1143
Field capacity and moisture loss during active deposition on Tailings Dams
Capacité au champ et perte d’humidité pendant le dépôt actif des résidus
MacRobert C.
University of the Witwatersrand
ABSTRACT: A common method to manage tailings in semi-arid environments is to self-impound the waste as it dries. This research
investigated the degree of in-situ drying of platinum tailings. Following sedimentation and drying a steady state developed. This was
marked by gravimetric water contents varying within a narrow range related to the materials field capacity. Low water contents
indicative of significant suctions were only recorded following 6 months of dormancy. Liquidity indices indicated that during normal
operation only the outer 50 m dried sufficiently to impound the waste stream.
RÉSUMÉ : Les résidus miniers dans des environnements semi-arides sont souvent mis en dépôt et font prise. Cet article présente
l’étude du degré de séchage in situ des résidus de platine. Suite à la sédimentation et au séchage, un état permanent d’hygrométrie est
atteint. Cet état est très proche de la capacité au champ du matériau. Les indices de liquidité montrent que pendant l’opération, un
séchage suffisant pour retenir l’écoulement des résidus, s’effectue seulement sur une profondeur externe de 50 m.
KEYWORDS: Tailings, Reference Evapotranspiration, Moisture Loss, Field Capacity, Strength Gain.
1 INTRODUCTION
The self-impoundment of mine tailings is dependent on
whether the geotechnical behaviour enables strength gain within
realistic time frames. In semi-arid environments this is aided by
the drying effect of evaporation.
This paper presents results of research carried out on the in-
situ drying behaviour of platinum tailings. This was done by
monitoring gravimetric water contents following successive
field depositions on two back-to-back tailings dams over eleven
months. The rate of drying is correlated with Reference
Evapotranspiration with the extent of drying illustrated to be
controlled by field capacity. Liquidity indices are presented to
illustrate the strength gain that occurs.
1.1
Test Work
Test work was carried out on two back-to-back facilities;
Dam 1 a 100 ha conventional upstream spigoted facility (raised
at 2.3 m•year
-1
) and Dam 2, a 100 ha waste rock impoundment
filled via a series of spigots (raised at 4.6 m•year
-1
). Two
separate processing plants supplied similar tailings; the South
Plant to Dam 1 and the North Plant to Dam 2.
Sampling took place every 50 m along a 400 m test section
on Dam 1 with access by a specially constructed catamaran
drawn by a steel cable. On Dam 2 sampling took place every 50
m along a shorter 200 m test section accessed via scaffold from
the pool wall. In both cases the test section ran from the spigot
points to the pool in the interior. Table 1 details the raw data
obtained during the study.
Test depositions were scheduled to deposit 400 mm of
material on each test strip at similar cycle intervals. However
the depth of material deposited was not uniform due to the
inherent beaching behaviour. This resulted in only the outer 100
m having a similar rate of rise of 2.5 m•year
-1
on both test strips.
On Dam 1 less material was deposited past 100 m whereas on
Dam 2 more material was deposited, resulting in rates of rise of
1.2 m•year
-1
and 4.0 m•year
-1
respectively for these sections.
Table 1. Raw Data
Activity
Raw data obtained
Beach sampling via
bulk samples, grab
samples, and auger
samples
Particle size distributions, particle specific
gravities, gravimetric water content,
calibrated gypsum block suction tests and
triaxial permeability tests.
Site climatic data
A-Pan evaporation, rainfall and daily
minimum and maximum temperatures.
South African Weather
Service, Mokopane
Station
Daily temperature, wind speed and relative
humidity.
Historical monitoring
and design data
Atterberg limits, evaporative drying tests,
filter paper suction tests
2 ANALYSIS OF RESULTS
2.1
Rate of Moisture Loss
The results following the sampling of three depositions on
Dam 2 were analysed to determine the rate of moisture loss
during sedimentation and drying to steady state.
The rate of sedimentation was determined by linear
regression using water contents determined from slurry
densities during deposition and grab samples following
cessation of deposition. The density of the slurry varied
considerably with the water content on average 94 % with a
standard deviation of 33 %. Sedimentation was observed to be
complete within 65 hours (7 hour standard deviation) with a
