coastal and Shelf Science 51:31-44 doi:hƩ p:// dx.doi.org/10.1006/ecss.2000.0617 Sagawa T, Boisnier E, Komatsu T, Mustapha KB, HaƩ our A, Kosaka N, Miyazaki S (2010): Using boƩ om surface refl ectance to map coastal marine areas: a new applicaƟ on method for Lyzenga’s model InternaƟ onal journal of remote sensing 31:3051-3064 doi:10.1080/01431160903154341. UNEP (2007): Coral Reefs DemonstraƟ on Sites in the South China Sea. UNEP/GEF/SCS Technical PublicaƟ on No.5. Yang D, Yang Y, Yang C, Zhao J, Sun Z (2011): DetecƟ on of seagrass in opƟ cal shallow water with Quickbird in the Xincun Bay, Hainan province, China Image Processing, IET 5:363-368 doi:10.1049/iet-ipr.2009.0392

Dr. Suriyan Saramul Lecturer, Chulalongkorn University, Thailand Wikipage: hƩ p://www.nf-pogo-alumni.org/~Suriyan+Saramul

Coastal environmental issues, such as oil spills and marine debris, are crucial problems that exist in many countries, and especially the developing countries. Such issues will directly impact marine organisms, for example, the oil slicks will block the sunlight to penetrate through the water, causing phytoplankton to be unable to photosynthesize. When an oil tanker crashes at sea, it is diffi cult to project the oil trajectories if there is no real Ɵ me surface current observaƟ on or operaƟ onal oceanography system in the area. Moreover, real Ɵ me current observaƟ on, such as the one using oceanographic buoys, is costly and requires a lot of buoys in order to observe the spaƟ al circulaƟ on paƩ ern of a large area. Remote sensing techniques are being developed to measure ocean surface properƟ es such as SST (sea surface temperature), SSH (sea surface height) and SSS (Sea Surface salinity). In addiƟ on, land-based remote sensing, e.g. high frequency radar (HF Radar), is developed to measure the surface current and wave characterisƟ cs of coastal areas. Surface currents obtained from HF Radar are esƟ mated based on the Doppler shiŌ of the radio signal that transmits from the staƟ on and receives the back scaƩ ering signal (see more detail about HF Radar in Joseph (2014)). This technique gives a good spaƟ al and temporal distribuƟ on paƩ ern of surface currents.

Since 2011, the Thai Meteorological Department (TMD) and Geo-InformaƟ cs and Space Technology Development Agency (GISTDA) has implemented the HF Radar system to measure surface currents and wave characterisƟ cs in Thailand waters.

In total, there are 16 staƟ ons, 6 from TMD and 12 from GISTDA (see Figure 1 for staƟ on names and locaƟ ons). In the inner Gulf of Thailand, there are lots of human acƟ viƟ es (e.g. tourism, fi shery, industry) that may cause a large environmental impact in coastal areas. The array of HF Radar systems will provide useful informaƟ on when there is an unexpected incident, such as the projecƟ on of oil spill, or search and rescue if something goes wrong. The circulaƟ on paƩ ern in the Gulf of Thailand is mainly controlled by monsoonal winds. Model study have shown that it shows changes in circulaƟ on paƩ ern: the clockwise and counter-clockwise circulaƟ on is formed during southwest and northeast monsoon, respecƟ vely (Buranapratheprat et al., 2009; Saramul and Ezer, 2014). In the current study, we found a seasonal circulaƟ on paƩ erns based on HF Radar system in the inner Gulf of Thailand (blue dashed box in Figure 1) will be described.

An hourly vector surface current data with the resoluƟ on of 2 km was retrieved from the GISTDA website between 24/01/2015 and 22/02/2015 for northeast monsoon and between 25/05/2015 and Figure 1 - HF Radar staƟ on’s locaƟ on in Thailand. Red and 23/06/2015 for southwest monsoon. Data points w contained less green dots represent that a staƟ on belongs to GISTDA or than 80% data were removed from the data set. The data was then TMD, respecƟ vely. The inner Gulf of Thailand is shown in averaged over the 30-day-period to obtain the residual current and the blue dashed box. to esƟ mate the average for the whole area to obtain the spaƟ ally averaged surface current. To invesƟ gate the infl uence of wind on the circulaƟ on paƩ ern, a re-analysis of the wind velocity retrieved from the EMCWF website will be analyzed (detail of EMCWF can be seen in hƩ p://apps.ecmwf.int) (Dee et al., 2012).

From Figure 2, variaƟ on of wind velocity in the inner Gulf of Thailand can be clearly seen. Northeasterly wind is mainly dominant during January 2015 and it changes to southerly wind during February 2015 (Figure 2a, shaded area). On the other hand, the southwest monsoonal wind is predominant during May to June 2015. The wind magnitude in both seasons is in the range of 2-5 m/s, which is quite calm. The wind is not only a factor that controls the circulaƟ on paƩ ern in the inner Gulf, but also Figure 2 - Wind velocity variaƟ on in the inner Gulf of Thailand during a) Jan-Feb the Ɵ des. Due to the locaƟ on and the shape of 2015 (northeast monsoon) and b) May-Jun 2015 (southwest monsoon). The the inner Gulf, the Ɵ dal current fl ows only in shaded area represents Ɵ me that HF Radar data are available. the north-south direcƟ on. It circulates counterclockwise as Kelvin wave propagaƟ on in the northern hemisphere. Water level at the Ko Sichang staƟ on is used as a representaƟ ve water level in the inner Gulf (see Figure 3). Generally, Ɵ des in the inner Gulf are mixed mainly semi-diurnal with a Ɵ dal range of 2-3 m. Tidal range during southwest monsoon (Figure 3b) seems to be larger than during northeast monsoon (Figure 3a).

The circulaƟ on paƩ ern in the inner Gulf derived from HF Radar is depicted in Figure 4. Figure 4bdisplays that the residual fl ow for the upper part shows strong movement towards the northeast. This causes the water level to pile up at the northeast corner of the inner Gulf as menƟ oned in Saramul and Ezer (2014). For the lower part, eastward and southeastward fl ows are present. These bring water out of the inner Gulf. Unlike the observaƟ ons of Buranapratheprat et al. (2009) and Saramul and Ezer (2014), the circulaƟ on paƩ ern during northeast monsoon is somewhat similar to the southwest monsoon (Figure 4a). This can be explained by the wind paƩ ern. Even during northeast monsoon, wind is mainly blowing from the south or southeast. In addiƟ on, Figure 3 - Water level at Ko Sichang staƟ on during a) January the lower part of the fl ow is more in the south. 24, 2015 to February 22, 2015 (northeast monsoon) and b) To invesƟ gate whether the fl ow is Ɵ dal driven, an area averaging May 25, 2015 to June 23, 2015 (southwest monsoon). was performed which is shown in Figure 4c and 4d for northeast and southwest monsoon, respecƟ vely. Figure 4c and 4d show Ɵ dal driven fl ows. The Ɵ dal fl ow in the inner Gulf is in north-south direcƟ on as menƟ oned above. The maximum and

minimum velociƟ es are coinciding with spring and neap Ɵ des as depicted in Figure 3 which are around day 12 and 26 for the maximum velocity and day 4 and 20 for the minimum velocity. In general, Ɵ dal velociƟ es in the inner Gulf are in the range of 10 – 35 cm/s.

Land-based remote sensing in the form of a high frequency radar system is installed and operated in Thailand both in the Gulf and the Andaman Sea. This system can be a reliable source of informaƟ on for surface current and wave characterisƟ cs in Thai waters (if needed). Surface current data retrieved from the HF Radar system was analyzed to invesƟ gate the surface fl ow in the inner Gulf of Thailand during northeast and southwest monsoons. The result confi rmed that the fl ow in the inner Gulf is Ɵ dal driven. Its propagaƟ on is aligned in the north-south direcƟ on with the maximum and minimum velociƟ es related to spring and neap Ɵ des. The infl uence of the monsoonal wind is demonstrated by performing Ɵ me averaging of January-February and May-June surface current. During southwest monsoon, it presents strong northeastward fl ow at the upper part, while the ouƞ low (eastward fl ow) can be found at the lower part. This causes the piling up of water at the northeast corner of the inner Gulf. Model studies predict a southwestward fl ow during northeast monsoon. However, this result shows a northeastward fl ow instead. This diff erence requires further study, which can be conducted by using modeling or acquiring a new data set during northeast monsoon to see the impact of the northeasterly wind.

Figure 4 - Monthly averaged surface velocity during a) northeast and b) southwest monsoon and spaƟ ally averaged surface velocity during c) northeast and d) southwest monsoon.

Buranapratheprat, A., Niemann, K.O., Yanagi, T., Matsumura, S., Sojisuporn, P., 2009. CirculaƟ on in the Upper Gulf of Thailand InvesƟ gated Using a ThreeDimensional Hydrodynamic Model. Burapha Science Journal. 14(1): 99-113 Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., et al., 2011. The ERA-Interim reanalysis: confi guraƟ on and performance of the data assimilaƟ on system. Quarterly Journal of the Royal Meteorological Society 137(656): 553-597. doi:10.1002/qj.828 Joseph, A., 2014. Remote Mapping of Sea Surface Currents Using HF Doppler Radar Networks. In Measuring Ocean Currents. Elsevier, Boston. 109- 137. doi:10.1016/B978-0-12-45990- 7.00004-1 Saramul, S.; Ezer, T., 2014. On the dynamics of low laƟ tude, wide and shallow coastal system: numerical simulaƟ ons of the Upper Gulf of Thailand. Ocean Dynamics. 64: 557-571. doi:10.1007/ s10236-014-0703-z

Photo by alumnus Lilian Krug