Open data · CC BY 4.0

Take the data with you

Counting the rows…

Open data · CC BY 4.0

Take the data with you

Every route, segment, audit, report, cleanup and episode on the map, as CSV and GeoJSON. Check it, reuse it, and credit Cleartrail Munnar.

LicenceCC BY 4.0

Credit “Cleartrail Munnar”

Column dictionary Audit method v1

Downloads below leave demo rows out (?demo=exclude). The counts leave them out too.

The datasets

Each one opens in a spreadsheet as CSV, or on a map as GeoJSON

LinesLast changeFixed, cut once

4routes

Routes

The audited routes, with their length and segment count.

SourceRoutes (reference data)

Columns

route_id
code
from_name
to_name
from_name_ml
to_name_ml

All 9 columns in the dictionary

CSV GeoJSON

LinesLast changeFixed, cut once

73segments

Segments

Every fixed segment, with its place along its route and whether it is beside water.

SourceSegments (reference data)

Columns

segment_id
route_id
route_code
seq
start_m
end_m

All 13 columns in the dictionary

Column dictionary

What every column in every file means

Every date is a calendar date in Munnar time (Asia/Kolkata), written YYYY-MM-DD. No time of day is published. A list is joined with | in a CSV and is an array in GeoJSON. An empty cell means there is no value. The episodes GeoJSON also gives each feature a steps array, in date order. Each step has a step code: reported (public reports on the segment; report_count, on, last_on), verified (the first audit; audit_id, severity, audit_count), cleanup_requested (request_count, on), cleaned (cleanup_id, last_cleanup_id, cleanup_count, on, last_on), follow_up (an audit after a cleanup; audit_id, severity, cleanup_id of the cleanup it follows, held true when it was in the Low band, decides true for the first audit after that cleanup, early true when it came before the 14 Munnar days were up) and verified_clean (the follow-up that closed the episode; audit_id, on).

Routes

9 columns

Column	Type	Meaning
route_id	Text	The route's identifier.
code	Text	The route code that starts every segment identifier on it, such as MN-MAT.
from_name	Text	Where the route starts.
to_name	Text	Where the route ends.
from_name_ml	Text	Where the route starts, in Malayalam.
to_name_ml	Text	Where the route ends, in Malayalam.
length_m	Whole number	The route's length in metres.
segment_count	Whole number	How many fixed segments the route is cut into.
demo	True or false	Always false: routes are reference data, not demo data.

Segments

13 columns

Column	Type	Meaning
segment_id	Text	The segment's identifier: the route code and a number, such as MN-MAT-007. Never changed.
route_id	Text	The identifier of the route the segment is on.
route_code	Text	The code of that route.
seq	Whole number	The segment's number along its route, from 1.
start_m	Whole number	Metres from the start of the route to the start of the segment.
end_m	Whole number	Metres from the start of the route to the end of the segment.
length_m	Whole number	The segment's length in metres. Most are 500; a route's last segment can be shorter.

Licence and method

Use it freely, and say where it came from

Use it freely, credit it

Licensed CC BY 4.0. Credit “Cleartrail Munnar”.

Data: Cleartrail Munnar, CC BY 4.0

CSV files carry their licence in licence.txt, because a CSV cannot hold a comment. Every GeoJSON file states the licence in its own properties, and every download names it in its Link header.

Read licence.txt

Audit method v1

Every audit row names the method version that scored it.

1.6weight of a glass item, every other item is 1
+120weighted items per km for bulk dumping
+40weighted items per km for an overflowing bin
+10index points beside water
4weighted items per km make 1 index point
100the highest the index goes

Read the method on About

near_water	True or false	Whether the segment runs beside water. An audit there adds points to its index.

start_lon	Number	Longitude of the segment's first point, WGS 84.

start_lat	Number	Latitude of the segment's first point, WGS 84.

end_lon	Number	Longitude of the segment's last point, WGS 84.

end_lat	Number	Latitude of the segment's last point, WGS 84.

demo	True or false	Always false: segments are reference data, not demo data.

Audits

20 columns

Column	Type	Meaning
audit_id	Text	The audit's identifier.
segment_id	Text	The segment that was counted.
counted_on	Date	The day the auditor counted the segment.
pet	Whole number	Plastic bottles counted.
wrappers	Whole number	Wrappers and sachets counted.
cups	Whole number	Cups and plates counted.
glass	Whole number	Glass bottles and pieces counted. Glass counts for more in the index.
other	Whole number	Other litter items counted.
bulk_dumping	True or false	Whether bulk dumping was present on the segment.
overflowing_bin	True or false	Whether an overflowing bin was present on the segment.
near_water	True or false	Whether the segment is beside water, as its plan says.
segment_length_m	Whole number	The segment's length in metres when it was counted.
method_version	Whole number	The version of the audit method that scored this row. See the method section of About.
items	Whole number	All items counted, unweighted.
items_per_km	Whole number	The unweighted items scaled to a kilometre.
mli	Whole number	The Munnar Litter Index, 0 to 100, higher is worse. Only audits have one.
severity	Text	The band of the index, as one code: low (0 to 24), moderate (25 to 49), high (50 to 69) or severe (70 to 100).
note	Text	The auditor's note, as published. Empty when there is none.
photo_count	Whole number	How many public photographs the audit has.
demo	True or false	True when the row is demo data, made up to show how Cleartrail works.

Reports

10 columns

Column	Type	Meaning
report_id	Text	The report's identifier.
category	Text	What the reporter saw, as one code: litter (litter on the ground), dumping (dumped waste), overflowing_bin (a bin that is overflowing), plastic_in_water (plastic in a stream or lake) or other (anything else).
amount	Text	The reporter's rough impression, as one code: a_little (a little), a_bagful (about a bagful) or a_lot (a lot). Empty when not given. Never a number and never a severity.
note	Text	The reporter's note, as published. Empty when there is none.
lon	Number	Longitude of the pin, WGS 84, as stored.
lat	Number	Latitude of the pin, WGS 84, as stored.
segment_id	Text	The segment within 100 m of the pin. Empty when none is.
reported_on	Date	The day the report was sent.
photo_count	Whole number	How many public photographs the report has.
demo	True or false	True when the row is demo data, made up to show how Cleartrail works.

Cleanups

10 columns

Column	Type	Meaning
cleanup_id	Text	The cleanup's identifier.
segment_id	Text	The segment that was cleaned.
done_on	Date	The day the cleanup happened, in Munnar.
group_name	Text	The group that did it: the name it gave, or the partner it was credited to. Empty when that partner is not named publicly.
people	Whole number	How many people took part.
bags	Whole number	How many bags were filled.
note	Text	The note on the cleanup, as published. Empty when there is none.
before_photo_urls	List	Addresses of the public before photographs, joined with \| in a CSV.
after_photo_urls	List	Addresses of the public after photographs, joined with \| in a CSV.
demo	True or false	True when the row is demo data, made up to show how Cleartrail works.

Episodes

24 columns

Column	Type	Meaning
episode_id	Text	The episode's identifier, made from the event that opened it. It is stable while that event stays public.
segment_id	Text	The segment the episode is on.
status	Text	One code: open (still going) or closed (a follow-up audit counted low).
opened_on	Date	The day of the earliest thing in the episode.
closed_on	Date	The day the episode closed. Empty while it is open.
opened_by_kind	Text	What opened it, as one code: report (a public report), audit (an audit) or cleanup (a recorded cleanup).
opened_by_id	Text	The identifier of the report, audit or cleanup that opened it.
report_count	Whole number	How many public reports joined the episode.
first_report_on	Date	The day the first of them was sent.
last_report_on	Date	The day the latest of them was sent.
first_audit_id	Text	The first audit counted before any cleanup.
first_audit_on	Date	The day that audit was counted.
first_audit_severity	Text	That audit's band, as one code: low, moderate, high or severe (as in the severity column of audits). Reports and audits are never blended: this comes from the audit alone.
audit_count	Whole number	How many audits were counted before any cleanup.
cleanup_request_count	Whole number	How many people asked for a cleanup.
first_cleanup_request_on	Date	The day the first request was made.
cleanup_count	Whole number	How many cleanups were recorded in the episode.
first_cleanup_on	Date	The day the first cleanup happened.
last_cleanup_on	Date	The day the latest cleanup happened. The follow-up audit falls due from this day on.
follow_up_audit_ids	List	Audits counted after a cleanup, joined with \| in a CSV.
follow_up_held	True or false	Whether the audit that decides the cleanup counted low (true, the cleanup held) or not (false). Empty until one is counted.
verified_clean_on	Date	The day a follow-up audit counted low, closing the episode.
verified_clean_audit_id	Text	That follow-up audit.
demo	True or false	True when the episode is made of demo data.