ISSS608 2017-18 T3 Assign Priyadarsan Shankar Conclusion

VAST Challenge 2018: Suspense at the Wildlife Preserve
Mini Challenge 2 - Like a duck to water

Background

Methodology

Findings

Conclusion

Feedback

Back to main

VAST Challenge 2018 - Mini Challenge 2 answers

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?

- In the Past vs Present analysis, we've seen that many of the harmful chemicals are on an increasing trend like, Methylosmolene the harmful VOC solvent which has peaks of values around 300 times higher than what the past has seen in locations Kohsoom and Somchair.
- Even for harmful chemicals on a downward trend like Chlorodinine there are spikes in the current year 2016, causing a reason for worry.
- Nitrates and Sulphates once on a decreasing trend are now suddenly increasing in the post Kasios era of the Boonsong preserve.
- There are instances where the level of dissolved oxygen has fallen below the standard levels for aquatic life to survive.
- The yearly maximums are always on the increase even if the measure levels doesn't show an overall increasing trend.
- From the analysis, we can say that the chemical contamination in the Boonsong preserve waters has increased in the post Kasios era but can not comment with accuracy on the trend as the time series on the data is very irregularly spaced.

Q2 A - What anomalies do you find in the waterway samples dataset?

- There are 6 years(1999-04) of missing sample for the highly toxic chemicals of our interest across all locations.
- There’s a two year window that varies across streams where no measurements are taken of the chemicals of interest and sparse reading of the other measures. This window varies across locations and streams. In Boonsri and Kohsoom the locations close to the dumping site, it’s 2008-09, then it’s Busarakhan, Chai, Kannika, Achara, Somchair and Sakda along the streams 1 and 2 in years 2012-13. Interestingly the gaps co-insides with the factory setup timeline of Kasios inside the preserve. It could be due to construction work ongoing inside the forest hampering the water quality measurement activity.
- Tansanee, Decha and Achara have no readings measured before 2009.
- The number of unique chemicals being measured, and the total number of samples being taken in each year is on the drop since 2009.
- For measures like Atrazine, AOX, Arsenic and Sulphides there are gaps of around 3000 days between subsequent samples in the data. This impacts negatively any possible trend analysis.

Q2 B - How do these affect your analysis of potential problems to the environment?

- The samples being irregularly spaced across the time series is a roadblock to identifying trends and seasonality for the measures in motive to characterizing the contamination levels.
- Decha and Tansanee are lone locations of measure on stream 3 and 4, so we are left with water readings data with low past samples which might lead to incorrect interpretation of the contamination levels in those streams.
- The sampling strategy also being haphazard impacts the analysis, as some locations have 3 readings of a measure taken at each measure day and some have no readings of a measure across 500+ days.
- The number of records being comparatively sparse in the winter months can cause the analysis conclusions to be biased to the summer characteristics of water as it's properties are temperature dependant.

Q2 C - Is the Hydrology Department collecting sufficient data to understand the comprehensive situation across the Preserve?

- No, the Hydrology department is not collecting sufficient data to comprehensively understand the waterways.
- The samples are very widely/irregularly spaced across time, which is a probable consequence of the measuring process being manual.
- The number of measurement locations where the water sensor readings are taken are irregular across the streams, the first has 5, second has 3 and the third and fourth have 1 location each, so there's not enough data on each waterway to characterize clearly the water quality scenario across the preserve.
- Vital measures like pH(Acidity) and Turbidity(Cloudiness or haziness in water) of the water that are universal standards to define the quality of water are never measured, this also resulted in an incomplete WQI(Water Quality Index) calculation in our analysis.

Q2 D, Q3 B - What changes would you propose to make in the sampling approach/strategy to best understand the situation?

A new sampling strategy I would like to propose is:
i) Install water way sensors at atleast 3 locations in each stream.
ii) Include sensors to measure all vital measures including the pH and the Turbidity of water, this way the water WQI can be measured accurately.
iii) Install atleast one sensor location at the closest proximity where water is present from the dumping site, this way any increase in contamination levels due to dumping can be characterized to it accurately.
iv) Install atleast one sensor location at the water waste outlet's connecting point to the waterways of each of the 5 furniture factories present in the southern zone of the preserve. This way any increase in contamination levels due to factory waste can be characterized to it accurately and can prevent companies like Kasios from falsely claiming themselves innocent for being the source of some chemicals.
v) Install atleast one sensor location at the closest proximity where water is present from the dumping site.
vi) Automate the measurement process and synchronize the measurement times across all locations in the preserve, so the time series data of the chemicals is evenly spaced out in time and complete enough to identify trends and spikes and accurately characterize them.
vii) Measure atleast 2 readings of each water contaminant measure in a day, preferably at times of factory operation and after the production day end when the waste is let out.
This will ensure accurate characterization of the contamination levels of all the waterways using the delta values of each measure between two consecutive measuring stations and how the levels vary across time and what is the root cause to be attributed for the change.

Q3 A - After reviewing the data, do any of your findings cause particular concern for the Pipit or other wildlife?

- The chemical contamination levels are on the increase in the waterways since 2009.
- The chemical contamination levels are highest in 2016 in the locations Kohsoom(Nearest to the known dumping site) and Somchair across all chemicals classified as HIGH MEDIUM and LOW toxicity.
- The water quality index has been stable until 2009 throughout the preserve and is falling at many locations since then. Busarakhan, Somchair and Chai have hit poor quality and Kohsoom is bordering the lower bound of medium quality at the end of 2016.
- An interesting find is the contamination level at Somchair is consistently high as compared to the water in stream 1 which is along the dumping site. So I believe there is possibility of the presence of a second illegal dumping site in the vicinity of Somchair.
- Highly toxic chemicals for birds such as Methylosmolene, Chlorodinine, Lead, Copper, Iron, Mercury, Nitrates, Nitrites have never seen in history peaks in the current period.
- Such rapidly decreasing water quality is harmful to any aquatic life of the Boonsong waterways and has a high chance to harm all the animals and birds including the Rose crested blue pipit living in that environment and using the water for drinking or finding food in fish.