The present study was conducted to evaluate 33 springs' hydrology (discharge and yield estimation) of Chandrabhaga and Danda watersheds of Uttarakhand, India. The springs were classified using Meinzer method and evaluated the relative performance for rejuvenation strategy. It was found that most of springs fall in sixth and seventh class order with flow rate 6.5 to 65.5 and 0.8 to 6.5 m3·day−1, respectively. The relative performance of springs were analyzed based on four methods: (i) spring flow variability, (ii) normalized spring flow (short and long duration), (iii) rainfall spring flow lag and (iv) spring flow gradient. The relative results of springs were analyzed on a scale of 0–5. The Chandrabhaga springs 01, 03, 4B, 05, 06 and 13 were found to be relatively good on a scale value of 4 out of 5 as compared to springs 4A, 07, and 10A with a scale value of 1. For the Danda watershed, the relative performance of springs 4A and 28 found on scale value of 5 and springs 4B, 11 and 20 with a scale value of 4 are relatively good compared to springs 02, 06, 07, 15 and 17. The cumulative flow of spring showed a linear response with cumulative rainfall for the period of June to September (monsoon period). The spring-shed was delineated and evaluated for optimization for the maximum efficiency, spring flow, ratio of area and relief versus maximum spring flow yield. The results revealed that the quantification of water fluxes for water balances, storage of groundwater and development of mathematical models can be used for sustainable water resources development and to revive the mountain springs which helped the adverse impacts of climate change.

The Himalayan region extends up to 2,400 km and runs Northwest to Southeast with a width of 150–400 km (Hasnain 1999; Jeelani et al. 2015). Springs are rich in natural resources and provide fundamental needs for the existence of life at mountain but it also provides water at downstream through its perennial river system (Agarwal et al. 2012; Kulkarni & Thakkar 2012; Lamban et al. 2015). Sometimes the availability of natural resources at hills goes to acute shortage due to uneven distribution of rainfall in space and time.

There are 5 million springs across India, of which nearly 3 million are in the Indian Himalayan region (Pant & Rawat 2015). Springs have not received their due attention and are today facing the threat of drying up. Spring discharge is reported to be declining due to increased water demand, changing land use patterns, and ecological degradation. With the climate change scenario, the spring-shed are observed in rising temperatures, rise in rainfall intensity, reduction in its temporal spread and a marked decline in winter rain (Negi & Joshi 2004; Razandi et al. 2015). The water quality is also deteriorating under changing land use and improper sanitation (Negi & Joshi 2004).

A spring is a place where water emerges from the ground and flows over the land surface. Water is continuously replenished from below through the pores of the ground over a considerable area. Springs are normally classified either by their origin or by the average size of flow rate (Mustafa et al. 2015). Springs in lesser Uttarakhand normally fall into the group as springs of impervious rocks. These springs are emanating naturally from unconfined aquifers. The discharge from springs has decreased drastically due to human activities (Mukhopadhyay & Khan 2014). In the Himalayan region mostly, unconfined aquifers are the water sources of the springs where the water flow comes out under gravity. The behavior of a spring can only be administered and forecasted by studying its temporal discharge variation hydrograph (Vashishth & Sharma 2007).

The spring yield during rainy and non-rainy seasons is mainly affected by rainfall and recharge area characteristics. In the Central Himalayas, around eight types of springs are recognized on the basis of the geology, nature of water bearing formations and the conditions related to their formation. The highest yielding spring (Negi & Joshi 2004) was obtained from fluvial originating springs and the rate was 405 × 103 L·d−1 whereas the colluviums originating springs were at the lowest rate and producing 7.2 × 103 L·d−1 (Kulkarni & Vijay 2014).

The lesser Mid-Western Himalayan watersheds are rich with perennial or non-perennial springs and provide the fundamental basis for settlement as well as the existence of life. These water resources have low discharge rates and most of them fall under the lowest flow category between 0.82 and 6.5 m3day−1 flow rate which is classified as the seventh or eighth category of Meinzer's (1918) spring classification. Springs can be classified according to origin and size with the average flow from the spring. The present study evaluated the spring flow rate, identification of relative spring flow performance and development of spring flow models based on spring-shed and topography.

The study springs lying in the Western Himalaya agro-ecological region of Uttarakhand are located between latitude N30018’ to N30019’ and longitude E78035’ to E78036’ (Chandrabhaga and Danda, Figure 1) at altitudes ranging from 1,070 m to 2,350 m with the average annual rainfall of 1,200 mm. The lithology of the study area is mainly phyllites and schists rocks of Chandpur formation (Lesser Himalaya). The rocks of Chandpur formation as ‘the olive green and grey phyllite inter-bedded and finely inter-banded with meta-silt stone and a very fine grained wackes with local metavolcanics’ to act as filter that regulates the groundwater paths for the spring as conduit or diffused or a combination of both (Tambe et al. 2012). The springs are gravitational fracture of perennial or seasonal in nature. The highly weathered and immensely fractured geological system allows a rapid water transit through the aquifer (Figure 1). The springs dry up with early summer as the soil, which has water retaining capacity, is being degraded due to deforestation and thinning of forest cover and due to rainfall pattern with increasing high intensity storms with longer dry spells.

Figure 1

Location for springs of Chandrabhaga and Danda watersheds. (a) Chandrabhaga watershed. (b) Danda watershed.

Figure 1

Location for springs of Chandrabhaga and Danda watersheds. (a) Chandrabhaga watershed. (b) Danda watershed.

Close modal

Himalayan springs, which can be used in water resources management in a mountain region emanating naturally from unconfined aquifers, are the primary source of water for rural households (Tambe et al. 2012). The Indian Himalayan springs are drying up due to climate change effect (Tambe et al. 2012). Springs of Western Himalayan (Uttarakhand) watershed are rain-fed springs and an uneven distribution of rainfall in space and time restricts the availability of natural spring flow (Negi & Joshi 2004; Pourtaghi & Pourghasemi 2014; Katsuyama et al. 2015). Climate change on precipitation pattern, reduction in temporal spread with other anthropogenic causes is another problem of dying springs in Himalayan region. Due to the impacts of climate change on precipitation patterns such as rise in rainfall intensity, reduction in its temporal spread, and a marked decline in winter rain, coupled with other anthropogenic causes, the problem of dying springs is being increasingly felt across this region (Vashishth & Sharma 2007; Tambe et al. 2012). Tambe et al. (2012) conducted study on springs in Sikkim of Himalayan region and found the discharge of the springs to peak at 51 L/min during the post monsoon months (September–November) and then diminish to 8 L/min during spring (March–May). The lean period (March–May) discharge is perceived to have declined by nearly 50% in drought-prone areas and by 35% in other areas over the last decade. In the Himalayan region, glaciers are receding, erratic rainfall and deforestation are leading factors to severe water shortages in the uplands (Tambe et al. 2012).

The field data of springs were collected automatically as well as manually. The rainfall measurements are based on five automatic rain-gauges installed in each watershed. Spring flow measurements are manually recorded by measuring the time taken for a specific amount of water coming out of the spring. During high flow season in monsoon months, the specific quantity was taken as 5 L and in non-monsoon months it was reduced to 1 L. A total of 37 (21 in Chandrabhaga and 16 in Danda) springs were under observation in the watersheds. The Meinzer's classification of springs based on average flow is illustrated in Table 1. Since the springs are distributed throughout in the watershed (Figure 1), the measurements are taken on alternate days or a maximum of 3 days' interval in both the watersheds. The data collection was started in year 1999 and continued until June 2013. All springs of the watersheds were considered for observation. The watershed runoff was measured by installing V-notch in each watershed with different pressure gauge recorder to record the water head on the notch. Continuous flow (CF) throughout the year was recorded for five springs. Few are with interpreted flow (IF) with once or twice and even more in the recorded period due to low rainfall. Few springs showed very high numbers of interrupted flow/almost dried/dead mainly due to construction and development activities in the area. Few springs are destroyed in the road cutting/ widening and could not be continued.

Table 1

Meinzer's classification of springs according to average flow

MagnitudeAv. flow (m3/day)
1st ≥244,657 
2nd 24,465–244,657 
3rd 2,446–24,465 
4th 654–2,446 
5th 65.4–654 
6th 6.5–65.5 
7th 0.8–6.5 
8th <0.8 
MagnitudeAv. flow (m3/day)
1st ≥244,657 
2nd 24,465–244,657 
3rd 2,446–24,465 
4th 654–2,446 
5th 65.4–654 
6th 6.5–65.5 
7th 0.8–6.5 
8th <0.8 

Since, most of the springs of the watershed fall in sixth magnitude with flow ranging from 6.5 to 65.5 m3 day−1. Therefore it is important to study the relative performance of springs for the purpose of identification of poor springs and its rejuvenation. For this, four methods viz. (1) spring flow variability, (2) normalized spring flow, (3) rainfall spring flow lag and (4) spring flow gradient were used to determine relative performance. The relative performance of springs also analyzed in monsoon and non-monsoon months for both places of springs.

The rate of flow change with flow for a differential time dt which is expressed as a function of Q (Equation (1));
(1)
where, ∫ (Q) is a functional characteristics for a given catchment. The equation for flow data can be approximated by actual flow measurements at Qt and Qt+1 at successive time Δt apart and the differential linear form can be expressed as shown in Equation (2).
(2)

The master recession curve is determined from the plot of dQ/dt against Q. The graphical analysis of Qt+Δt against Qt or the equivalent form of dQ/dt against Q represents the upper and lower observed recession rates representing maximum and minimum recession rates. The plots obtained are sensitive to the data quality as well as the selected time interval. Daily time interval has normally been recommended for the analysis where stress is on low flow. A 2 day time interval is recommended the by Institute of Hydrology as a compromise between data and accuracy. A period longer than 2 days may provide the general behavior of flow representing maximum (upper) and minimum (lower) recession rates. The upper and lower recession rates against higher time month interval was used for spring classification.

Mass spring flow is another spring indicator when divided by mass rainfall. The results of normalized mass spring flow give the spring's performance with time. Thus the normalized mass spring flow is independent of rainfall and more reflects the effects of spring-shed changes on spring flow. These plots were developed for all springs of Chandrabhaga and Danda watersheds. Monthly total rainfall and spring flow values were used for development of normalized mass spring flow plots. These plots provided points where springs are changing its behavior but also the long term spring health.

Spring flow response to rainfall

The cumulative rainfall (mm) and cumulative spring flow (m3) of water year (June–May) was extracted from the data of all the springs of the two watersheds. For 10 randomly selected springs average cumulative flow with cumulative rainfall for the monsoon and non-monsoon period was plotted, shown in Figures 2 and 3, for Chandrabhaga and Danda springs, respectively. It is evident from Figure 2 that the cumulative spring flow showed almost a linear response with cumulative rainfall for the period of June to September (monsoon period). The linear response during monsoon suggested that the recharge area of spring is in close vicinity of the spring within the watershed and flow is mainly due to rainfall. A non-linear response from September to January, possibly because this period is non-monsoon period and the rainfall in this period is too less. Again, for the period from January to February, the spring showed a linear response as this period receives the winter rain. Further the period February to May (non-monsoon period) is dry and the spring response is again non-linear (Figures 2 and 3). The behavior suggests that rainfall is an important source of recharging the springs and the impact of winter rains is well recognized in sustaining the spring flow for the forthcoming summer. Exponential or power regression exists between two rainfalls and cumulative spring flow.

Figure 2

Cumulative rainfall with cumulative spring flow monsoon and non-monsoon period for 10 selected springs of Chandrabhaga watershed.

Figure 2

Cumulative rainfall with cumulative spring flow monsoon and non-monsoon period for 10 selected springs of Chandrabhaga watershed.

Close modal
Figure 3

Cumulative rainfall with cumulative spring flow for monsoon and non-monsoon period of 10 selected springs of Danda watershed.

Figure 3

Cumulative rainfall with cumulative spring flow for monsoon and non-monsoon period of 10 selected springs of Danda watershed.

Close modal

Spring classification

Average daily spring flow was estimated using total spring flow which was recorded for all springs for classification of springs as per Meinzer's (1918) classification system. It is found that most of the springs in the area fall under sixth magnitude, with flow ranging from 6.5 to 65.5 m3·day−1. Springs 08 and 11 of Chandrabhaga watershed and springs 4B and 16 of Danda watershed fall under the seventh magnitude class with flow ranging from 0.8 to 6.5 m3day−1. In general, the spring has shown low discharging and most springs fall under the lowest flow class: see Meinzer's (1918) spring classification as summarized in Table 2.

Table 2

Spring data and classification of Chandrabhaga and Danda springs

Springs of Chandrabhaga watershed
Springs of Danda watershed
S. no.Spring no.Start dateEnd dateElevation (m a.s.l)RemarksSpring no.Start dateEnd dateElevation (m a.s.l)Remarks
1-Jul-99 30-Jun-10 1,432 CF 1-Jul-99 30-Jun-13 1,249 CF 
1-Jul-99 30-Jun-10 1,413 CF 1-Jul-99 30-Jun-13 1,295 CF 
1-Jul-99 30-Jun-10 1,387 CF 1-Jul-00 30-Jun-13 1,269 CF 
4A 1-Jul-99 30-Jun-10 1,169 CF 4A 1-Jul-99 30-Jun-13 1,242 CF 
4B 1-Jul-99 30-Jun-10 1,220 CF 4B 1-Jul-99 30-Jun-13 1,249 CF 
1-Jul-99 30-Jun-10 1,415 IF 01 1-Jul-99 30-Jun-13 1,237 CF 
1-Jul-99 30-Jun-10 1,460 CF 1-Jul-00 30-Jun-13 1,194 IF 01 
1-Jul-99 30-Jun-10 1,569 IF 01 1-Jul-99 30-Jun-13 1,121 CF 
1-Jul-99 30-Jun-10 1,543 IF 02 1-Jul-99 30-Jun-13 1,157 IF 01 
10 1-Jul-99 30-Jun-10 1,431 CF 1-Jul-99 30-Jun-13 1,213 IF 01 
11 10A 1-Jul-99 30-Jun-10 1,608 CF 11 1-Jul-00 30-Jun-13 940 IF 02 
12 11 1-Jul-99 30-Jun-10 1,506 IF 07 13 1-Jul-00 30-Jun-13 1,191 IF 03 
13 12 1-Jul-00 30-Jun-02 1,558 Dead 15 1-Jul-00 30-Jun-02 1,157 CF 
14 13 1-Jul-03 30-Jun-10 1,115 CF 16 1-Jul-03 30-Jun-04 1,121 IF 01 
15 14 1-Jul-00 30-Jun-06 1,681 Dead 20 1-Jul-00 30-Jun-13 1,266 CF 
16 15 1-Jul-00 30-Jun-10 1,742 IF 01 28 1-Jul-02 30-Jun-13 938 CF 
17 16 1-Jul-00 30-Jun-10 1,661 IF 01      
18 17 1-Jul-05 30-Jun-10 1,694 IF 01      
Springs of Chandrabhaga watershed
Springs of Danda watershed
S. no.Spring no.Start dateEnd dateElevation (m a.s.l)RemarksSpring no.Start dateEnd dateElevation (m a.s.l)Remarks
1-Jul-99 30-Jun-10 1,432 CF 1-Jul-99 30-Jun-13 1,249 CF 
1-Jul-99 30-Jun-10 1,413 CF 1-Jul-99 30-Jun-13 1,295 CF 
1-Jul-99 30-Jun-10 1,387 CF 1-Jul-00 30-Jun-13 1,269 CF 
4A 1-Jul-99 30-Jun-10 1,169 CF 4A 1-Jul-99 30-Jun-13 1,242 CF 
4B 1-Jul-99 30-Jun-10 1,220 CF 4B 1-Jul-99 30-Jun-13 1,249 CF 
1-Jul-99 30-Jun-10 1,415 IF 01 1-Jul-99 30-Jun-13 1,237 CF 
1-Jul-99 30-Jun-10 1,460 CF 1-Jul-00 30-Jun-13 1,194 IF 01 
1-Jul-99 30-Jun-10 1,569 IF 01 1-Jul-99 30-Jun-13 1,121 CF 
1-Jul-99 30-Jun-10 1,543 IF 02 1-Jul-99 30-Jun-13 1,157 IF 01 
10 1-Jul-99 30-Jun-10 1,431 CF 1-Jul-99 30-Jun-13 1,213 IF 01 
11 10A 1-Jul-99 30-Jun-10 1,608 CF 11 1-Jul-00 30-Jun-13 940 IF 02 
12 11 1-Jul-99 30-Jun-10 1,506 IF 07 13 1-Jul-00 30-Jun-13 1,191 IF 03 
13 12 1-Jul-00 30-Jun-02 1,558 Dead 15 1-Jul-00 30-Jun-02 1,157 CF 
14 13 1-Jul-03 30-Jun-10 1,115 CF 16 1-Jul-03 30-Jun-04 1,121 IF 01 
15 14 1-Jul-00 30-Jun-06 1,681 Dead 20 1-Jul-00 30-Jun-13 1,266 CF 
16 15 1-Jul-00 30-Jun-10 1,742 IF 01 28 1-Jul-02 30-Jun-13 938 CF 
17 16 1-Jul-00 30-Jun-10 1,661 IF 01      
18 17 1-Jul-05 30-Jun-10 1,694 IF 01      

CF: Continuous flow.

IF: Interrupted flow x times.

Dead: spring dead due to C&D (construction and demolition).

Meters above sea leve (m a.s.l).

Relative performance of springs based on spring flow variability

Average minimum flow and average of maximum–minimum flow over years of the spring are related with average maximum spring flow for the variability of spring flow which is illustrated in Figure 4. The points falling between two graphs are identified as good springs in comparison to others that fall on line or out of the area. It is evident from Figure 4 that the identified springs has moderate discharge (good) or less discharge (bad) was identified. The springs identified comparison with discharge (good) or less discharge (bad) are shown in Table 3 and Figure 4. It is evident from Table 3 that the springs 01, 02, 03, 06, 09, 13, and 16 have good discharge and springs 4A, 4B, 05, 07, 08, 10A, 11, 14, 15, and 17 have less discharge for Chandrabhaga watershed springs. In Danda watershed, the springs 02, 03, 4A, 4B, 05, 11, and 15 are identified as good discharge springs and 01, 06, 07, 08, 09, 13, 16, 20 and 28 as bad discharge springs.

Table 3

Relative performance of springs based on the spring flow variability for Chandrabhaga and Danda springs

Springs of Chandrabhaga watershed
Springs of Danda watershed
S. no.Spring no.Maximum flow (m3/day)Average flow (m3/day)Spring classSpring no.Maximum flow (m3/day)Average flow (m3/day)Spring class
15.1 9.5 6th 88.8 26.9 6th 
21 12.4 6th 93.2 31.4 6th 
117.4 44 6th 69.2 25.6 6th 
4A 49.8 17.5 6th 4A 121.2 43.5 6th 
4B 48 14.1 6th 4B 20.4 6.1 7th 
28.3 8.2 6th 85.8 31.5 6th 
25.5 12.2 6th 31.0 7.2 6th 
49.3 12.5 6th 74.4 15.1 6th 
17.4 4.8 7th 54.4 11.9 6th 
10 46.3 19.8 6th 92.0 26.1 6th 
11 10A 27.9 10.2 6th 11 46.6 18.7 6th 
12 11 14.7 4.5 7th 13 47.3 15.5 6th 
13 13 49.6 23.4 6th 15 33.5 11.5 6th 
14 14 34.4 12.7 6th 16 22.7 5.9 7th 
15 15 57.3 19.7 6th 20 96.6 31.6 6th 
16 16 18 8.1 6th 28 161.0 56.9 6th 
17 17 34.9 10.2 6th     
Springs of Chandrabhaga watershed
Springs of Danda watershed
S. no.Spring no.Maximum flow (m3/day)Average flow (m3/day)Spring classSpring no.Maximum flow (m3/day)Average flow (m3/day)Spring class
15.1 9.5 6th 88.8 26.9 6th 
21 12.4 6th 93.2 31.4 6th 
117.4 44 6th 69.2 25.6 6th 
4A 49.8 17.5 6th 4A 121.2 43.5 6th 
4B 48 14.1 6th 4B 20.4 6.1 7th 
28.3 8.2 6th 85.8 31.5 6th 
25.5 12.2 6th 31.0 7.2 6th 
49.3 12.5 6th 74.4 15.1 6th 
17.4 4.8 7th 54.4 11.9 6th 
10 46.3 19.8 6th 92.0 26.1 6th 
11 10A 27.9 10.2 6th 11 46.6 18.7 6th 
12 11 14.7 4.5 7th 13 47.3 15.5 6th 
13 13 49.6 23.4 6th 15 33.5 11.5 6th 
14 14 34.4 12.7 6th 16 22.7 5.9 7th 
15 15 57.3 19.7 6th 20 96.6 31.6 6th 
16 16 18 8.1 6th 28 161.0 56.9 6th 
17 17 34.9 10.2 6th     
Figure 4

Relative springs' performance of springs based on the spring flow variability for Chandrabhaga and Danda springs.

Figure 4

Relative springs' performance of springs based on the spring flow variability for Chandrabhaga and Danda springs.

Close modal

Relative performance of springs based on normalized spring flow

Accumulated rainfall and accumulated spring flow and their normalized spring flow for springs are presented in Figures 5 and 6, respectively, for Chandrabhaga and Danda watersheds. Accumulated mass and normalized mass spring flow of springs (03 and 07) is shown in Figure 5. The accumulated mass spring flow is found to be increasing but has not followed the path of the rainfall. This behavior of these two springs is not similar to rainfall. This non-similarity is clearly seen in normalized mass spring flow. An increasing trend in normalized value is a clear indication of the spring improvement. Spring number 03 is almost stable up to year 2006 with discharge fluctuations. The same spring has shown improvement from 2006 to 2010; further, it has shown a declining trend. The spring number 07 has shown a slow, continuous decline with discharge fluctuation in the initial years. Accumulated mass spring flow of springs 01, 02 and 03 are almost similar but not exactly as that of rainfall pattern which is shown in Figure 6. The normalized mass spring flow of springs 01 and 02 is almost similar with the increasing trend up to year 2005 and has further shown a decreasing trend. Spring 03 shows a continuous improvement up to year 2013 with some oscillations.

Figure 5

Accumulated mass rainfall and accumulated mass spring flow and their normalized spring flow for Chandrabhaga watershed.

Figure 5

Accumulated mass rainfall and accumulated mass spring flow and their normalized spring flow for Chandrabhaga watershed.

Close modal
Figure 6

Accumulated mass rainfall and accumulated mass spring flow and their normalized spring flow for Danda watershed.

Figure 6

Accumulated mass rainfall and accumulated mass spring flow and their normalized spring flow for Danda watershed.

Close modal

For relative performance of the spring, the total recorded period is divided in two categories: (i) two short terms (June 2001 to May 2005 and June 2005 to May 2013) and (ii) one long term (June 2001 to May 2013) to study the behavior change of normalized mass spring flow. The difference between the beginning and the end of the period was identified and used as an index to say a spring has good discharge or less discharge in that period. A negative value defines the spring as bad and a positive value defines it as good. Some short peaks and falls can also be seen in the period and are considered short term responses on the spring flow. The performance results of two short and one long period for the springs of both watersheds are illustrated in Tables 4 and 5. The relative performance for the short term and long term are self-explanatory in Table 4. Further for long term, from June 2001 to May 2013, the Chandrabhaga springs 01, 03, 4B, 05, 06, 08, 13, 15, 16, 17 are observed as good (discharge) and 02, 4A, 07, 09, 10A, 11, 14 as bad (less discharge). For the same period, the Danda springs 01, 02, 4A, 4B, 05, 07, 08, 09, 11, 13, 15, 16, 20, 28, 29 are identified as good and spring 06 as bad spring.

Table 4

Relative performance of spring based on normalized mass spring flow for two short and one long period for Chandrabhaga springs

Springs of Chandrabhaga watershed
Springs of Danda watershed
Spring no.Av. max flow (m3/day)Av. min flow (m3/day)Max-min flow (m3/day)Based on variabilitySpring no.Av. max flow (m3/day)Av. min flow (m3/day)Max-min flow (m3/day)Based on variability
15.1 6.2 9.0 Good 88.8 5.3 83.5 Bad 
21.0 7.4 13.6 Good 93.2 12.1 81.1 Good 
117.4 17.2 100.1 Good 69.2 8.6 60.6 Good 
4A 49.8 6.5 43.4 On line 4A 121.2 13.1 108.1 Good 
4B 48.0 2.9 45.2 Bad 4B 20.4 3.6 16.9 Good 
28.3 1.9 26.4 Bad 85.8 12.6 73.2 Good 
25.5 6.6 18.9 Good 31.0 0.8 30.2 Bad 
49.3 1.2 48.1 Bad 74.4 0.9 73.5 Bad 
17.4 1.0 16.5 Bad 54.4 1.3 53.1 Bad 
46.3 9.4 36.9 Good 92.0 2.1 89.9 Bad 
10A 27.9 3.0 24.8 On line 11 46.6 7.3 39.3 Good 
11 14.7 0.7 14.0 Bad 13 47.3 3.5 43.8 Bad 
13 49.6 9.5 40.2 Good 15 33.5 4.0 29.5 Good 
14 34.4 2.9 31.5 Bad 16 22.7 1.0 21.7 Bad 
15 57.3 5.5 51.8 Bad 20 96.6 8.9 87.6 On line 
16 18.0 4.2 13.9 Good 28 161.0 15.1 145.9 On line 
17 34.9 3.0 31.9 Bad      
Springs of Chandrabhaga watershed
Springs of Danda watershed
Spring no.Av. max flow (m3/day)Av. min flow (m3/day)Max-min flow (m3/day)Based on variabilitySpring no.Av. max flow (m3/day)Av. min flow (m3/day)Max-min flow (m3/day)Based on variability
15.1 6.2 9.0 Good 88.8 5.3 83.5 Bad 
21.0 7.4 13.6 Good 93.2 12.1 81.1 Good 
117.4 17.2 100.1 Good 69.2 8.6 60.6 Good 
4A 49.8 6.5 43.4 On line 4A 121.2 13.1 108.1 Good 
4B 48.0 2.9 45.2 Bad 4B 20.4 3.6 16.9 Good 
28.3 1.9 26.4 Bad 85.8 12.6 73.2 Good 
25.5 6.6 18.9 Good 31.0 0.8 30.2 Bad 
49.3 1.2 48.1 Bad 74.4 0.9 73.5 Bad 
17.4 1.0 16.5 Bad 54.4 1.3 53.1 Bad 
46.3 9.4 36.9 Good 92.0 2.1 89.9 Bad 
10A 27.9 3.0 24.8 On line 11 46.6 7.3 39.3 Good 
11 14.7 0.7 14.0 Bad 13 47.3 3.5 43.8 Bad 
13 49.6 9.5 40.2 Good 15 33.5 4.0 29.5 Good 
14 34.4 2.9 31.5 Bad 16 22.7 1.0 21.7 Bad 
15 57.3 5.5 51.8 Bad 20 96.6 8.9 87.6 On line 
16 18.0 4.2 13.9 Good 28 161.0 15.1 145.9 On line 
17 34.9 3.0 31.9 Bad      
Table 5

Relative performance of spring based on normalized mass spring flow for two short and one long period for Danda springs

S. no.Spring no.June 01 to May 05June 05 to May 13June 01 to May 13June 01 to May 05June 05 to May 13June 01 to May 13
01 −0.79 0.98 0.04 Bad Good Good 
02 −0.91 0.82 −0.31 Bad Good Bad 
03 −1.14 2.74 0.59 Bad Good Good 
4A −0.43 0.56 −0.31 Bad Good Bad 
4B 1.09 1.79 2.77 Good Good Good 
05 −0.53 1.01 0.32 Bad Good Good 
06 0.12 0.20 0.05 Good Good Good 
07 −1.52 −1.38 −3.40 Bad Bad Bad 
08 NA 0.78 0.78 NA Good Good 
10 09 −3.54 −0.21 −4.34 Bad Bad Bad 
11 10A −1.21 0.00 −1.48 Bad Bad Bad 
12 11 −0.64 −0.33 −1.10 Bad Bad Bad 
13 13 2.56 3.53 5.95 Good Good Good 
14 14 −1.59 0.64 −1.19 Bad Good Bad 
15 15 0.26 1.23 1.07 Good Good Good 
16 16 0.44 0.78 1.10 Good Good Good 
17 17 NA 1.15 1.15 NA Good Good 
S. no.Spring no.June 01 to May 05June 05 to May 13June 01 to May 13June 01 to May 05June 05 to May 13June 01 to May 13
01 −0.79 0.98 0.04 Bad Good Good 
02 −0.91 0.82 −0.31 Bad Good Bad 
03 −1.14 2.74 0.59 Bad Good Good 
4A −0.43 0.56 −0.31 Bad Good Bad 
4B 1.09 1.79 2.77 Good Good Good 
05 −0.53 1.01 0.32 Bad Good Good 
06 0.12 0.20 0.05 Good Good Good 
07 −1.52 −1.38 −3.40 Bad Bad Bad 
08 NA 0.78 0.78 NA Good Good 
10 09 −3.54 −0.21 −4.34 Bad Bad Bad 
11 10A −1.21 0.00 −1.48 Bad Bad Bad 
12 11 −0.64 −0.33 −1.10 Bad Bad Bad 
13 13 2.56 3.53 5.95 Good Good Good 
14 14 −1.59 0.64 −1.19 Bad Good Bad 
15 15 0.26 1.23 1.07 Good Good Good 
16 16 0.44 0.78 1.10 Good Good Good 
17 17 NA 1.15 1.15 NA Good Good 

Relative performance of springs based on rainfall spring flow lag

Daily values of rainfall and spring flow were used for estimation of spring-rainfall lag for springs of Chandrabhaga and Danda watersheds. Analysis indicates a lag varying from 9 to 30 days for different springs. A low value of spring-rainfall lag in days suggested that the springs of watershed springs are fast responding, highly variable and may not be considered as good springs. The estimated lag values for the springs of watershed are shown in Table 6. The lag values are used to classify the spring as being good or bad. If lag is more than the average lag in days, the springs are relatively good, and for less than average value, then the spring is considered bad. The springs classified as good and bad are reported in Table 6.

Table 6

Relative performance of springs based on rainfall spring flow lag

S. noSpring no.June 01 to May 05June 05 to May 13June 01 to May 13June 01 to May 05June 05 to May 13June 01 to May 13
01 7.63 −0.69 5.55 Good Bad Good 
02 9.55 −2.93 4.78 Good Bad Good 
03 4.19 3.99 7.44 Good Good Good 
4A −3.02 10.94 7.21 Bad Good Good 
4B NA 1.62 1.62 NA Good Good 
05 6.75 2.90 8.97 Good Good Good 
06 −0.06 0.03 −0.35 Bad Good Bad 
07 2.34 −0.55 0.93 Good Bad Good 
08 1.63 1.97 3.40 Good Good Good 
10 09 10.71 −4.08 4.73 Good Bad Good 
11 11 8.64 −3.43 4.02 Good Bad Good 
12 13 4.05 0.08 3.43 Good Good Good 
13 15 3.18 −0.27 2.42 Good Bad Good 
14 16 3.75 −1.29 2.12 Good Bad Good 
15 20 8.35 −0.29 6.48 Good Bad Good 
16 28 17.19 1.27 16.07 Good Good Good 
S. noSpring no.June 01 to May 05June 05 to May 13June 01 to May 13June 01 to May 05June 05 to May 13June 01 to May 13
01 7.63 −0.69 5.55 Good Bad Good 
02 9.55 −2.93 4.78 Good Bad Good 
03 4.19 3.99 7.44 Good Good Good 
4A −3.02 10.94 7.21 Bad Good Good 
4B NA 1.62 1.62 NA Good Good 
05 6.75 2.90 8.97 Good Good Good 
06 −0.06 0.03 −0.35 Bad Good Bad 
07 2.34 −0.55 0.93 Good Bad Good 
08 1.63 1.97 3.40 Good Good Good 
10 09 10.71 −4.08 4.73 Good Bad Good 
11 11 8.64 −3.43 4.02 Good Bad Good 
12 13 4.05 0.08 3.43 Good Good Good 
13 15 3.18 −0.27 2.42 Good Bad Good 
14 16 3.75 −1.29 2.12 Good Bad Good 
15 20 8.35 −0.29 6.48 Good Bad Good 
16 28 17.19 1.27 16.07 Good Good Good 

Relative performance of springs based on spring flow gradient

The upper and lower recession rates against higher time interval of months is used for spring classification. Master recession curves were developed from the plot of dQ/dt against Q by combining all the springs of each watershed (Appendix A1) and the master recession rate was estimated by the curve fitting of power relationship. The upper and lower recession rates for Chandrabhaga and Danda springs are approximated by plotting the upper and lower envelop and the recession values obtained by the curve fitting which is given in Table 7. A correction constant was estimated for upper and lower recession rates for both the watersheds by dividing upper and lower recession rates by respective master recession (Table 7).

Table 7

The recession rates for the springs of Chandrabhaga and Danda watersheds

S. noChandrabhaga springs
Danda springs
Spring no.Lag in daysSpring conditionSpring no.Lag in daysSpring condition
13 Bad 22 Good 
13 Bad 12 Bad 
27 Good Bad 
4A 13 Bad 4A 24 Good 
4B 25 Good 4B 22 Good 
27 Good Bad 
29 Good 12 Bad 
28 Good 10 Bad 
30- Good Bad 
10 29 Good 18 Good 
11 10A 17 Bad 11 29 Good 
12 11 21 Good 13 Bad 
13 13 28 Good 15 Bad 
14 14 16 Bad 16 17 Good 
15 15 14 Bad 20 25 Good 
16 16 14 Bad 28 25 Good 
17 17 11 Bad    
 Av. lag 20.3  Av. lag 15  
S. noChandrabhaga springs
Danda springs
Spring no.Lag in daysSpring conditionSpring no.Lag in daysSpring condition
13 Bad 22 Good 
13 Bad 12 Bad 
27 Good Bad 
4A 13 Bad 4A 24 Good 
4B 25 Good 4B 22 Good 
27 Good Bad 
29 Good 12 Bad 
28 Good 10 Bad 
30- Good Bad 
10 29 Good 18 Good 
11 10A 17 Bad 11 29 Good 
12 11 21 Good 13 Bad 
13 13 28 Good 15 Bad 
14 14 16 Bad 16 17 Good 
15 15 14 Bad 20 25 Good 
16 16 14 Bad 28 25 Good 
17 17 11 Bad    
 Av. lag 20.3  Av. lag 15  

The upper and lower envelope for each spring is to be drawn from the respective plot of dQ/dt against Q and to estimate the upper and lower recession rates. Since it was difficult to estimate the upper and lower recession rates, the master recession values of all the springs were first estimated. The upper and lower recession rates were estimated by multiplying master recession value with respective spring correction constant as estimated and given in Table 7.

The upper and lower recession rates and the master recession rates are used for the classification of springs. If the recession rate (master, upper and lower) is less than its mean, the spring is good, otherwise the spring is bad. The results for both Chandrabhaga and Danda springs are illustrated in Table 8. A relative spring performance scale was developed considering the performance of springs through each method. The best performing spring obtains five goods out of five through five procedures and the worst obtains zero value. For Chandrabhaga watershed, the springs 01, 03, 4B, 05, 06 and 13 are relatively good as compared to others.

Table 8

Relative performance of springs based on spring flow recession values

Recession ratesSprings of Chandrabhaga
Springs of Danda
RecessionCorrection constantRecessionCorrection constant
Master recession rate 0.9669 – 1.1122 – 
Upper recession rate 3.1414 3.2489 3.1414 2.8244 
Lower recession rate 0.2284 0.2362 0.2284 0.2053 
Recession ratesSprings of Chandrabhaga
Springs of Danda
RecessionCorrection constantRecessionCorrection constant
Master recession rate 0.9669 – 1.1122 – 
Upper recession rate 3.1414 3.2489 3.1414 2.8244 
Lower recession rate 0.2284 0.2362 0.2284 0.2053 

Spring flow modeling was attempted with a new concept by delineating the boundaries of the individual spring or a group of springs. Geographic information system (ArcGis) software was used to draw the flow direction on the surface of the watershed with reference to the spring's outlets of the springs. Using this flow direction and adopting the assumptions 1, 2 and 3, the approximate area as spring-shed was developed for all springs. Following the approximate spring shed area and the topographic variable (approximate area/relief) both were related with the maximum spring flow. The relative performance of Chandrabhaga and Danda spring based on all methods is shown in Figure 7. Adjustment of the area and relief were made in such a way that the relationships' efficiency should improve in each adjustment. The analysis of springs was carried out in combination for the area based on their measured discharge. It is evident from Figure 7 that the Chandrabhaga springs 01, 03, 4B, 05, 06 and 13 were found to be relatively good on a scale value of 4 out of 5 as compared to springs 4A, 07, and 10A with a scale value of 1. For Danda watershed, the springs 4A and 28 with a scale value of 5, and springs 4B, 11 and 20 with a scale value of 4 are relatively good as compared to springs 02, 06, 07, 15 and 17.

Figure 7

Relative performance of Chandrabhaga and Danda springs based on all methods.

Figure 7

Relative performance of Chandrabhaga and Danda springs based on all methods.

Close modal

The relationships were obtained after minor adjustment of spring shed area and maximum elevation of spring's shed which are shown in Figure 8(a) and 8(b) for Chandrabhaga and Danda springs, respectively. It is evident from Figure 8(a) that the measured maximum spring flow with maximum elevation and area/relief of spring shed are in close agreement with R2 value 0.84 and 0.64 for Chandrabhaga springs. It is shown in Figure 8(b) that the measured maximum spring flow with maximum elevation and area/ relief of spring shed are in close agreement with R2 value 0.75 and 0.66 for Danda springs.

Figure 8

Spring flow model considering spring shed area, topography and maximum spring flow for Chandrabhaga and Danda springs.

Figure 8

Spring flow model considering spring shed area, topography and maximum spring flow for Chandrabhaga and Danda springs.

Close modal

Springs are the major source of freshwater in the Himalayan region. The present study was conducted to evaluate 33 springs' hydrology (discharge, yield estimation and rejuvenation strategy). The springs were classified using field-based empirical methods which helped to characterize geology, hydrology, climate and land use patterns. The springs were classified and, posteriorly, the relative performance for rejuvenation priorities evaluated. It was found that most of the springs fall in the sixth class magnitude with flow rates ranging from 6.5 to 65.5 m3 day−1 and in the seventh class with flows ranging from 0.8 to 6.5 m3 day−1. The adapted strategies were developed based on hydrogeological investigation and demand-supply model for vulnerable springs. The methodology can be used for quantification of water fluxes to revive mountain springs which can help counter the adverse impacts of climate change.

The authors are highly grateful to the Department of Science & Technology (DST), project ES/11/741/2003 for providing data and an internal study of National Institute of Hydrology, Roorkee Uttarakhand for providing all the facilities to conduct this study. The authors gratefully acknowledge the critical reviews of anonymous reviewers and the editors, which improved the manuscript significantly.

The Supplementary Material for this paper is available online at https://dx.doi.org/10.2166/ws.2019.191.

Agarwal
A.
Bhatnagar
N. K.
Nema
R. K.
Agrawal
N. K.
2012
Rainfall dependence of springs in the Midwestern Himalayan hills of Uttarakhand
.
Mountain Research and Development
32
(
4
),
446
455
.
Hasnain
S. I.
1999
Himalayan Glaciers: Hydrology and Hydrochemistry
.
Allied Publishers Limited
,
New Delhi, India
.
Jeelani
G.
Kumar
U. S.
Bhat
N. A.
Sharma
S.
Kumar
B.
2015
Variation of δ18o, δD and 3H in Karst Springs of South Kashmir, Western Himalayas (India)
.
Hydrological Processes
29
(
4
),
522
530
.
Katsuyama
M.
Yoshioka
T.
Konohira
E.
2015
Spatial distribution of oxygen-18 and deuterium in stream waters across the Japanese archipelago
.
Hydrological Earth System Science
19
(
3
),
1577
1588
.
Kulkarni
H.
Thakkar
H.
2012
Framework for India's strategic water resource management under a changing climate
. In:
Handbook on Climate Change and India
(
Dubash
N.
, ed.).
Oxford University Press
,
Oxford, UK
, pp.
328
340
.
Kulkarni
H.
Vijay
S. P. S.
2014
Groundwater resources in India: an arena for diverse competition
.
Local Environment: The International Journal of Justice and Sustainability
19
(
9
),
990
1011
. http://dx.doi.org/10.1080/13549839.2014.964192.
Lamban
L. J.
Jodar
J.
Custodio
E.
2015
Groundwater isotopic characterization in Ordesa and Monte Perdido National Park (Northern Spain)
. In:
Hydrogeological and Environmental Investigations in Karst Systems
(
Andreo
B.
Carrasco
F.
Durán
J. J.
Jiménez
P.
La Moreaux
J. W.
, eds).
Springer
,
Berlin, Germany
, pp.
245
253
.
Meinzer
O. E.
1918
Bibliography and Index of the Publication of the United States Geological Survey Relating to Ground Water
.
U.S. Geological Survey
,
Reston, VA, USA, Water Supply paper 427
.
Mukhopadhyay
B.
Khan
A.
2014
Rising river flows and glacial mass balance in central Karakoram
.
Journal of Hydrology
513
,
192
203
.
Pant
C.
Rawat
P. K.
2015
Declining changes in spring hydrology of non-glacial river basins in Himalaya: A case study of Dabka Catchment
. In:
Dynamics of Climate Change and Water Resources of Northwestern Himalaya
(
Joshi
R.
Kumar
K.
Palni
L. M. S.
, eds).
Part of Society of Earth Scientists Series
,
Springer Nature, Basel
,
Switzerland
, pp.
151
179
.
Tambe
S.
Kharel
G.
Arrawatia
M. L.
Himanshu
K.
Kaustubh
M.
Ganeriwala
A. K.
2012
Reviving dying springs: climate change adaptation experiments from the Sikkim Himalaya
.
Mountain Research and Development
32
(
1
),
62
72
.
Vashishth
A. K.
Sharma
H. C.
2007
Study on hydrological behavior of a natural spring
.
Current Science
93
(
6
),
837
839
.

Supplementary data