AT the end of 2025, 468 of Malaysia’s 672 monitored rivers were classed as clean, eighteen fewer than the year before.
A further 171 sat in Class III and 33 in Class IV. Behind those labels sits one of the most extensive environmental datasets the country produces, with more than 1,300 monitoring stations and records reaching back to 1978.
And yet, if you asked a hundred Malaysians on a bridge what Class III meant, almost none could tell you.
That gap between how much we measure and how little the public understands is the quiet failure at the centre of our water policy.
It is not a failure of science, and more science will not close it. It is to look back on the policy design.
The Water Quality Index combines six parameters into a single score from 0 to 100, mapped onto classes I to V.
A river scoring 62 clears the bar for “slightly polluted” while sitting in a class that requires extensive treatment before supply.
This is rigorous science, and almost perfectly designed to be ignored by anyone who is not a chemist. A resident beside that river cannot see, from the label, what it would take to make the water drinkable.
We pour the findings out faithfully, year after year, bagai mencurah air ke daun keladi. The water is poured, the leaf stays dry.
The information is public but not legible, and information that cannot be understood cannot drive a decision.
We have solved this before, and not with more data. Appliances carry energy-efficiency stars; nobody reads the kilowatt-hour tables, but everyone grasps five stars versus two.
Air-quality bands turn a hazy afternoon into a warning rather than a particulate count. Neither spread on charm.
Both exist because regulation obliged them to be displayed at the point of decision. Each was a policy innovation before it was a design one.
This is the logic behind the Eco-Heart Indicator, the symbol derived from the six water quality parameters that one of us developed, copyright-protected under Universiti Malaya.
It renders the classes the DOE already publishes as a heart for healthy, diamond (in relation with economic value) or failing in a symbol of tadpole facing downward, that anyone can read at a glance.
It measures nothing new. It carries an existing number the last mile into the public imagination. But an invention with no policy channel remains a demonstration.
Opening that channel takes three things, none requiring new science or new money. The first is a legibility mandate. Every release of water-quality data should carry a standardised, plain-language signal, issued alongside the technical figure rather than months later in a report almost nobody opens.
Standardisation is the harder half of that sentence. Water and land are largely state matters, so a signal adopted well in one state can mean nothing across the next border.
A signal has to mean the same thing in every state, or it teaches the public a language it cannot use twice. The second is to treat comprehension as a measurable objective.
Our indicators count stations, samples and reports, none of which asks whether a riverside community can state the condition of its own river.
Publishing the data in open, machine-readable form would let schools, journalists and state planners build their own translations, so that legibility depends on no single device, including ours.
The third is an adoption pathway. Malaysia funds invention generously and adopts it timidly, and a regulatory sandbox letting a river basin authority trial a public-signal tool, then report what changed in public understanding, would cost little and reveal much.
What fails to be understood upstream does not stay upstream. Every one of those parameters is eventually carried downstream and discharged into the sea.
A polluted river is not a contained problem. It is a delivery pipe to the estuaries, fisheries and coral systems the same public depends on.
Malaysia monitors marine water through a separate index, under separate agencies and separate laws, as though the water changed character at the river mouth.
It does not. The innovation required is institutional: a source-to-sea mandate under which river and coastal indicators are reported against each other.
Fragmented mandates are a policy choice, and policy choices can be redesigned. Our innovation system is heavily weighted toward measurement: new sensors, new models, new networks.
A system that rewards instruments but not interfaces will keep producing this result. Rivers do not deteriorate in laboratories.
They deteriorate through thousands of small decisions made by people who never saw the data, or could not read it.
Malaysia does not need to measure its rivers better. We need to make what we measure legible: to turn a class into a signal, and a river’s decline into a warning the public can read long before it reaches the coast.
Translation is not a cosmetic step that comes after the science. It is itself a policy function, and it needs the mandates and indicators any policy function requires.
Enforcement is where river governance is weakest, and enforcement depends on a public that understands why it matters.
We may have all the tools and technology we need, but are we ready to enforce the law and ensure it is adhered to?
The river has been saying this for years, and the sea has been repeating it. We simply never built the policy that would tell the public what it means.
Dr Azizi Abu Bakar is a Research Officer at the UM Sustainable Development Centre and Redza Zakaria is a Senior Lecturer at the Faculty of Law, Universiti Malaya.
The views expressed are solely of the author and do not necessarily reflect those of MMKtT.
- Focus Malaysia.
