Simulation of Drainage Discharge

Although some of the estimates ( PTF derived dirt hydraulic parametric quantities, unifying irrigation and rainfall informations into one input file, hydraulic caput of aquifer, deepness of impermeable bed, etc. ) were used in this survey, the predicted GWL divergences from the ascertained one are rather comparable to the values obtained by other research workers ( Gupta et al. , 1993 ; Sinai & A ; Jain, 2006 ; Skagss et al. , 2012 ; Yang, 2008 ) . Hence, the theoretical account anticipation of GWL seems to be satisfactory under Fergana status.

5.7.2 Simulation of drainage discharge

Figure 5.30 nowadayss graphical comparings of fake and ascertained day-to-day subsurface drainage flow for Azizbek site ( CHD-2 ) and Akbarabad site ( CHD-3 and CHD-6 ) . Based on this figure, it can be seen that drain flow anticipations extremely deviated from the ascertained values for all the drains considered. Drain flow extremums were ill predicted, i.e. , there are disagreements in the magnitude of some events, particularly during irrigation of an country at one field or both Fieldss relative to the drain outflow point.

It is evident that drainage rate (Q) in DRAINMOD is a map of center H2O table tallness above the drains (m, refer to Equation 4.34 ) that satisfies q=f ( m ) relationship ( Skaggs et al. , 2008 ; 2012 ) . However, multi-tier irrigation within a field and irrigation consequence of two next Fieldss have complicated the q=f ( m ) relationship. In fact, periodical irrigation in 15-20 yearss creates uneven drainage tonss and causes the scattering of a groundwater hill (m) to a non-irrigated country within a field ( Dukhovny et al. , 2007 ) . This produces comparatively of import differences in the infiltration chance clip between upstream and downstream and therefore accession and recession divergences in theoretical account predictedQ.

Field observations reveal that there was a instance when maximal drainage rate has non occurred even GWL between the drains was shallow ( piezometer # 2 in Akbarabad and # # 2, 5 and 8 in Azizbek ) . This leads to miss-estimation of the accession of drainage flow. In add-on, recessions were normally under predicted, intending that the theoretical account did non react every bit rapidly as the Fieldss really did under drain flow recession conditions. Furthermore, H2O table recession ( see Figure 5.29, right ) , due to evapotranspiration ( Yin et al. , 2013 ) , is bit by bit cut downing groundwater flow into the drains ( Kellogg et al. , 2008 ) . Therefore, both the ocular ( Figure 5.30 ) every bit good as the statistical ( Table 5.11 ) comparings indicate that predicted day-to-day drainage rates were in hapless understanding with mensural values. Overall, the theoretical account did non demo marks of consistent scope ( zero and top out discharges ) of drain flow under irrigation status. Separate tally of the theoretical account for each harvest twelvemonth at different field and amount of theQwithin one drain country could be some ground for such hapless statistics. Manual measuring of theQwith 5 to 10 yearss interval may besides be another ground. Furthermore, one drainage in both sites crosses two next Fieldss ( see Figures 4.1 and 4.2 ) , which vary by harvest type and irrigation programming. Hence, irrigation in two next Fieldss, due to upslope consequence and depending on groundwater ( hydraulic ) gradient creates assorted drain H2O interception e.g. , interception is increasing as the hydraulic gradient additions ( Khan et al. , 2003 ) . These all might be influenced on DRAINMOD predicted drainage rate. In general, the amount of theoretical account predicted drainage volume at CHD-2 in Azizbek site was 225 and 370 millimeter, severally during flora period ( April-September ) in 2010 and 2011, severally. This corresponds to 4 % and 19 % less than entire ascertained ( measured ) subsurface drainage volume during that period, severally. Predicted drainage volume at CHD-3 and CHD-6 in Akbarabad site was 168 and 218 millimeter, severally during flora period in 2011. This values about twice ( 48 % ) less than ascertained drainage volume during that period.

Based on observations, coincident irrigation in two next Fieldss increased the designed drainage discharge ( drainage modulus, see Tables 4.1 and 4.2 ) by 2.3-fold in CHD-2 and 2.0-fold CHD-6. The maximal drainage discharge ( 7.2 cubic decimeter s-1) in CHD-3 has corresponded to designed drainage faculty ( 0.6 cubic decimeter s-1hour angle-1) . Hence, drainage extremum has occurred when ~80 % of one field ( comparatively near to run out mercantile establishment into the aggregator ) or ?50 % of the above drain country ( both Fieldss ) were irrigated at the same time in a twenty-four hours. Similar consequences were observed by Dukhovny et Al. ( 2007 ) in Azizbek site. In their surveies, coincident irrigation on more than 50 % of the country between drain spacing increased the drainage modulus by double. Ayars et Al. ( 2006b ) reviewed that the bulk of sub-surface drainage systems across Australia were over-drained and drainage rates exceeded designed drainage modulus.

It is besides of import to observe that Skaggs et Al. ( 2008 ; 2012 ) and other research workers ( Wang et al. , 2006 ; Dayani et al. , 2009 ) simulated hydrology in DRAINMOD is differ by status that they have been considered ( e.g. , humid clime, where daily rainfall is uniformly distributed over the full field ) . Whereas in waterless clime, peculiarly in Fergana part, irrigation, due to uneven distribution over the field, and drawn-out continuance within one event, is complicated the environment that need to be considered in any theoretical account ( including DRAINMOD ) . In add-on, the length and deepness of the subsurface drainage, constructed in survey country, is rather long ( L=500-1670 m ) and deep ( b=2.3-3.2 m ) compared to those constructed in humid part ( for detailed description on drainage design parametric quantities in humid part refer to Ale et al. , 2009 ; Breve et al. , 1998 ; Dayani et al. , 2009 ; El-Sadek et al. , 2003 ; Skaggs et al. , 2012 ) . Therefore, deep drainage consequences deeper H2O flow waies into the dirt profile and deeper GWL that allow increased infiltration from irrigation and uninterrupted drainage flow ( van Beers, 1976 ; Huo et al. , 2011 ) .

Due to run out incline, there might besides be important differences in the harvest irrigation demand across every bit good as between Fieldss ( Ayars et al. , 2006a ; 2006b ) . Dukhovny et Al. ( 2005a ) observed that the caput above drainage pipe along the drains was ranged from 0.5 to 1.4 m, with maximal caput at drain mercantile establishments. Hence, capillary rise could be high at center between the drains every bit good as near to run out outflow point. The latter, nevertheless is controlled by H2O degree at having aggregator.

At first glimpse, the DRAINMOD simulated hydrology seemed to be acceptable, nevertheless, literature about comprehensive analysis of DRAINMOD predicted hydrology ( comparative analysis of GWL and drainage discharge, together ) were light. The group of research workers ( except Helwig et al. , 2002 ; Skaggs et al. , 2012 ; Tian et al. , 2012 ) either vis-a-vis compared theoretical account foretelling GWL ( Gupta et al. , 1993 ; He et al. , 2002 ; Sinai & A ; Jain, 2006 ; Yang, 2008 ) or drainage discharge ( Bechtold et al. , 2007 ; Dayani et al. , 2009 ; El-Sadek & A ; Vazquez, 2012 ; Fernandez et al. , 2007 ; Luo et al. , 2010 ; Salazar et al. , 2008 ; Singh et al. , 2006 ; Wang et al. , 2006 ) .

Based on the consequences ( including extended literature reappraisal ) , it can be concluded that better public presentation of the theoretical account simulated GWL ( Table 5.10 ) does non connote that best consequence in drainage discharge ( Table 5.11 ) can be achieved under the same environment, peculiarly in irrigation status. It was found that by changing the theoretical account input informations ( particularly, hydraulic caput of an aquifer, that performed better H2O table simulations ) , resulted different dirt H2O volume drained into drains ( informations is non shown ) . Furthermore, drainage strength ( cm twenty-four hours-1) in irrigated Fieldss vary depending on day-to-day rate of irrigation country covered with each twenty-four hours irrigation ( IAC ) above the drain. Analysis of Dukhovny et Al. ( 2005b ) showed that formation of groundwater hill (m) as a consequence of alterations in the flow form associated with IAC has little consequence on drainage flow (Q) . Despite on above mentioned restrictions, there are still high possible utilizing DRAINMOD to foretell GWL every bit good as drainage discharge in waterless parts, which may ensue in minor divergences from ascertained values. For that, ( 1 ) the drain, under consideration is needed to be served merely one field ( comparatively little field country ) with homogenous harvest type ; ( 2 ) irrigation in a field is needed to be applied at the same rate within a twenty-four hours whereas prolonged irrigation ( e.g. , irrigation continuance & gt ; 1 yearss ) consequences different H2O burden into a drain, therefore Hooghoudt equation [ refer to Eq. 4.34 ] does non depict the q=f ( m ) relationship for the full drainage period ( Skaggs et al. , 2008 ; 2012 ) and ( 3 ) sidelong flow is needed to be in a steady-state status ( Hooghoudt equation ) , e.g. , the cross-section of H2O flow country into the drains ( see shaded country in Figure 4.10 ) is changeless for the full drainage line over the period. However, the latter is rarely happening in pattern ( van Beers, 1976 ) . Further research is needed to understand the behaviour of groundwater flow into sub-surface drains under irrigation status ( including precipitation and belowground H2O recharge ) . Field probes with installing of machine-controlled devices ( with hourly measurings ) , such as Odyssey ( Dataflow systems Pty LTD, New Zealand ) or RDS ( Remote Data Systems, Inc. , Wilmington, NC ) for GWL monitoring, NMM ( neutron wet metre, Campbell Pacific Nuclear Intl, Martinez, CA, Evett et al. , 2007 ) or Em50 Series Data Collection System ( Decagon Devices, Inc. Pullman, WA ) for dirt wet monitoring every bit good as tipping pail device ( Tait et al. , 1995 ) for irrigation influx ( including IAC ) and drainage outflow measurings can manage that.

Figure 5.30: Comparison of the ascertained and predicted drainage rate ( cm twenty-four hours-1) affected by irrigation patterns conducted at ( a ) C-13 & A ; C-15, ( B ) C-165 & A ; C-172 and ( degree Celsius ) C-174 & A ; C-176. ( Pi: predicted drainage ( summed from all harvest old ages from two Fieldss ) ; Oi: ascertained drainage ; Irr: irrigation of several Fieldss ; Pre: precipitation )

Table 5.11: Statistical comparing between observed and simulated day-to-day drainage discharge ( cm twenty-four hours-1) for CHD-2 ( plots C-13 and C-15 ) in Azizbek site every bit good as CHD-3 ( C-165 and C-172 ) and CHD-6 ( C-174 and C-176 ) in Akbarabad site during 2009-2011

Time period

Drain

William claude dukenfields

N

R2( – )

RMSE ( centimeter )

MAE ( centimeter )

Tocopherol ( – )

vitamin D ( – )

10.2009-10.2011

CHD-2N’

C-13a„¦, C-15

150

0.43

0.112

0.076

0.403

0.762

07.2010-10.2011

CHD-3

C-165, C-172

97

0.41

0.118

0.079

0.333

0.707

07.2010-10.2011

CHD-6

C-174, C-176

94

0.33

0.172

0.108

0.075

0.584

the finding coefficient ( R2) from additive arrested development ;

N’closed horizontal drainage ( CHD ) ;

a„¦field contours ( C ) ; Other abbreviations are same as described in Table 5.10.