Aquifer response to recharge–discharge phenomenon: inference from well hydrographs for genetic classification
- 1.2k Downloads
The continuous groundwater level data emanating from a high-frequency automatic water level recorder installed in a purpose-built piezometer provides a true hydrograph. Analyses of such hydrographs fairly reflect the aquifer character and can be used to draw inference for genetic classification of hard rock aquifers. The signature shape of annual water level fluctuation curve (annual cycle) of a piezometer is due to the specific character of the aquifer and the way it responds to the recharge–discharge phenomenon. The pattern of annual cycle remains identical year after year, although its magnitude may vary with the annual quantum of recharge–discharge. Lithology of the aquifer does not control the shape of the curve. Based on the crest and trough shape, the hard rock aquifers of Peninsular India, where the monsoonal pattern of rainfall occurs, have been classified into genetic groups. It is also found that the nature of the aquifer can be determined by visual comparison of apparent line thickness of the hydrograph, where thin lines denote unconfined aquifer and the apparently thicker lines correspond to confining condition. The response of an aquifer to a pumping event can be identified and separated by its pattern. Thus, the aquifer classification can be automated by adopting the proposed classification scheme.
KeywordsWell hydrograph Piezometer Aquifer response Genetic classification Peninsular India
In the present paper, an attempt has been made for the first time using an exhaustive database to classify the hard rock aquifers of Peninsular India based on their response to recharge–discharge phenomenon recorded in the form of a hydrograph. The main objective of the study is to (i) identify the regular patterns of response of aquifer, if any (ii) interpret the characteristic behaviour responsible for such a regular patterns of response of aquifer and (iii) classify the hard rock aquifer based on similarities of aquifer response.
Under World Bank assisted Hydrology Project (Phase-I) purpose-built piezometers were drilled by the Central Ground Water Board (CGWB) in hard rock terrain of the Peninsular India to depths of 30, 60, and 90 m. The depth range selected was to tap weathered zone or shallow fracture/fracture zone down to 30 m, between 30 and 60 m and in the range of 60–90 m depth separately. Groundwater level data and well information were collected. Many of these wells were fitted with a pressure transducer type high frequency digital automatic water level recorder (DWLR) scheduled to record six hourly data to obtain a continuous groundwater level spectrum. Further, these DWLRs also had the capability to automatically record any extreme event (anomalous rise or fall) that may occur intermittently in groundwater levels outside of defined interval. The data so obtained from each piezometer was downloaded onto a computer. Dedicated software “GEMS” (“Groundwater Estimation and Management software” developed by CGWB) was used to store and retrieve these data along with its well information and to generate a hydrograph for a requisite period from all over the country. This was to ensure measurement of the undistorted piezometric head at the desired frequency, which was larger than the traditional frequency. In fact, the frequency is so high that the resulting piezometric hydrograph may almost be continuous. The GEMS can also incorporate corresponding daily rainfall of the piezometer site. More than 250 representative true hydrographs so obtained in uniform format, having time series data of more than one annual cycle, were compared and analyzed. Selected hydrograph were from hard rock area of eight Indian States namely Andhra Pradesh, Chhattisgarh, Gujarat, Karnataka, Maharashtra, Madhya Pradesh, Orissa and Tamilnadu (Fig. 1). Based on identified marked similarities in the crest and trough shape of true hydrographs and by comparison of apparent line thickness of curves representing different litho-type which were spatially distributed over about 1.7 million km2 area, the hard rock aquifers of the Peninsular India have been classified into genetic groups.
In response to monsoon recharge groundwater level rises to its shallowest level, generating the rising limb. Based on the rainfall pattern and amount during monsoon, the peak takes its shape as per the aquifer character.
Shape of crest
Hydrograph crests from study area can be described as pointed, flat or rounded in shape. The order of abundance of crest shape in the Peninsular hard rocks is pointed > flat > rounded (Fig. 4).
The crest shapes are characteristic and do not change for a particular well/piezometer, however the magnitude or width may vary annually. This reflects the aquifer character control over the crest shape, whereas its magnitude and width is controlled by the quantum of recharge. For example flat top crest remains flat top whatever is the magnitude and so on.
Pointed crests are either single peak or multi peaked. The number of peak/peaks and its width varies annually based on rainfall pattern and quantity.
Thus, each well/piezometer has a signature curve pattern. The signature pattern of individual piezometer reflects heterogeneity of the aquifer. However, the curves can be classified into groups based on gross similarity.
V type curves are those curves where both rising and recession limbs have nearly same slope angle though the rising limbs may sometimes be steeper then recession limbs. Also, they generally don’t show any break in slope trend throughout the limb (Fig. 5).
U type curves are those curves where rising limb has a nearly uniform slope, but the recession limb shows a definite break in the slope. The initial discharge is gentle followed by a fast phase of discharge and then again a slowdown giving rise to U shaped curve. Sometimes the slope is initially steep and subsequently gentle after some period of discharge and a combination of U and V type is developed. The lowest part of the trough is not very sharp in angle, rather it is nearly flat or curved (Fig. 6).
S type curves are those, which have sharp initial and ultimate-slope of recession limb and in between flat or very gentle slope giving rise to two breaks in the actual slope of the recession limb. The S type curves indicate low seasonal fluctuation and are not very common, e.g. Sarangarh (Fig. 7).
The significance of the hydrograph shape is in the interpretation of aquifer characteristics and stress. The V type curve indicates quick dissipation, U type indicate initial quick adjustment and then slow dissemination whereas the S type curves are indicative of poor response of the aquifer. The U shape indicates comparatively best sustain diffusivity of the aquifer. The order of abundance found in these shapes in the Peninsular hard rock is V > U > S.
Line patterns of true hydrograph
The true hydrograph having a fixed scale can be used to predict the nature of aquifer based on its apparent line thickness. When annual cycle of true hydrograph due to the scale of potting of time series data produced a compact form and the minute fluctuations in groundwater levels plotted close together, this provides an apparent line thickness and can be visually compared as thin or thick apparent line (Figs. 5, 6, 7). The minute fluctuations of groundwater levels may be caused due to various factors like frequent change in saturation level or in pressure head within aquifer. The detail of this is discussed in the subsequent paragraphs.
Recharge has been defined as “the entry into the saturated zone of water made available at the water-table surface, together with the associated flow away from the water table within the saturated zone” (Freeze and Cherry 1979). The manner in which infiltrating water is transmitted through the system controls system response to recharge. If the rate of recharge were constant and equal to the rate of drainage away from the water table, water levels would not change. For unconfined aquifers, a time lag occurs during which the pressure change in the land surface is propagated through the unsaturated zone to the water table. Air must move through the unsaturated zone to transmit a pressure change. Therefore, an imbalance exists between the pressure on the water in the well and the water in the aquifer until the pressure front arrives at the water table. This imbalance produces a change in the observed water level in the well. Weeks (1979) and Rojstaczer (1988) present in-depth analyses of this phenomenon. The length of the time lag increases with increasing depth to the water table and with decreasing vertical air diffusivity of the unsaturated zone sediments. The barometric efficiency of an unconfined aquifer is not constant, because these two factors can change over time. Techniques for identifying and removing the effects of atmospheric-pressure changes from observed water levels are described in Weeks (1979) and Rasmussen and Crawford (1997). Pressure transducers are often used to monitor water levels; these devices can be affected by atmospheric pressure changes. Non-vented or absolute, transducer readings may reflect atmospheric-pressure changes and could give the false impression that water-level fluctuations are much greater in magnitude than they are in reality. However, all the above-discussed deviations are not significant in the context of the present paper. Well hydrographs have been classified earlier based on its segment slope and are used to compute recharge, specific storage and time lag (Raj 2004). The shape of crest and trough and therefore the shape of the annual cycle is the product of aquifer character, recharge and discharge quantum and its pattern and have been utilized in this study as a tool to arrive in for genetic classification of hard rock aquifer (see the supplementary sheets-1 & 2).
Identification and interpretation of a particular shape of curve is significant in this process. The pointed crest reflects immediate diffusion of water, whereas flat top and rounded crest indicates delayed and slow diffusivity respectively.
Effects of pumping in nearby area can be observed in many of the hydrographs. However a small amount of pumping does not affect the overall shape of recession curve. Thus, comparison of curves, so generated in a constant scale format for any region, can eventually be compared the aquifer character and can be used for classification of aquifers. Phreatic, semi-confined or confined aquifer may show different response for recharge or discharge event based on aquifer character or its position. Further, comparison of any two or more phreatic/semi-confined or confined aquifer is possible based on true hydrograph patterns. The magnitude of fluctuation depends on the quantum of recharge or discharge event and aquifer character.
In Peninsular India, all the hard rock true hydrographs show a predominant annual cycle in response to the wet and dry season. With the onset of monsoon, throughout the monsoon the curve form rising limb and once the monsoon retrieve the curve starts forming recession limb, which continue until the next monsoon (annual cycle). Successive annual cycles join together to form long-term hydrograph. With the full knowledge of fluctuation of groundwater level, the hydraulic conductivity/filtration conditions can be calculated by analyzing recession curve of well hydrograph (Ferdowsian et al. 2001; Gailuma and Vitola 2009; Posavec et al. 2006; Sujono et al. 2004). The rapid decline in recession curve corresponds to better filtration conditions (Gailuma and Vitola 2009). Therefore, it can be stated comparatively that the U shape curves correspond to better hydraulic conductivity (k) than the V shape, whereas the S shape curves represents the lowest hydraulic conductivity of the aquifer. So, based on hydraulic conductivity (k) aquifers can be classified as U > V > S type curve. In the Peninsular hard rock V type is more abundant over U type, while S type is the least abundant type since it largely corresponds to aquicludes.
It has also been observed that the curve types are independent of lithology of the aquifer. Litho-type curve shape correlation indicates both U and V type curve is found in granite, basalt and limestone aquifer (Figs. 5, 6), whereas all granite or basalts do not form any specific type of curve (See the supplementary sheets-1 & 2). Therefore, similarity in porosity and permeability pattern within different litho-unit is responsible for the broad similarity of curve shape.
Implication of automated genetic classification
Once the value of aquifer parameters for aquifer classes in an area is fixed, these can be used for various computations of resource and prediction for its availability.
Managed aquifer recharge can be planned based on the class of the aquifer.
Water use efficiency can be improved upon based on the class of aquifer for food and water security.
The aquifer response to recharge–discharge phenomenon of hard rock aquifers has been investigated by analyzing over 250 well hydrographs, collected from eight States of Peninsular India. The true hydrographs are based on continuous (six hourly) time series data recorded with the help of digital automatic water level recorders fitted in purpose-built piezometers. Due to the high heterogeneity of hard rock aquifers each well/piezometer produces a hydrograph that is unique and does not match with hydrographs of other wells. However, the successive annual curves of the hydrograph of a well remain nearly similar in shape though the magnitude of fluctuation varies based on the quantum of recharge–discharge and its pattern. This clearly reflects that the shape of the hydrograph is controlled by aquifer characteristic and its response and does not entirely depend on recharge–discharge quantum and pattern. The combination of rising and recession limbs produces different crest and trough shapes. Similarity in crest and trough shape represents similarity in aquifer characteristics. As found during this research shapes of the crests and troughs of the true hydrographs can form a basis for devising a genetic classification of the aquifers. Three crest types namely ‘point’, ‘rounded’ and ‘flat top’ crest and three trough types viz. V shape, U shape and S shape have been identified in hard rocks of Peninsular India. The V type curve indicates quick dissipation; U type indicates initial quick adjustment and then slow dissemination whereas the S type curves are indicative of poor response of the aquifer. The U shape indicates relatively better sustained diffusivity of the aquifers. Confined and unconfined aquifers can be differentiated based on the apparent line thickness. Thin line denotes unconfined and thick line denotes confined aquifer. Intermittent apparent thickness of line corresponds to semi-confined aquifer. Fluctuation produced by pumping of well can be segregated based on the fluctuation pattern.
The identified patterns in the hydrographs provide newer insights into the response of hard rock aquifers to recharge–discharge phenomenon. Types of aquifers and their behaviors can be derived from the patterns in the hydrographs. Relative potential of the aquifers can also be assessed from the patterns in the hydrographs. Analysis of available true hydrographs in association with other related data can help in planning groundwater management interventions in hard rock areas.
The proposed aquifer classification has great potential for applied use; several potential lines of research exist that could further the usefulness of the method.
The authors are grateful to Sh. Sushil Gupta, Chairman, CGWB for allowing us to publish the paper and for providing the logistical support during the study. Fruitful discussions with our colleague Sh. S. K. Sinha have been beneficial in finalization of this paper. Sincere thanks are due to Sh. H. K. Sahu, Member Secretary, Upper Yamuna River Board for his encouragement. The paper has been benefited by review comments of an anonymous reviewer and the able editorial handling and suggestions of Dr. Abdulrahman I. Alabdulaaly. The views expressed in the paper are the authors own and not of the department to which they belong.
- Ferdowsian R, Pannell DJ, McCarron C, Ryder AT, Crossing L (2001) Explaining groundwater hydrographs: separating atypical rainfall events from time trends. AJSR 39(4):861–875. http://www.general.uwa.edu.au/u/dpannell/dpap0012.htm
- Freeze RA, Cherry JA (1979) Ground water. Prentice-Hall, Englewood CliffsGoogle Scholar
- Gailuma A, Vitola I (2009) Recession curve analysis approach for groundwater. www.puma.lu.lv/fileadmin
- Hydrology Project (2000) Understanding conventional and DWLR assisted water level monitoring, training modules #1 to 5 by DHV consultants BV & DELFT Hydraulics with Halcrow Tahal, CES, ORG and JPS. pp 1–68. (http://www.cwc.gov.in/main/HP/download/DWLR)
- Kumar CP (2010) Groundwater data requirement and analysis. http://www.acadamea.edu
- Maréchal JC, Sarma MP, Ahmed S, Lachassagne P (2010) Establishment of earth tides effect on water level fluctuations in an unconfined hard rock aquifer using spectral analysis. http://arxiv.org/pdf/1002.3916
- Posavec K, Bacani A, Nakic Z (2006) A visual basic spreadsheet macro for recession curve analysis. Ground water 44(5):764–767Google Scholar
- Singhal BBS (2007) Nature of hard rock aquifers: hydrogeological uncertainties and ambiguities. Groundwater dynamics in hard rock aquifers. Capital Publishing Company, New DelhiGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.