Ambient Noise Mapping across Zone Types: A Leq Study Judged against the Noise Rules, 2000

What this study does. Noise is the pollutant everyone hears and nobody measures: hospitals wait beside honking traffic, 'silence zones' outside schools are nothing of the kind, and everyone knows the local junction is loud — but by how much, against which limit, at what hour? India actually answers this on paper: the Noise Pollution (Regulation and Control) Rules, 2000 set ambient standards per zone — residential 55 dB(A) day / 45 night, commercial 65/55, industrial 75/70, and silence zones 50/40 — with day defined as 6 a.m.–10 p.m. This project checks where a real Bengaluru neighbourhood stands against those numbers.

Twelve fixed points — three each in silence-zone (hospital and school frontage), residential, commercial and major-traffic locations — are measured with a borrowed Class 2 sound level meter on a tripod, following the CPCB 15-minute Leq protocol. Each point is sampled in three daily bands (morning, evening, night) on a weekday and a Sunday, the meter calibrated at 94.0 dB before and after every session with drift logged. The meter records Leq plus statistical levels L10, L50 and L90, so each site yields a shape: steady droning streets show a narrow L10–L90 spread, horn-dominated junctions a wide one.

Analysis is honest about scale: per-zone mean ± SD Leq by band, exceedance tables against the Rules' limits, day–night differences, and a hand-drawn noise map. No claim of citywide monitoring is made — a one-observer, 12-point survey is exactly what it is, and the report says so.

Every neighbourhood complains about noise; almost none can quote its own measured numbers. The Noise Pollution (Regulation and Control) Rules, 2000 define legal day and night limits for silence, residential, commercial and industrial zones, but without local measurements nobody — resident, school, hospital or ward office — knows which junctions exceed which limits, at which hours, or by how much. Complaints therefore remain anecdotes that no authority has to act on.

At the student level, noise 'studies' usually mean a few phone-app readings taken at random times with no calibration and no defined geometry, then compared loosely with limits that may not even apply to the location's zone. Such numbers cannot be defended, because nothing about their provenance is controlled or recorded.

This project therefore measures ambient noise at twelve fixed, zone-classified points with a calibrated Class 2 meter under a declared protocol — 15-minute Leq windows, three daily bands, weekday and Sunday, percentile levels included — and delivers the missing local evidence: a noise map and an exceedance table against the correct legal limits, with the survey's scale and limits stated honestly.

Abstract

This single-member monitoring study maps ambient noise across twelve fixed points in three zone-legal categories (silence, residential, commercial) plus a major-traffic corridor in one Bengaluru neighbourhood. A Class 2 sound level meter on a tripod records 15-minute Leq and percentile levels (L10, L50, L90) in three daily bands on a weekday and a Sunday per point, with 94.0 dB calibration before and after each session. Results are summarised per zone and band, compared against the Noise Pollution (Regulation and Control) Rules, 2000 day/night limits, and drawn as a neighbourhood noise map with exceedances marked.

Introduction

India's noise standards are unusually clear: the 2000 Rules name four zone categories with separate day and night limits, and they protect silence zones around hospitals, schools and courts. Enforcement aside, the first missing ingredient is local data — residents argue about noise with anecdotes because nobody in the neighbourhood has ever put a calibrated meter at the school gate at 9 p.m.

Existing knowledge and the gap

CPCB runs a national ambient noise monitoring network in major cities and publishes its protocol; the Rules define the legal limits precisely. What student projects usually produce instead is a scatter of random readings with a phone app: no calibration, no fixed measurement geometry, no defined duration, and no legal zone classification, so the numbers cannot be compared with the standards at all. The gap this project fills is a defensible neighbourhood survey — declared points, calibrated instrument, CPCB-style 15-minute Leq windows, percentile levels for texture, and a verdict per point against the correct day/night limit for its zone.

Proposed study

1. Classify twelve locations into silence, residential, commercial and traffic categories per the Rules' definitions; fix their coordinates and photograph each geometry.
2. Borrow a Class 2 meter and acoustic calibrator; rehearse the 15-minute protocol and record the meter's specifications.
3. Measure every point in morning, evening and night bands on one weekday and one Sunday, logging Leq, L10, L50, L90 and any anomalous events.
4. Verify calibration before and after each session; reject any session with drift beyond 1 dB and re-measure.
5. Analyse per-zone summaries, exceedance against the correct day/night limit, day–night gaps and percentile spreads.
6. Draw the neighbourhood noise map and write the plain-language verdict for residents and the school/hospital administration.

Feasibility

• Technical: a borrowed Class 2 meter and calibrator, a tripod and a fixed protocol are all an undergraduate can handle competently; the statistics are simple summaries, honestly labelled.
• Economic: no purchase is needed if the college or a local consultant lends the meter; transport is the only cost.
• Operational: night measurements need a safety plan — an escort and lit public footpaths only, never private premises. Cooperation from the school or hospital front office makes silence-zone sampling smooth. The known risks — wind and passing vehicles biasing single windows, and repositioned meters between visits — are handled with the wind shield, marked tripod spots and the drift rule.

In scope

• Twelve fixed points in one Bengaluru neighbourhood, three daily bands, one weekday and one Sunday round per point.
• Class 2 meter metrics: Leq, L10, L50, L90; calibration drift records; zone-legal classification per point.
• Descriptive statistics, exceedance analysis against the Noise Rules, 2000, day–night comparison and a noise map.

Out of scope

• Continuous 24-hour monitoring stations and Ldn/Lden indices — a fixed-installation exercise beyond a borrowed meter.
• Source attribution (traffic counts, horn-type surveys) beyond qualitative notes at each window.
• Occupational or indoor noise, construction-site measurements and loudspeaker-event chasing — each is a regulated study of its own.
• Annoyance or health surveys of residents — the design measures sound, not people.

Research design: a fixed-point observational monitoring study with a declared instrument protocol, timed sampling bands and a standards-based verdict. Individual study, 12 weeks at 8 h/week.

Phase 1 — Classification and geometry (Weeks 1–2). Twelve points are chosen and zone-classified strictly by the Rules' definitions: three within 100 m of hospital/school frontage (silence zone), three on interior residential lanes, three on commercial stretches, three on the arterial road edge. Each point's tripod spot is marked and photographed (distance from facades, height 1.2–1.5 m), and a map sheet is drawn.

Phase 2 — Instrument and rehearsal (Week 3). The borrowed Class 2 meter's specifications are copied into the report; wind shield fitted; calibrator checked. Rehearsal on two spare points fixes the routine: mount, calibrate (94.0 dB), record 15-minute window, note events, calibrate again, log drift.

Phase 3 — Field campaign (Weeks 4–9). Each point is measured in three bands — morning 08:00–10:00, evening 18:00–20:00, night 22:00–24:00 — on one weekday and one Sunday, six windows per point, seventy-two windows total. Night rounds run with an escort on lit public footpaths. Per window the log captures Leq, L10, L50, L90, calibrator drift, weather, and one-line source notes (horn bursts, temple loudspeaker, generator). A session whose pre/post calibration drifts beyond 1 dB is rejected and re-measured.

Phase 4 — Analysis (Weeks 10–11). Per point: Leq by band; day value as the mean of morning and evening windows per CPCB convention; night from the night window. Per zone: mean ± SD by band; weekday vs Sunday comparison; percentile spread (L10–L90) as texture. Exceedance table: every window against the zone's day or night limit from the Rules' schedule. Neighbourhood map hand-drawn with per-point values and exceedance marks. The survey's scale is stated in every summary: twelve points, two days, one season.

Timeline: W1–2 classification and geometry; W3 instrument rehearsal; W4–9 campaign; W10–11 analysis; W12 report, map, slides, viva.

The survey is a calibration-gated measurement pipeline: no reading enters the dataset unless its session passed the drift check, and every reading carries its zone-legal context. The workflow below is the report's methods spine.

flowchart TD
  A["Classify 12 points: silence, residential, commercial, traffic"] --> B["Mark tripod spots; photograph geometry"]
  B --> C["Borrow Class 2 meter + calibrator; rehearse protocol"]
  C --> D["Calibrate at 94.0 dB before window"]
  D --> E["Record 15-min Leq + L10/L50/L90; log sources and weather"]
  E --> F["Calibrate again; drift <= 1 dB?"]
  F -->|Yes| G["Accept window: morning, evening, night x weekday, Sunday"]
  F -->|No| H["Reject session; re-measure point"]
  G --> I["Per-zone summaries; percentile spreads"]
  I --> J["Exceedance table vs Noise Rules 2000 day/night limits"]
  J --> K["Neighbourhood noise map"]
  K --> L["Plain-language verdict for residents and institutions"]

Data organisation

One row per window: point code, zone class, date, band, weekday/Sunday, Leq, L10, L50, L90, calibration drift, weather and source notes. Zone summaries, exceedance flags and the map values all regenerate from this table; two windows are hand-averaged in the appendix. Nothing is averaged before the exceedance table — the Rules judge each band against its own limit.

Core calculations (abridged)

Day Leq   = mean of morning and evening window Leq values
Night Leq = night-window Leq
Exceedance = window Leq - zone day/night limit (dB), positive = breach
Texture   = L10 - L90 (percentile spread at the same window)

Results tables to fill (no numbers are pre-filled)

• Seasonal rounds: festival-season and monsoon windows to capture the events that dominate Indian noise complaints.
• Continuous monitoring: a 24-hour station at the two worst points for proper Ldn computation and diurnal curves.
• Source attribution: pairing the meter with a traffic-count log and horn-event tally to explain the exceedances.
• Indoor transmission: façade-versus-indoor pairs at the school and hospital to quantify how much of the outdoor breach reaches occupants.
• Ward expansion: a students'-group replication across more wards, turning the project into a standing monitoring cell for the ward office.
• A complaint-pack template: the exceedance table reformatted as the annexure a residents' association can attach to a formal complaint under the Rules.