4. India – Village – Methodologies - Fluoride
4.3.
Participatory fluoride surveys
This section describes a rapid, reliable method of determining fluoride concentrations
in groundwater sources that is compatible with a participatory planning process.
4.3.10. Background
The initial objective of the fluoride surveys undertaken as part of the WHiRL project
was to establish whether domestic water sources were safe. Although official fluoride
statistics suggested little case for concern, observations in the study villages in
Kalyandurg, Anantapur suggested severe health problems were widespread including
dental and skeletal fluorosis. The wider aim was to identify possible alternative
options for water supply in villages where people were actually consuming unsafe
water. Background on the occurrence of fluoride in drinking water is given in Box
4.3.2.
4.3.11. Methodology
Several methodologies are available to monitor water quality issues like fluoride in
groundwater. These include:
 laboratory analysis: which is reliable, but slow, and especially leads to long
delays in giving feedback on results to communities
 portable field kits: these can be reliable if the kits are used properly, and make it
possible to feedback results quickly, as well as sampling a large number of
sources across areas quickly
 indicator test strips: these can make analysis very participatory, but were not
sufficiently reliable
In the WHiRL study, a HACH portable colorimeter was used (HACH pocket
colorimeter Type I, Hach Company, Loveland, Colorado). This meter was selected
because of the lower cost compared to ion selective electrodes while still giving the
level of accuracy needed.
Box 4.3.2 Fluoride in drinking water: an overview
Fluorine is a highly-reactive element that is found in a number of commonly-occurring minerals of
which fluorspar, cryolite and fluorapatite are the most widespread. Fluoride occurs naturally in some
groundwater sources and in a range of different food items that have been grown or produced using
water with high fluoride content. In some countries, fluoride is added to drinking water and
toothpastes in the belief that low levels of fluoride consumption reduce tooth decay. Fluorine and
various fluorine compounds are often released into the environment by industrial processes (e.g.
production of aluminium, phosphate fertiliser, bricks, tiles and ceramics).
Experts disagree as to whether any health benefits result from low concentrations of fluorides in
drinking water, however, it is agreed that ingestion of large amounts of fluoride, whether via water or
food, can cause serious health problems for humans and animals. These range from discoloured teeth
(i.e. dental fluorosis) to aching joints, brittle bones, stunted growth and deformed limbs (i.e. skeletal
fluorosis).
The current WHO permissible limit for the fluoride concentration of drinking water supplies is 1.5 mg/l
(or ppm) with the added recommendation that “climatic conditions, volume of water consumed and
intake from other sources should be considered when setting national standards”. In 1993, the Bureau
of Indian Standards set a maximum permissible fluoride concentration of 1.0 mg/l, although
concentrations of up to 1.5 mg/l are considered to be acceptable in the absence of an alternative safer
source.
Training and testing
Samples were analysed using the manufacturers prescribed ‘pipette’ method. This
requires use of standard fluoride solutions to calibrate the meter, and adding reagents
to water samples, so a careful and experienced user is vital. Training to check the
procedure recommended by the manufacturer, and to check results for consistency (by
repeating tests) and reliability (by comparison with results from laboratories) should
not be overlooked. Our testing period showed that we would also need to dilute
samples with distilled water by a ratio of 2:1 in order to remain within the
measurement limits of the equipment.
Sampling
In the vicinity of the four study villages (Battuvanipalli, Pathacheruvu, Mairevu and
Obulapuram), water samples were collected from each well within approximately 2
km of the village centre. This included the main domestic water supply sources for the
villages, but also other sources like irrigation wells in order to provide information on
variations in fluoride levels around the villages and areas around the villages where
better quality water might be available. A major factor in determining the timing of
sampling was that we had to visit irrigation wells when power was available. Basic
information on the type and use made of each well was also collected, and using a
GPS system, the exact location was also logged. Occasional measurements were also
made of surface water in tanks and check dams that are used to recharge groundwater.
Analysis
Samples were analysed using the colorimeter in a temporary laboratory, a clean room
with space to layout the samples and equipment, usually on the same day as sample
collection, either within the village or nearby. After solving problems revealed by the
initial cross-checks (poor results due to procedure, dilution etc) it was possible to
complete 10-15 tests in an hour.
Calibrations of the portable meter were regularly checked against standard fluoride
solutions, and samples sent to other laboratories for cross-comparison. The results of
analysis using the portable colorimeter compared favourably with results from
samples sent to water quality laboratories in nearby Kurnool and Anantapur. In fact,
we believed the above method to be more reliable than some of the laboratory
analyses: especially those determined using a visual assessment method. However,
regular checking of different batches of samples over time, tests by different
operators, and with different laboratories is important to reveal possible errors. It is
useful to plot the field and laboratory results together in a scatter graph and to look for
differences and patterns.
In order to further triangulate the findings of the surveys, results were compared with
results from a rapid assessment in 2001. In an earlier study (3.3 India-DistrictMethodologies-QPA) in Kalyandurg, water users had been asked in late 2001 to
assess the quality of water from domestic sources on the basis of taste and whether
they felt the water contained fluoride. Despite the differences in timing of sampling
there was good agreement between these results and the results of fluoride analysis.
For the 13 sources covered by both surveys, the fluoride values for the three sources
ranked by users as having fluoride problems had contents of 1.47, 1.74 and 2.94 mg/l.
For the other 10 sources not considered to have fluoride problems by users, results
were all in the range 1.00 – 1.44 mg/l.
Data analysis and discussion
The WHiRL project had wider objectives to identify solutions to fluoride
contamination of drinking water following a participatory planning process linked to
implementation of watershed development in the study villages. Participation of the
community in collecting water samples was encouraged, so that people understood the
objectives of the survey and felt some level of ownership of the findings. Equally
important, results were then rapidly reported back to the community and well owners.
Wells where fluoride levels exceeded the safe limit of 1.5 mg/l (Table 4.3.3) were
also marked with a red cross to indicate that the water was considered unsafe for
drinking.
After the initial survey, fluoride levels were again routinely measured, roughly once
every two months, for the main domestic water sources and within areas (using the
irrigation wells to collect groundwater samples) where it seemed possible to locate
new safe drinking water sources.
Table 4.3.3 Effects of fluoride in drinking water
Fluoride in
drinking water
mg/l
Below 1.0
1.0-1.5
1.5-3.0
3.0-10.0
Effects
Safe
Marginal
High risk of dental fluorosis
Leads to skeletal fluorosis with
adverse changes in bones
Crippling skeletal fluorosis
More than 10.0
4.3.12. Example results
The initial surveys revealed significant differences in fluoride levels both within and
between the four villages studied (Table 4.3.4). Flouride levels were high in all
villages, but on average are within safe drinking water limits (1.5 mg/l) in Manirevu
and Obulapuram. Some sources in Battuvanipalli and Pathacheruvu by contrast had
levels around 4 mg/l, and since health problems were also reported in these villages,
later surveys focused on these two settlements.
Table 4.3.4 Flouride levels in groundwater, October-November 2002
Habitation
No of
samples
Battuvanipalli
Manirevu
Obulapuram
Pathacheruvu
68
60
71
72
Minimum
fluoride content
(mg/l)
1.04
0.49
0.71
0.88
Maximum
Average
Std dev
fluoride content fluoride content
(mg/l)
(mg/l)
4.28
2.6
0.86
2.42
1.3
0.39
1.82
1.4
0.23
3.96
2.1
0.56
Of the 26 mainly ‘domestic’ public water sources in the four villages, 10 had fluoride
values above the recommended safe limit of 1.5 mg/l. The situation was most severe
in Battuvanipalli where supply for the main piped system had high fluoride levels
(2.94 mg/l). Other alternative sources here had other water quality problems: the
water was brackish. Fluoride levels were also found to be a major problem in
Pathacheruvu (the analysis here also includes some sources in nearby Golla village)
where 5 out of 6 sources failed. None of sources in the hamlet of Pathacheruvu itself
were safe, and the only safe source is located some distance away in Golla. The new
source where the hamlet was to be resettled also had high levels (1.9 mg/l). In
Manirevu and Obulapuram one handpump in each village exceeded the safe limits,
but other safe sources were available.
No statistical significant difference was found between fluoride levels for different
well types (e.g. borewell, dugwell, dug-cum-borewell), and furthermore there was no
statistically valid difference between values from drinking (average 1.6 mg/l) and
irrigation sources (1.92 mg/l).
Routine monitoring of fluoride levels showed that values fluctuated significantly
during the year, and especially during drought periods. The extended period of
drought (and falling groundwater levels) throughout 2003 and early 2004 led to an
average increase in fluoride concentrations in observations wells of around 30%. This
finding suggests that the impact of droughts can include severe water quality
problems, and that similar increases in fluoride levels and health problems might be
expected as a result of groundwater overexploitation (for irrigation) in these areas.
4.3.13. Use of findings
The results of fluoride surveys were accepted readily by residents of Battuvanipalli
and Pathacheruvu and the day after the surveys, people were busily exchanging
results and had memorised most of them. They wanted information to understand why
they were suffering health problems, and more importantly, so that they could find
solutions. The interest in and acceptance of results contrasted starkly with the
reaction from district-level authorities who stuck firmly to the official line that there
was not a fluoride problem in these villages. It was only after approximately 12
months of interactions that these authorities started to accept the findings of the
preliminary surveys and to recognise that people in these villages have been suffering
because of the water quality of their drinking water supply.
As part of the participatory planning process, ideas for solutions were developed. In
Battuvanipalli for example, options included:
 Alternative source for drinking: Eight out of a total 65 irrigation wells sampled in
the vicinity of Battuvanipalli recorded fluoride levels below 1.5 mg/l (in the range
1.04-1.46 mg/l). There was a cluster of wells about 2 km south of the village where
a new source could be sited, however this area was disputed with a neighbouring
village.
 Recharge of existing source and management: It was thought that enhanced
recharge in the zone close to the existing well and management of water levels in
the area (e.g. by restricting pumping from adjacent irrigation sources) might
influence fluoride levels.
 Treatment: Given the existing good water supply, apart from very high fluoride
levels, household or community level treatment of water was another option.
Eventually, an irrigation well with low fluoride levels 1.5 mg/l (compared to the 3.8
mg/l of the existing source at this time) was used to develop a new drinking water
supply for the village, in addition to the existing supply. The farmer who provided this
well was compensated.
4.3.14. Lessons learnt
The main findings and lessons learnt during the participatory fluoride surveys were
that:
 cross-checking of results between different methods is valuable and essential
 a large variability in fluoride levels between different sources can be expected
 fluoride levels increased during a period of drought and similar impacts might be
expected as a result of overexploitation of groundwater
 the wells with the best quality water may often being used for irrigation
 there were high levels of fluoride in some domestic water sources, and
discrepancies with official statistics, but high levels were supported by health
observations and community reports. Cross checking the results of water quality
monitoring with health statistics, health observations, and community interviews
should be more widely adopted, and
 a high-level of resistance from relevant district authorities can be expected if
participatory surveys produce findings that do not support the official statistics.
This resistance can be broken down over time but patience and sensitive
interactions are required.
Read more
WHiRL. undated. Beware of fluoride in water. Briefing note, India No. 2. Accion
Fraterna, Anantapur.
References and links
www.hach.com is the website of a popular manufacturer of water quality testing
equipment, including the colorimeter used in this study. Hach have agents in
India and other countries.
Seeley, J. 2001. Recognising diversity: disability and rural livelihoods approaches in
India. Natural Resource Perspectives, Number 72. ODI, London.
Smith, M. D. 1999. Study of fluoride treatment. Consultancy Report (for proposed
DFID-supported Andhra Pradesh Water and Environmental Sanitation
Project). WEDC, Loughborough.
van der Sommen, J. J., Gischler, M. A. and Krupanidhi, K. V. J. R. 1998. Monitoring
of groundwater in fluoride-affected areas in Andhra Pradesh (India), in H. A.
J. v. Lanen (editor), Monitoring for Groundwater Management in (semi-) arid
regions. UNESCO, Paris.