ISSS608 2017-18 T3 Assign ONG WAH JIE Observations & Insights

ISSS608 Assignment Vast Challenge Mini-Challenge 2 Ong Wah Jie

Background

Data Preparation

Visualization

Observations & Insights

Q1: Characterize the past and most recent situation with respect to chemical contamination in the Boonsong Lekagul waterways. Do you see any trends of possible interest in this investigation?

Data selection for chemical components

From the calendar plot, selected the chemicals which had records from 2011 – 2016. This was to get the average for the past 6 years. This resulted in 30 chemical components.

Using the Chemical (Readings) Worksheet, the trends of these 30 chemical components were looked at and the significant ones for 2013 – 2016 were selected for the trend analysis presented below.

Spike in 2014 readings

For the following chemicals, ammonium, anionic active surfactants, nitrites, sodium and total nitrogen, it was observed that there was a spike in the 2014 readings

For ammonium and total nitrogen, this was contributed by the locations Kohsoom and Tansanee.

For anionic active surfactants, this was contributed by Kohsoom (increasing from 2014 – 2016).

For nitrites and sodium, this was due to Tansanee.

Yearly readings trend

Readings by location

For the diagram above, the blue bold line refers to readings taken and the red dotted lines refer to 3 times standard deviation

Readings table

Materials	Locations	2014 Readings	Average reading per day
Ammonium	Kohsoom	3.884 mg/l	0.552 mg/l
Ammonium	Tansanee	5.803 mg/l	0.552 mg/l
Total Nitrogen	Kohsoom	5.81 mg/l	2.55 mg/l
Total Nitrogen	Tansanee	12.11 mg/l	2.55 mg/l
Anionic Active Surfactants	Kohsoom	0.1626 mg/l	0.335 mg/l
Nitrites	Tansanee	0.3908 mg/l	0.0404 mg/l
Sodium	Tansanee	233.3 mg/l	27.5 mg/l

Arsenic levels increasing since 2014

Levels of arsenic has been increasing since 2014, peaking in 2015 before slightly dipping in 2016 (remaining above the average number of readings from the second half of 2014 – 2016).

The locations which had above average readings for this period were Decha, Kannika, Sakda and Tansanee.

This was especially so for Tansanee which recorded a reading of 4.468 µg/l in 2015 which was 2.7 times the average reading per day of 1.618 µg/l.

Yearly readings trend

Spike in 2013 readings

Observed spike in 2013 for the following chemicals bicarbonates, calcium, chemical oxygen demand (Cr) and total dissolved salts.

For bicarbonates and calcium this was contributed by the locations Kohsoom and Tansanee, which remained above average for 2013-2016.

For chemical oxygen demand (Cr), this was due to Tansanee

For total dissolved salts, this was due to Tansanee which had elevated levels above the average from 2013-2016.

Readings by location

Readings table

Materials	Locations	2013 Readings	Average reading per day
Bicarbonates	Kohsoom	196.2 mg/l	191.7 mg/l
Bicarbonates	Tansanee	253.9 mg/l	191.7 mg/l
Calcium	Kohsoom	61.03 mg/l	55.17 mg/l
Calcium	Tansanee	89.57 mg/l	55.17 mg/l
chemical oxygen demand (Cr)	Tansanee	68.65 mg/l	25.95 mg/l
total dissolved salts	Tansanee	787.5 mg/l	383.9 mg/l

Copper Levels

Copper levels has remained above average since 2014 – 2016. This was mainly due to the locations Kannika and Sakda which had higher than average copper levels in the same period.

Yearly readings trend

Lead Levels

Lead has been above the average reading in 2016. These are due to the locations Busarakhan and Somchair which have higher lead levels in 2016.

Yearly readings trend

Q2a: What anomalies do you find in the waterway samples dataset? How do these affect your analysis of potential problems to the environment?

Different number of sampling stations at the different waterways

With reference to the diagram below, the waterways which are a part of the red and brown boxes, only have 1 water sensor to monitor the entire waterway. The waterway (box in blue) has 5 water sensor and the remaining waterway (not boxed) has 3 sensors.

With a limited number of water sensors (e.g. 1), it would not be enough to monitor the health of the entire waterway. This would lead us to draw wrong conclusion of its health based on just one point in the waterway.

Extremely high readings taken at location Tansanee

Extremely high readings (greater than 3 times standard deviation) have frequently been taken at the location Tansanee.

For the diagram above, the blue bold line refers to readings taken and the red dotted lines refer to 3 times standard deviation

The situation along this waterway would warrant further investigation. However, there is currently only one water sensor along this waterway.

Readings taken are not consistent

From the calendar chart below, it can be seen that there is no consistency in the number of samples taken for each chemical from year to year. For the same measure, the number of samples can differ from year to year (e.g water temperature 313 taken in 2015 and 649 taken in 2016).

There can be years in which totally no readings are taken at all (e.g total hardness in 2013).

Without a consistent approach, it would be hard to understand year-on-year trends of the chemical measure and the health of the waterway might be misrepresented in the year of interest.

Q2b: Is the Hydrology Department collecting sufficient data to understand the comprehensive situation across the Preserve?

They are not collecting sufficient data. The waterways should only have 1 water sensor should have more water sensors installed along the waterways, with at least one further upstream and another downstream in order to better understand the whole waterway.

In addition, the number of samples collected are not consistent with varying number of samples taken for each measure in different years and there could be times when no samples are collected for a measure at all.

Q2c: What changes would you propose to make in the sampling approach to best understand the situation?

A systematic method of sampling should be established. A fixed number of samples should be collected each month and at the same period and time each month (e.g 3 samples, 1st,15th and 30th of the month, at 9am each time).

This should be done across all the water sensor locations and the water sample collected should be tested for all the chemical components. Null should be entered as a record. There should not be duplicate records for the same location on the same day.

Q3a: After reviewing the data, do any of your findings cause particular concern for the Pipit or other wildlife?

Lead levels at Busarakhan and Somchair have been increasing since 2015.

(2016 readings of 6.06 µg/l and 7.14 µg/l vs average daily readings 2.63 µg/l) This is worrying as in animal systems, lead has been incriminated in a wide range of toxic effects.

For Busarakhan, Kannika which lies further downstream does not display elevated lead levels for 2015-2016. This could indicate the presence of a factory in Busarakhan which is discharging lead waste.

For Somchair, Sakda which lies further downstream does not display elevated lead levels for 2015-2016. This could indicate the presence of a factory in Somchair which is discharging lead waste.

For the diagram above, the blue bold lines refer to readings taken and the red dotted lines refer to 3 times standard deviation

Methylosmolene Levels

Methylosmolene (the toxic manufacturing chemical in the suspected dumping) levels are extremely high (2016 readings of 58.2µg/l vs average daily readings of 9.9 µg/l). This is the nearest water sensor to the waste dumping ground. Chai a downstream water sensor does not display such high readings.

It appears that the chemical dumped might have entered the water supply through seepage in the soil.

Elevated levels at Kohsoom

Apart from Methylosmolene mentioned earlier, Kohsoom has elevated readings for other chemicals as well from 2014-2016 (2016 readings anionic active surfactants 0.4546 mg/l, bicarbonate 194 mg/l and calcium 60.14 mg/l). This is in contrast to downstream Chai which does not display such elevated readings.

Q3b: Would you suggest any changes in the sampling strategy to better understand the waterways situation in the Preserve?

• To better understand the lead situation at Busarakhan and Somchair, water sensor readings should be taken on-site (i.e. close to the factory) and downstream of the factory to determine if it is the cause of the contamination
• To better understand the situation at Kohsoom, a water sensor should be placed at the waterway (box in red) near the waste location. This will allow us to determine if it is the waste location that is causing the elevated readings that is seen at Kohsoom.

• Currently, it seems that the water sensors are taken manually, with varying number of samples being taken on the same day. Automatic water sensors should be installed along these key locations with automated readings taken at determined intervals.