USVI Public Records

Document text

EFFECTS OF VARIOUS FACTORS ON THE SIZING OF RAIN WATER CISTERN SYSTEMS Henry H. Smith Project No. 375006 Agreement No. 14-08-0001-G-875 September 1983 [ [ [ [ [ [ i The research on which this report is based was financed in part by the United States Department of the Interior as authorized . f by the Water Research and Development Act of 1978 (P.L. 95-467). ~@ i) Technical Report No. 19 PrP Caribbean Research Institute College of the Virgin Islands St. Thomas, U.S.V.I. 00802 3 [ [ DISCLAIMER Contents of this publication do not necessarily reflect the views and policies of the United States Department of the Interior, nor does mention of trade namés or commercial products constitute their endorsement by the U. S. Government. -ii- ABSTRACT The sizing of rain water cistern systems is properly done by con- sidering several factors. Among these are rainfall frequency, duration and magnitude, demand on the system, characteristics of the catchment area, and the reliability level desired. Models developed in this study consider these factors. Results of simulation varying all of these fac- tors show the changes in volume storage which must be provided. The models demonstrate their usefulness in the proper sizing of cistern systems to satisfy the peculiarities of specific situations. [ -iii- Assistance in the analysis of data in this study as well as in the development of the models was provided by Mr. Clement Browne, the. student research assistant at the Caribbean Research Institute. Mr. Browne's persistence, constancy and willingness to do more than re- quired positively affected the outcome of this study. -iv- [ [ [ j f [ [ [ i [ [ [ [ [ [ [ [ [ [ TABLE OF CONTENTS ABSTRACT ACKNOWLEDGEMENT LIST OF FIGURES LIST OF TABLES INTRODUCTION HISTORICAL BACKGROUND QUALITY CONSIDERATIONS QUANTITY CONSIDERATIONS MODEL DEVELOPMENT AND APPLICATION CONCLUSIONS AND RECOMMENDATIONS REFERENCES APPENDIX A - Design Curves for Sprat Hole St. Croix, U.S.V.I. --------- ; APPENDIX B - Listing of Models Developed - E iv ounef f 10 29 30 AP1 AP9 re LIST OF FIGURES . FIGURE 10. ll. Typical Cistern in the U. S. Virgin Islands ------ Computation of Minimum Feasible Required Storage-- Flowchart: of CISTERN RELIABILITY 21-Year Weekly Rainfall Averages at Sprat Hole, St. Croix, U.S.V.I. Design Curves for Sprat Hole, St. Croix for Several Demand Levels with 85% Catchment Efficiency ------ Average Monthly Rainfall in Tortola, B.V.I and Anguilla, B.W.I Average Monthly Rainfall at George Town, Grand Cayman, British West Indies and Koror, Belau ----- Design Curves for Anguilla, B.W.I. with 85% Catchment Efficiency for Several Demands Design Curves for Grand Cayman, B.W.I with 852 Catchment Efficiency for Several Demands --------- Design Curves for Koror, Belau, with 85% Catch- ment Efficiency for Several Demands Design Curves for Tortola B.V.I with 85% Catchment Efficiency for Several Demands - 17 20 21 23 24 25 26 27 28 LIST OF TABLES TABLE Page 1. Relative Efficiences of Various Foul Flush Devices w2woenn 11 2. Catchment Efficiencies Used in the Literature j [ [ -vii- [ \ a 3 ga: a INTRODUCTION Rain water cistern systems are a widely used but not very heralded supply of water throughout the world. These systems have for centuries been the only source of water in some areas and are once again being turned to as a supplemental source where water demands have increased and where modern technology, largely on account of the associated expense, has failed. While the original intentions of this study were abridged because of tightened time constraints, the investigations described would be useful in the development of guidelines for the utilization of cistern systems either individually or in conjunction with supplemental sources of water for domestic use. A short historical overview of rain water cistern useage is made followed by an examination of a few of the quality aspects which must be considered with rain water cistern systems, An in depth investigation then is made of several factors affecting the sizing of - these systems. These factors are the expected demand on the systems, the required reliability, the rainfall amount and distribution and the catchment efficiency. [ [ [ j [ [ [ [ [ [ [ — 9 HISTORICAL BACKGROUND While no one can document the earliest cisterns, their usage have been recorded as far back as 2000 B.C. in the Middle East where typical middle class dwellings stored rain water in cisterns. Throughout history, the use of these systems have continued. In the eighth century the Greeks constructed houses with cisterns and Hipprocrates advocated boiling water for disinfection as did the Egyptians in Alexandria where _ prominent families utilized cisterns. In ancient Turkey, cisterns were included in the water works for as far back as the second millenium B.C. during the Hittite Period. Cisterns were the principal source of water in elevated locations in Anatolia and several other cities, and were constructed to serve as | emergency sources in case of warfare. In the Yucatan peninsula, recent discoveries have shown that cisterns were in use during what is refered to as the Maya Period, beginning about 300A.D. Cisterns later played a vital part in the development of Mayan culture and in the lives of the present inhabitants of the Yucatan peninsula, ! | In the eastern United States, during the early settlement of the country, rain was collected from rooftops and used in households. In the early part of the 19th century a law was passed in New York to construct public cisterns to collect rain from public buildings for use in fire protection. In Hawaii, rainfall catchment has long been a source of domestic and livestock water supply. The earliest report on rain catchment in = = Hawaii (1915) estimated an average catchment area of 100 square feet per person with an average tank capacity of 1000 gallons per person and an average water use rate of 10 gallons per person per day.” Throughout the Caribbean, cisterns are often the principal source of water and in some areas are required by law. In the Cayman Islands cisterns provide 68% of the household water. 3 In the United States Virgin Islands and the British Virgin Islands all residential buildings must provide at least ten gallons of cistern storage for each square of roof area. In Bermuda, a minimm of ten gallons of elisvers storage must be provided for at least 80% of the roof area. In Bermuda cisterns provide approximately 1.6 million gallons of water per day and ‘in the U.S. Virgin Islands they serve as the principal source of water for over 75% of the residents. Cisterns in the Virgin Islands generally form an integral part of residental structures. Though often many innovative and imaginative approaches have been used in the design of these cisterns in incorporating them into buildings, the practice generally is to have them form a part of the foundation. Typically they appear and are constructed as shown in Figure 1. Number of Students ICWRM CURRENT STATUS STUDENT ENROLLMENT Academic Year 100 {00-- go ao 77 70+ ml 57 50-4. 42 40+ 35 30+- 25 20 10 40 | 10+ eres + aaes * asso | oyse * sea sa/ea * aa/e5 Projected fl] Actual [ [ [ [ [ ae) [ [ [ [ [ ur a : GUTTER Ejez06 =| DownsPour Tr | CISTERN (NLET 1 > gq OUTLET WITH SCREEN —&--—-— aL 22'-0" le. fo'—o" . 1o'~o" A f------- . TT =a eer 4 a "Bh et te . Pay 7 eexaa) ~ — ~ 7 TT shlr ob tat aie 'T" lleover! : & < COVER I 1. yy! 3 a 1 Low eat 6g} FS fe 1|_|#40/6"0.¢. To? s7RAIGHT _|- Fi. ele Wy T ¥ Th T wa T z= | §] ly “ a} & wy +] P| j|teeato.c.aw. aorrom If tl 8 3 _- ry) Std ; he, « | s] 8 | ld R16"0.6. Tavss@ %5 SPAN| | 3 ar S He | Wee LY mee ae FT | | | Ss 2 1 | | tad B3 Vy. yy i + g 7-3 Ye ee ee . ‘. at A L., —\e ee eer re ee eer eel el ll 4 ( 3S INLET 4! ovreer CI ST; STE PL y¢4- DOWELS #48 BY O.c. B.w. ONE 4" Sthecwen BRS Sz, OTH SIDE Figure 1 Typical Cistern in the U. S. Virgin Islands (Not to scale indicated) [ [ [ [ [ ' [ ' [ [ [ [ [ [ [ [ QUALITY CONSIDERATIONS Collected rain water may be contaminated from a myriad of sources. As expressed by Rinehart: Cistern water contains atmospheric dust and aerosols, accumulated dust and debris from roofs, breakdown products from roofing materials, organic debris from trees, micro-organisms, fecal material from rodents, birds and lizards, and salt deposited from sea spray. During storage, it interacts with cistern walls. Frogs.may visit or reside in the cistern. Under special circumstances, zthe cistern itself can be subjected to ground water seepage. ‘While no recent study has been made of rainfall quality in the Virgin Islands, the potential for acid rain and dry deposition as described by Hicks does exist.” Pesticide contamination of Virgin Islands cistern water has been documented in a report by Lenon, Curry, Miller and Patulski.© Of greater importance than the quality of the rain, is the quality of the water when it is to be put to use. This quality may be influenced by the characteristics of the roofing material used, such as its roughness and likelihood of retaining pollutants, the paint or other | agent used to coat the roof, its slope, the proximity of trees, birds and other sources of deposits on the roof. ) To enhance the quality of the harvested water before it enters the cistern, various strategies for diversion of the initial five to ten minutes of highly polluted foul flush are practiced. Jenkins and ~~ J “J 9 om) Pearson ’ have presented a very good analysis of this foul flush. These strategies are said to for a small investment yeild a marked improvement in the quality of water in storage and are often recommended over filters which tend to clog and become contaminated resulting in relatively high maintenance requirements. Table 1 based on information 8 from Michaelides and Young’ indicates the relative efficiencies of ten foul flush separation devices in preventing soluble and particulate pollutants from entering storage. . Jenkins and Pearson” and Young and Sharpe? has detailed elaborate and very effective procedures to treat cistern water. Canoy and epee a Knudsen’~ recommend a far simpler procedure: Every household with a cistern water supply should practice chlorination for their cistern storage water; sufficiently - strong solutions of chlorine such as 'Clorox', should be added to the cistern each night so that the residual total chlorine the next morning is on the order of 0.5 mg Cl /L or greater, based on measurements with a DPD chlorine — colorimetric test kit. New cistern water supply systems should be required to be constructed to filter debris and to facilitate the addition of chlorine to the storage tanks. . One way to accomplish this would be to provide plastic piping which would allow the addition of chlorine from the main part of the house, and which would preferably distribute the chlorine to several locations within the tank. A water depth indicator in the cistern supply tank would also be desirable to aid in determining how much chlorine is needed. It is realized that the addition of chlorine in this manner will increase the chlorinated organics content of the cistern water supply. However, in the opinion of the authors, the health risk of the increased concentrations is small when compared to the risk, of promoting entric disease if the chlorine were not added. It is interesting to note that a study by Coffin and Richardson! for the Virgin Islands Water Resources Research Center revealed that from a survey conducted, 68.4 percent of the respondents used cistern water as their principal source of drinking water and 65 percent of- -6- 4IhITo.SAe es TTF Ds a I) a ST SI) ea AF a aS a.) a a as aS) a) TABLE 1 RELATIVE EFFICIENCES OF VARIOUS FOUL FLUSH DEVICES ‘ Movable | Movable Removable] Diversion] Simple F 4 ippi i i’ Tank Flush Box | Flush Box’ | #26atfoh°at] Box with") RazR'"9) brvagh"s Downpipe) Tank Valve . h Box F Box Covers with Drain | with Tric- | Simple Foul] Floating Funnel Tap kle Drain Flush Box Ball Operation Manual | Manual Manual Manual Semi-Auto- | Automatic Semi-Auto- Semi-Auto}| Automa-{ Automatic : matic matic or matic or tic Automatic Automatic Enables Re- gection of . an Initial xX xx XX xx 00 00 00 XxX XX XX Flush of . Soluble Pol- lutants Reduces Un- desirable . Flows of xx XX XX XX To Some Extent Suspended . if Maintained XX XX XX XX Particles , these respondents felt that they would choose cisterm water over all other sources if they had no constraints. 51 percent of the respondents said that they choose not to boil their drinking water. rr rr a i nr QUANTITY CONSIDERATIONS The five major design factors for cistern systerns are: i. The amount and distribution of rainfall available for ii. The amount and characteristics of the catchment area, ii. The water demand. iv. - The storage capacity. v. Economics. Large cisterns are costly. The cost of materials for construction of the cistern illustrated in Figure 1 is approximately $5000.00 . However in many areas cisterns may be the most reliable or the only source of water. In such areas, water is very dear and is used accordingly. | The designer of the cistern has no control on the demand that will be exerted on the cistern system by the user. Nor does the designer control the amount and distribution of rainfall or to a significant extent the size of the catchment area. What the designer must concentrate on is the optimm storage capacity when the other factors are considered. Schiller and Latham’? reviewed several methods which were modifications of the popular Rippl mass curve method. The methods they reviewed included the mass curve based on hydrologic data, the yield after storage (YAS) method, the rationing and stocking model and Ree's statistical method. In addition to these methods, several others were presented at the First and Second International Conferences on Rain Water Cistern Systems and are detailed in the proceedings for these conferences. 14, 15 MODEL DEVELOPMENT AND APPLICATION In the present study, two simulation models were developed to -examine the effects of various parameters on the cistem capacity. While several approaches to this problem have been taken in reservoir sizing studies, because of limitations of the computer facilities available for this study, model development was forced to consider computer storage limitaitons. Particularly benefical in developing a model with these constraints is that the end result will be a product that is usable to others with similar constraints. ‘The model presented was developed using an Apple II micro-computer. A first approach to the model then would consider for each time period the rainfall input (RA), the volume of water harvested (HA) which is fimction of the amount of rainfall, the catchment area (CA) and the catchment efficiency (CE). HA= RA (CA x CE) (1) Catchment efficiency is a term that accounts for several factors which may affect the eventual discharge of the rain that comes in contact with the roof surface. Among these are the catchment slope, the degree of imperviousness, and the roughness. Various coefficients have been used throughout the literature to serve the same role that the catchment efficiency factor does here. Several of these are listed in Table 2. . From the harvested rainfall of Equation (1), the weekly demand (DM) is subtracted. This procedure of subtracting the demand from the harvested amount is similar to the Jenkins and Pearson's yield before -10- { es ne ee: ee l" r TABLE 2 Catchment Efficiencies Used in the Literature Catchment Source Efficiency 0.70 Waller and Inman, 198226 0.70 Piggot, et al, 198217 0.90 Pompe, 198218 0.75 Waller, 198229 0.90 Rakocy, 198429 0,750.80 Buros, 197621 -1l- [ [ [ [ [ [ [ [ ' ) [ [ [ [ storage (YBS) model.2* Alternatively the harvested rain could have been added to the amount in storage where spillage would occur if necessary and then the demand satisfied. This approach is analogous to Jenkins -and Pearson's yield after storage (YAS) model. 29 In the model being developed demand is first satisfied from the harvested rain, the excess rain added to storage and then spillage, if any, takes place. If the harvested rain does not satisfy the demand, the excess demand is added to storage. Jenkins and Pearson@" also proposed a yield and rainfall distributed (YRD) model for use in estimating day by day system performance. For the week by week simulation in the model being developed, the approach similar to the YBS model seemed more suitable. At each time step the volume of water remaining in storage is determined as well as whether or not the demand was satisfied. . The cycle is repeated for each week throughout the simulation period. The reliability of the system is defined to be the fraction of the simulation period that the system will meet demands. Each time the system 'fails" (does not meet demand) a defecit (DE) occurs. Reliability (DR) is calculated as: DR< 1-[DE/ (NY*52) ] (2) DR = desired reliability DE = cummlative occurances of deficits in the simulation period NY = number of years of record -12- ee -and Pearson's yield after storage (YAS) mode storage (YBS) model.2* Alternatively the harvested rain could have been added to the amount in storage where spillage would occur if necessary and then the demand satisfied. This approach is analogous to Jenkins 1.23 In the model being developed demand is first satisfied from the harvested rain, the excess rain added to storage and then spillage, if any, takes place. If the harvested rain does not satisfy the demand, the excess demand is added to storage. Jenkins and Pearson“* also proposed a yield and rainfall distributed (YRD) model for use in estimating day by day system performance. For the week by week similation in the model being developed, the approach similar to the YBS model seemed more suitable. At each time step the volume of water remaining in storage is determined as well as whether or not the demand was satisfied. . The cycle is repeated for each week throughout the simulation period. The reliability of the system is defined to be the fraction of the simulation period that the system will meet demands. Each time the system "fails'' (does not meet demand) a defecit (DE) occurs. Reliability (DR) is calculated as: DR< 1-[DE/ (NY*52) J (2) DR = desired reliability DE = cummilative occurances of deficits in the simulation period NY = number of years of record -12- At the end of the specified simulation period, the reliability of a system with a specified roof area, demand rate and cistern capacity is determined. Since it may be desirable to determine the effectiveness of various storage capacities, a procedure was incorporated into the model that pemmits a user to specify a reliability and demand and the model will determine the corresponding storage, if possible, that would meet these requirements. | Additionally, at times it may be desirable to determine for a specified demand the storage that would be required to meet several reliabilities. Alternatively, for a specified reliability, the storage which must be provided to meet various demands may be desired. The first case was addressed in a variation of the original model called CISTERN DEMAND and the second case in a similar variation called CISTERN RELIABILITY. As work progressed with the models it was determined that modifications could be made to reduce computational time and steps. The first of these modifications would halt execution in the simulation period as soon as the minimm desired reliability was not met. This procedure is best illustrated by using an extreme case as an example. Assume that the unrealistic reliability of 100% is specified for | a 10 year similation period. If in the second week of the third year the system fails then there is no need for the simulation to proceed further. In this example, DR in Equation (1) is 1, and NY is 10. When the system fails for the first time, DE becomes 1 and Equation (2) cannot be satisfied. With other specified reliabilities DE may increase until Equation (2) can no longer be satisfied. -13- The model then increments the specified minimmm storage and the simulation begins again and continues either to the end of the similation period or until the desired reliability cannot be met. The incremental increase in minimum storage (gallons/square foot of ‘roof area) was used as 0.5. This may be changed as the model user prefers. A maximm limit of 30 gallons storage/per square foot of roof area has been placed in the model to prevent unrealistic determinations. Storage is not always the limiting factor in rain water cistern systems. While impressive volumes of storage may be provided, unless the rainfall and demand are such that the storage is at times all used, then a wasteful excess of storage has been provided. In the simulation model, if for the simulation period a specified reliability has not been met, unless during the period some spillage has occured then it is useless to provide additional storage and go through the simulation again. In the model each time the storage volume is exceeded and a spill occurs, a variable called ''SPILL'' is set to a non-zero value. At the point in the simulation where it is determined that the desired reliability cannot be met, a check is made to see if SPILL is non-zero. If it is, then the storage is increased and the simulation repeated. If it is zero, then the desired reliability camnot be met for provision of additional storage will not result in any water savings. , Previously, it was explained that if in determining the minimm required storage to meet some capacity, it became apparent that the required storage would be greater than or equal to 30 gallons per square foot of catchment area, the computations would cease. Provisions were also made in the model to discontinue calculations when the difference between required storage for two consecutive reliabilities (or demands) -14- exceed five gallons per square foot of catchment area. This provision was also made to curtail unrealistic determinations of cistern capacity. In the model, when the minimm required storage for a series of increasing reliabilities is being determined, it is obvious that for a reliability level above the next lower level, the required storage is at a minimm the required storage of the lower level. More significantly, there will be a minimm increase in storage between incremental reliability levels equal to at least the differences in storages corresponding to the previous two consecutive reliability levels. To illustrate this, (Figure 2), let S,; and S, be the minimm required storage which must be provided to obtain respective reliabilities of Ry and R,. Let S5_, be the difference between S, and S S, then is a 1 minimm of S, + S,_)- A flowchart of the complete CISTERN RELIABILITY model appears in Figure 3. Listings for CISTERN RELIABILITY and CISTERN DEMAND are to be found in Appendix B. The model was used with 21 years of weekly rainfall data from Sprat Hole, St. Croix, U. S. Virgin Islands. The weekly averages of this data are as shown in Figure 4. For catchment efficiencies ranging from 0.70 to 1.00, the cistern storage capacity necessary to provide water over -the historical period for demands ranging from 500 gallons to 1200 gallons per week were determined. Reliability of these systems ranged from 50 percent to 95 percent. The catchment area remained constant at 1600 square feet. The results obtained for a catchment efficiency of 85Z are presented in Figure 5 and for the other catchment efficiency levels in Appendix A. -15- onan = = CISTERN SAND FILTERS REMOVABLE COVER DOWNSPOUT e- COVERED WITH METAL SHEETING a i eee —~ SS | FLAPPER VALVE OVERFLOW “4 & 6 N va REMOVABLE META id eo WASTE WITH +" HOLES ee a:& - be —_—-: eat] ee ee et 6 ‘e, » * & fo APPROVED FureR SAND Alt a > a '¢@ =! — tee fT i 44 i} 4 3 ~f. FINE GRAVEL _______. - ote 4 * is a a e". 3" COARSE GRAVEL a 4 - ~T a a ae SO ae San Cet” <® -—=* ais a & a’ TS Seale of am. * es ft [Sa i ] \ Ava C.4, WALA C4I7/NG 4"C.1.PIPE 7 %. a ~/eAaubKED 4EAD_SOINT —-— yr —>_ a— —_— 4 inte - a fe CISTERN aa’ a ee a’® a ~ -_ — 2g. a a > ‘ay 1B) PROVIDE ONe SQ.FT, OF Ve FILTER SURFACE FOR EACH 100 SO. Fr. OF ROOF ARBA. td SAND FILTER MAY BE USED IN PLACE OF ROOF WASHER WITH APPROVED CISTERN NoT fo SCALE 128 DISTRIBUTION OF TYPES OF DEVELOPMENTS Figure 44. GALLIA COUNTY Spring Pond & Other Drilled 9.0% 30.1% 0.4% b Driven 0.8% Cistern Dug — 43.6% 16.1% Table XIX. COMPARISON OF DEVELOPMENTS BY % SAFE % Safe # of Systems Sampled Drilled 73.5 294 Driven 50.0 8 Pond & Other 50.0 4 Dug 32.9 425 Cistern 32.5 157 28.4 88 Spring $ 81 ip Minimum Required Storage wn By n uo Computation of Minimum Feasible Required Storage -16- min S9-} I T a $2-1 Min So + ( +. i J I I |__.. Ro Ry ) RQ R3 R4 Reliability Figure 2 s§ 2-1) FLOWCHART OF CISTERN RELIABILITY = in Rainfall Sequenot 145 675 Run No Yes Input Minimm and or several Maximum Demands | Demands [input Demand ' Input Catchment Area Efficiency and Mini mum Required Storage 475 No Run Yes Input Minimum and or Several eS Maximum Rel iabili: i Input Reliability | Start of Loop 107 t Determine Storage Volume SV 225 Computes Rain Harvested and Satisfies Demand 240 { i cs Computes Cumulative Storage Figure 3 r -17- a ' @ 245 | No es,1CS=SV. SPILL! 250 i es =| [Bseitiea . 0 M &0 ° a A Ww u ed Ww % 255 lo No Reliability, Yes in Violated Ne} Fo ° + No torage Yes 0) gal/£t2 No pillage yes pm 2 Occurred i ' 245 i 530 No first: time Yes = in Loop R ired Yes equire Increment Stor. torage un- 2 565 (Flowchart of CISTERN RELIABILITY Cont'd) Go to 210 (Beginning of Loop) Higher reliability Unlikely EN? -18- (Flowchart of CISTERN RELIABILITY Cont'd) 585 No unning Yes for Several Demand 590 690 Rerun Demand INO Program OXIA Deman ? [ 605 695 Reset all Variables Increment Demand Reset Appropriate ' Variables 615 END Note: Numbers on flowchart refer to statement numbers in the program. [ -19- Average Rainfall (Inches) } 2.0L 1.54b 1.0- O.5F 0.0 A A 4 ! 13 26 39 52 Week Figure 4 21-Year Weekly Rainfall Averages at Sprat Hole, St. Croix, U.S.V.I -20- 12 > 10 vo MH Css} ay 8 u 700 ‘o 8 YS uy ~ 0 ed (39) 2 6 650 5) oD [9] fH fo} Ww on ci 600 a 4 pa} a 3 550 0 500 u rt o & 2 0 1 1 rn 1 ny ry N rn ad 50 60 70 80 90 % Reliability Figure 5 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with 85% Catchment Efficiency -21- [ [ [ [ [ [ [ [ ' [ J i To illustrate the effects that rainfall quantity and distribution might have on desirable storage volumes, the models were applied to monthly rainfall data from Tortola, British Virgin Islands,”6 snguilla, British West Indies, 2’ and Grand Cayman, British West Indies,2° all in the Atlantic Ocean. Also the models were applied to monthly rain data from Koror, Belau2? in the Pacific Ocean. Average monthly rainfall for these sites are plotted in Figures 6 and 7. In these simulations a catchment area of 1600 square feet with a catchment efficiency of 852. Several monthly demand levels were used. Results are presented in Figures 8,9,10 and 11. ~ 3s “Q 7S GQ TZ Average Monthly Rainfall (Inches) Anguilla Tortola Jan Feb Mar Apr _ May Average Monthly Rainfall in Tortola, B:V:I.and Anguilla, BiwW.I. Jun Jul Figure 6 -23- [ Grand Cayman 15.0 Belau 10.0 [ BEE] Sa 5.0 ES al ped Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 7 Average Monthly Rainfall at George Town, Grand Cayman, British West Indies and at Koror, Belau -24- Required Cistern Storage (Gals./ft? of roof area) 12.0 2750 10.0 8.0 2500 6.0. 2250 4.0 2000 2.0 0.0 i Li 50 60 70 80 90 4 Reliability Figure 8 Design Curves for Anguilla, BWI with 85% Catchment Efficiency For Several Demands -25- | 3500 ‘3000 2500 2000 Required Cistern Storage (Gals./£t?2 of roof area) 50 60 ' 70 80 90 % Reliability Figure 9 Design Curves for Grand Cayman, BWI with 85% Catchment Efficiency For Several Demands -26- Required Cistern Storage (Gals. /£t of roof area) A 10000 9500 10.0 L 8.0 LL. 9000 8500 6.0 L 8000 750) 4.0 7000 2.0 | —<-- rT A i — 50 60 70 80 90 % Reliability Figure 10 Design Curves for Koror, Belau with 85% Catchment. Efficiency for Several Demands ~27- ee | | “9 a 3° 9 2 Required Cistern Storage (Gals./ft of roof area) 3750 pt 12.0 r 10.0 oa 4000 3500 8.0 5 3250 6.0 ~— 3000 4.0 . 2750 2500 2250 2.0 a 2000 0.0 ~—y = : i a iT ~~ 50 60 70 80 90 % Reliability Figure 11 Design Curves for Tortola, BVI with 85% Catchment Efficiency -28- [ i [ [ —¥ [ [ [ i; [ [ [ [ ' CONCLUSIONS AND RECOMMENDATIONS The models developed in this study are capable of determining the minimm cistern storage capacity that would be required to satisfy a specified demand at a desired reliability level. Required inputs are a rainfall sequence of a chosen length, the area of the catchment surface, and a coefficient relating to its efficiency in transmitting intercepted water. , . These models may be used in several ways. Planners and decision makers can use these models in formulation of rules and regulations pertaining to cistern sizing. Alternatively, for cistern systems already in operation, these models can assist in determination of current storage levels. Such information is critical when conjunctive use strategies apply. | Further work on these models may include provision for changes in demand with changes in storage levels. Such a provision would eliminate the possible inaccuracies that may occur when an average demand ,is utilized. Additionally, it may be desirable to increase demand as time progresses in keeping with the assumption that the standard of living will improve with time as will water use. -29- REFERENCES 1. Reid, George W., "Lessons of History in the Design and Acceptance of Rain Water Cistern Systems,"" in Proceedings of the International Conference on Rain Water Cistern Systems, Honolulu, Hawaii, June 1982 (Ed. Faith N. Fujimura). 2. Wentworth, C.K., "Rainfall, Tanks, Catchment and Family Use of Water,'' Hawaii Water Authority, Honolulu, Hawaii, 1959. 3. Beswick, R.G.B., "Water Resources Situation Report - Cayman Islands ,"" Mosquito Research and Control Unit, Barbados, B.W.I. 1980. 4, Rinehart, F., et. al., "Water quality of Cistern Water in St. Thomas, U.S.V.1I.," Technical Report No. 15, Caribbean Research Institute, St. Thomas, September 1983. 5. King, Thomas, L. and P.B. Bedient, "Effect of Acid Rain on Cistern Water Quality," in Proceedings of the International Conference on Rain Water Cistern Systems, Honolulu, Hawaii, June 1982. 6. Lenon, H., L. Curry, A. Miller and D. Patulski, ‘Insecticide Residues in Water and Sediment from Cisterns in the U. S. and British Virgin Islands," in Pesticides Monitoring Journal, Volume 6, Nunber 3, pp 188-193, December, 1972. 7. Jenkins, D. and F. Pearson, "Feasibility of Rain Water Collection Systems in California,"’ Contribution No. 173, California Water Resources Research Center, University of California, Berkeley, 1978. 8. Michaelides, Georghios and Robert J. Young, "Protection of Water Quality from Roof Catchments by Appropriate Design and Maintenance," in Proceedings of the Second International Conference on Rain Water Cistern systems, St. Thomas, U. S. Virgin Islands, Jume, 1984 (Ed. Henry H. Smith). 9. See Supra, Note 7. 10. Young, Edward S. and William E. Sharpe, "Rainwater Cisterns Design, Construction and Water Treatment,"’ The Pennsylvania State University, University Park, Pennsylvania (undated). ; ll. See Supra, Note 4. 12. Coffin, and Richardson, Inc., "Water Conservation Under Conditions of Extreme Scarcity: The U. S. Virgin Islands," Technical Completion Report for U. S. Department of the Interior, Office of Water Research and Technology, Boston Massachusetts, December, 1981. 13. Schiller, E. J. and B. Latham, "Computerized Methods in Optimizi Rainwater Catchment Systems," in Proceedings of the Internationa Conference on Rain Water Cistern Systems, Honolulu, Hawaii, June, 1982 (Ed. Faith N. Fujimura). 30 { ! [ APPENDIX A [ gee | i [ DESIGN CURVES FOR SPRAT HOLE, ST. CROIX, USVI [ WITH CATCHMENT EFFICIENCIES FROM 70% to 902 } mpg | | -AP I- 4 Qa “” 7 —y “939 7 7 i 850 650 550 500 Required Cistern Storage (gals./£t? of roof area) 50 60 70 80 90 % Reliability Figure A - 1 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 70Z -AP 2- l 12 + 800 . 00 750 e cy) te] U4 ° [e) M4 Get fe) N 8 Fw) U4 ~ a e w 2 yo © 60 % wu ° v nN . o 4 Ww n ro oO ae] a ha 3 2 o ov % 0 50 60 70 80 90 4% Reliability Figure A - 2 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 75% -AP 3- 12 700 650 600 550 500 Required Cistern Storage (Gals./ft* of roof area) 50 60 70 80 90 4 Reliability Figure A - 3 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 80% “AP -4- Sant i; 12 [" o~ lr’ co) fa « 10 im Get [" ° o ui Ut l ° N vp 8 Ue ~ ia ° l" é et bY Oo i ww oo 0 +S) lr" ° Ww wn & Ww n ord oO ~ ) Mt | 3 2 ou. ov [a2] 0 = at a a | " i i d. y 50 60 70 80 90 % Reliability Figure A - 4 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 85% \ 1000 950 ,900 800 850 | 750 700 650 600 , 550 0 i 1 n 1 1150 1050 10 Required Cistern Storage (Gals. /£22 of roof area) a L | > 50 60 70 80 90 % Reliability Figure A - 5 Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 902% “~AP 6- “9 of roof area) 2 Required Cistern Storage (Gals./ft 12 10 800 750 700 650 600 550 500 50 60 70 80 % Reliability Figure A - 6 Y Design Curves for Sprat Hole, St. Croix for Several Demand Levels with Catchment Efficiency of 952 -AP 7- 12 “- 3 o M4 3 10 Gel ° fo) MY Gey ° ~ 8 pa) Leng ~ ) 4 we io) a w 6 00 « u fe) wu wn c u a 4 Ww Q oar oO ~~ o u ae 3 = 2 o [= 0 Design Curves For Sprat Hole 1000 1100 1050 A. A ry 4. 4. 950 900 850 800 750 700 650 600 550 500 50 60 70 80 % Reliability Figure A - 7 Catchment Efficiency of 100% 90 St. Croix for Several Demand Levels with ee ieee eee: eee eee tae ee ee eee ieee ioe eee eee Bee ee lone Oller ieee Bites APPENDIX B LISTINGS OF CISTERN RELIABILITY AND CISTERN DEMAND -AP 9- [ f [ [ 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 4110 115 4120 © 125 130 135 140 145 150 155 160 165 170 175 180 185 190 195 200 205 210 215 220 225 230 235 240 245 LISTING OF CISTERN DEMAND REM PROGRAM CISTERN DEMAND REM DETERMINES REQUIRED STORAGE FOR SEVERAL’ DEMAND LEVELS REM WITHIN SPECIFIED RELIABILITY RANGES REM BY C. BROWNE AND H. SMITH - CARIBBEAN RESEARCH INSTITUTE REM ST.THOMAS, VIRGIN ISLANDS REM SUMMER, 1984 DIM L$(53),HL(2,53) , ID$(25) ,D(25,53) ,HA(21,53) ,CS(21,52) D$ = CHR$ (4): REM CONTROL-D FILE REM POSITION TO START OF FILE PRINT D$;"OPEN STT WEEKLY" PRINT D$;"READ STT WEEKLY" INPUT NC INPUT NV INPUT MC FOR A = 1 TO NV INPUT L$(A) INPUT HL(1,A) INPUT HL(2,A) NEXT A INPUT Q$ REM CASES INFORMATION FOR B = 1 TO NC INPUT ID$(B) FOR A = 1 TO NV INPUT D(B,A):D(B,A) = D(B,A) / 12: IF D(B,A) < O THEN D(B,A) = 0 NEXT A: NEXT B PRINT D$;"CLOSE STT WEEKLY": HOME INPUT "NUMBER OF YEARS OF RECORDS? ";NY PRINT "RUN FOR SEVERAL RELIABLITIES?" GET SR$: IF SR$ = "" GOTO 150 IF SR$ = "Y" GOTO 675 DE = O:COUNT = 0:CX = 99 HOME INPUT "CATCHMENT EFFICIENCY? ";CE:CE = CE / 100 INPUT "CATCHMENT AREA? ";CA INPUT "DESIRED RELIABILITY?";DR:LS = (100 - DR) / 100 INPUT "MINIMUM REQUIRED STORAGE?";RS:MX = RS PRINT "FOR SEVERAL DEMANDS?" GET A$: IF A$ = "" THEN 195 IF A$ = "Y" GOTO 475 INPUT "WEEKLY DEMAND IN GALLONS?" ;DM COUNT = COUNT + 1:SPILL = 0 FOR Y = 1 TO NY: FOR WK = 1 TO 52:PW = WK - 1 SV = CA * RS HA(Y,WK) = ((CA * D(Y,WK) * CE) * 7.48) - DM IF Y = 1 AND PW = O THEN CS(Y,WK) = HA(Y,WK): GOTO 245 IF PW = O THEN CS(Y,WK) = CS((Y -71),52): GOTO 245 CS(Y,WK) = CS(Y,PW) + HA(Y,WK) IF CS(Y,WK) > SV THEN CS(Y,WK) = SV:SPILL = 1 -AP 10- —~FJ (Listing of CISTERN DEMAND Cont'd) 250 IF CS(Y,WK) < O THEN CS(Y,WK) = O:DE = DE + 1 lr 255 IF (DE / (NY * 52)) > LS THEN GOTO 620 260 D(Y,WK) = INT (D(Y,WK) * 100 + .5) / 100 265 HA(Y,WK) INT (HACY ,WK) * 35 + .5) / 10 Pr 270 CS(Y,WK) INT (CS(Y,WK) * 35 + .5) / 10 275 IF RS > 30 GOTO 615 280 NEXT WK: NEXT Y = 285 PRINT "FOR A DEMAND OF "DM; r 290 " REQUIRED STORAGE IS "RS" GALS./SQ.FT." 295 IF CX = 99 THEN LRS = RS 300 REM STORING MIN RS FOR NEXT RELIABILITY SATISFACTION P 305 TPRS = RS:CX = O:COUNT = O: IF GG = 1 GOTO 535 310 PRINT "DO YOU WANT A LISTING ?" 315 GET X$: IF X$ = "" THEN 315 ir 320 IF X$ = "N" THEN GOTO 340 l" 325 FOR Y = 1 TO NY: FOR WK = 1 TO 52 330 PRINT "RAIN="D(Y,WK)" CATCH= "HACY, WK)" STORED="CS(Y, WK)" 335 NEXT WK: NEXT Y 340 PRINT "DO YOU WANT A PRINTED OUTPUT? © 345 GET X$: IF X$ = "" THEN 345 . 350 IF X$ = "N" THEN GOTO 470 c 355 INPUT "STARTING WITH YEAR";BY 360 INPUT "ENDING WITH YEAR";EY 365 PR# 1: PRINT ™ - 370 FOR I = BY TO EY STEP 2 f 375 PRINT "" TAB( 3)"WEEK YEAR RAIN STORAGE"; -380 PRINT "" TAB( 5)"YEAR RAIN STORAGE": PRINT "" 385 FOR WK = 1 TO 52:J =I +1 f 390 PRINT "" TAB( 4)WK; ian | “9 395 PRINT "" TAB( 7 - LEN ( STR$ (WK)))I; 400 PRINT "" TAB( 9 - LEN ( STR$ (I)))D(I,WK); 405 PRINT "" TAB( 9 - LEN ( STR$ (D(I,WK))))CS(I,WK); 410 PRINT "" TAB( 14 - LEN ( STR$ (CS(I,WK))))J;— - 415 PRINT "" TAB( 5 - LEN ( STR$ (J)))D(J,WK); 420 PRINT "" TAB( 9 - LEN ( STR$ (D(J,WK))))CS(J,WK) fl 425 NEXT WK: PRINT "": PRINT "0 430 IF I < (EY - 1) GOTO 460 435 PRINT "FOR A RELIABILITY OF "DR"% REQUIRED STORAGE IS "RS; i 440 PRINT "GALS./SQ.FT." 445 PRINT NY"YEARS",WEEKLYDEM AND OF"DM"AREAOF"CA; 450 PRINT "SQ.FT. AND EFFICIENCYOF"CE m 455 GOTO 465 i‘ 460 FOR C = 1 TO 5: PRINT CHR$ (10): NEXT C: NEXT I 465 PR# O 470 GOTO 585 r 475 REM SETTING UP FOR SEVERAL INCREMENTS 480 GG = 1:PRS = RS 485 INPUT "MINIMUM DEMAND?" ;MM - 490 DM = MM r | | -AP 14,- [ [ [ ' ; 7,1 73) “SD re —~9 [ P 500 505 510 515 520 525 530 935 540 545 550 555 560 565 570 575 580 585 590 595 600 605 610 615 620 625 630 635 640 645 650 655 660 665 670 675 680 685 690 695 700 705 710 715 720 725 730 735 (Listing of CISTERN DEMAND Cont'd) PRINT "DO YOU WANT A PRINTED OUTPUT? " GET X$: IF X$ = "" THEN 505 IF X$ = "N" GOTO 530 . PR# 1: PRINT "": PRINT "RELIABILITY IS "DR; PRINT "CATCHMENT IS "CA" SIMULATION FOR "NY" YEARS" PRINT "CATCHMENT EFFICIENCY IS "CE: PRINT "" GOTO 210 REM DM = DM + 50: IF DM > MD THEN GOTO 585 GOSUB 650 DE = 0: GOTO 21 GOTO 210 . IF X$ = "N" GOTO 585 PRINT "A HIGHER DEMAND SATISFACTION IS UNLIKELY": GOTO 585 REM REM REM | IF SR$ = "Y" GOTO 690 PRINT "RERUN THE PROGRAM?" GET R$: IF R$ = "" THEN 595 IF R$ = "N" GOTO 615 DE = O:COUNT = 0:CX = 99 GOTO 165 END REM PRINT "SPILL="SPILL: IF SPILL = 0 GOTO 565 IF CX = 99 THEN COUNT ='0: GOTO 640 IF COUNT > 10 GOTO 565 PRINT "Y="Y"WK="WK"DE="DE"RS="RS"SPILL="SPILL DE = O:RS = RS + 0.5: GOTO 210 REM INCREMENTING 'RS' BASED ON PAST TRENDS DI = RS ~ PRS: IF DI > 0.5 GOTO 665 PRS = TPRS: GOTO 670 RS = RS + (RS - PRS):PRS = TPRS RETURN INPUT "MINIMUM REQUIRED RELIABILITY";DR INPUT "MAXIMUM REQUIRED RELIABILITY" ;MAXR GOTO 160 IF DR > = MAXR GOTO 615 DR = DR + 5:LS = (100 - DR) / 100 DM = MM: REM RESETTING DEMAND RS = LRS: REM RESETTING TO LEAST FEASIBLE REQUIRED STORAGE DE = O:COUNT = O:CX = 99 PRINT : PRINT "RELIABILITY IS NOW "DR: PRINT IF GG = O GOTO 735 REM REM GOTO 210 -AP 12- LISTING OF CISTERN RELIABILITY 5 REM PROGRAM CISTERN RELIABILITY 10 REM DETERMINES REQUIRED STORAGE FOR SEVERAL RELIABILITY 15 REM LEVELS WITHIN SPECIFIED DEMAND RANGES. 20 REM BY C. BROWNE AND H. SMITH - CARIBBEAN RESEARCH INSTITUTE 25 REM ST. THOMAS, VIRGIN ISLANDS 30 REM SUMMER, 1984 | 35 DIM L$(53),HL(2,53) ,ID$(25) ,D(25,53) ,HA(21,53) ,CS(21,52) 40 D$ = CHR$ (4): REM CONTROL-D FILE 45 REM POSITION TO START OF FILE 50 PRINT D$;"OPEN STT WEEKLY" 55 PRINT D$;"READ STT WEEKLY" 60 INPUT NC 65 INPUT MC 70 FOR A = 1 TO NV 75 INPUT L$(A) 80 INPUT HL(1,A) 85 INPUT HL(2,A) 90 NEXT A 95 INPUT Q$ 100 REM CASES INFORMATION 105 FOR B = 1 T0 NC 110 INPUT ID$(B) ’ 115 FOR A = 1 TO NV 120 INPUT D(B,A):D(B,A) = D(B,A) of 12 125 IF D(B,A) < O THEN D(B,A) = 130 NEXT A: NEXT B °4135 PRINT D$;"CLOSE STT WEEKLY": HOME 140 INPUT "NUMBER OF YEARS OF RECORDS? ";NY 145 PRINT "RUN FOR SEVERAL DEMANDS?" 150 GET SR$: IF SR$ = "" GOTO 150 155 IF SR$ = "yY" GOTO 675 160 DE = O:COUNT = O:CX = 99 165 HOME 170 INPUT "CATCHMENT EFFICIENCY? ";CE:CE = CE / 100 175 INPUT "CATCHMENT AREA?";CA 180 INPUT "WEEKLY USAGE RATE?";DM 185 INPUT "MINIMUM REQUIRED STORAGE?";RS:MX = RS 190 PRINT "FOR SEVERAL RELIABILITIES ?" 195 GET A$: IF A$ = "" THEN 195 200 IF A$ = "Y¥" GOTO 475 205 INPUT "DESIRED RELIABILITY? ";DR:LS = (100 - DR) / 100 210 COUNT = COUNT + 1:SPILL = 0 215 FOR Y = 1 TO NY: FOR WK = 1 TO 52:PW = WK - 1 220 SV = CA * RS 225 HA(Y, WK) = ((CA * D(Y,WK) * CE) * 7.48) - DM 230 IF Y = 1 AND PW = 0 THEN CS(Y,WK) = HA(Y,WK): GOTO 245 235 IF PW = 0 THEN CS(Y,WK) = es((y¥ - 1),52): GOTO 245 240 CS(Y,WK) = CS(Y,PW) + HA(Y,WK) 245 IF cs(y, WK) > SV THEN cs(Y, WK) = SV:SPILL = 1 ~AP 13- [ [ —7FI7 450o0Ule a4). a 4) a a.) aT 94) aaa) or) cS 250 255 260 265 270 275 280 285 290 295 300 305 310 315 320 325 330 335 340 345 350 355 360 365 370 375 _ 380 385 390 395 400 405 410 415 420 425 430 435 440 445 450 455 460 465 470 475 480 485 490 ——_— (Listing of CISTERN RELIABILITY Cont'd) IF CS(Y,WK) < O THEN CS(Y,WK) = O:DE = DE + 1 IF (DE / (NY * 52)) > LS THEN GOTO 620 D(Y,WK) = INT (D(Y,WK) * 70 + .5) / 100 HA(Y, WK) INT (HA(Y, WK) *5 + 0.5) / 10 CS(Y, WK) INT (CS(Y,WK) * 5 + .5) / 10 IF RS > 30 GOTO 615 NEXT WK: NEXT Y: IF DE < 4 GOTO 585 PRINT "FOR A RELIABILITY OF"DR; PRINT "% REQUIRED STORAGE IS "RS" GALS./SQ.FT." IF CX = 99 THEN LRS = RS REM ABOVE STORES MIN RS FOR NEXT DEMAND SATISFACTION TPRS = RS:CX = O:COUNT = O: IF GG = 1 GOTO 535 PRINT "DO YOU-WANT A LISTING ?" GET X$: IF X$ = "" THEN 315 IF X$ = "N" THEN GOTO 340 FOR Y = 1 TO NY: FOR WK = 1 TO 52 PRINT "RAIN="D(Y,WK)" CATCH="HA(Y,WK)" STORED="CS(Y,WK)" NEXT WK: NEXT Y ‘PRINT "DO YOU WANT A PRINTED OUTPUT? " GET X$: IF X$ = "" THEN 345 IF X$ = "N" THEN GOTO 470 INPUT "STARTING WITH YEAR";BY INPUT "ENDING WITH YEAR?";BY PR# 1: PRINT "" FOR I = BY TO EY STEP 2 PRINT "" TAB( 3)"WEEK YEAR RAIN STORAGE"; PRINT "" TAB( 5)"YEAR RAIN STORAGE": PRINT "" FOR WK = 1 TO 52:J = I + 1 PRINT "" TAB( 4)WK; PRINT "" TAB( 7 - LEN ( STR$ (WK)))I3 PRINT "" TAB( 9 = LEN ( STR$ (1I)))D(I,WK); PRINT "" TAB( 9 - LEN ( STR$ (D(I,WK))))CS(I,WK); PRINT "" TAB( 14 - LEN ( STR$ (CS(I,WK))))d; PRINT "" TAB( 5 - LEN ( STR$ (J)))D(J,WK); PRINT "" TAB( 9 - LEN ( STR$ (D(J,WK))))CS(J,WK) NEXT WK: PRINT "": PRINT "" IF I < (EY - 1) GOTO 460 PRINT "FOR A RELIABILITY OF "DR"% REQUIRED STORAGE IS "RS 5 PRINT " GALS./SQ.FT." PRINT NY"YEARS",WEEKLYDEM AND OF"DM"AREAOF"'CA; PRINT "SQ.FT. AND EFFICIENCYOF"CE GOTO 465 FOR C = 1 TO 5: PRINT CHR$ (10): NEXT C: NEXT I PR# O GOTO 585 REM SETTING UP FOR SEVERAL INCREMENTS GG = 1:PRS = RS INPUT "MINIMUM REQUIRED RELIABILITY?";RR DR = RR:LS = (100 - DR) / 100 -AP 14- =a a es i Se i nr: ce: rn 500 505 510 515 520 525 530 535 540 545 550 555 560 565 570 575 580 585 590 595 600 605 610 615 620 . 625 630 635 640 645 650 655 660 665 670 675 680 685 690 695 700 705 710 715 720 725 730 735 (Listing of CISTERN RELIABILITY Cont'd) PRINT "DO YOU WANT A PRINTED OUTPUT? " GET X$: IF X$ = "" THEN 505 IF X$ = "N" GOTO 530 PR# 1: PRINT ""; PRINT "DEMAND IS "DM" CATCHMENT IS "CA; PRINT " SIMULATION FOR "NY" YEARS" PRINT "CATCHMENT EFFICIENCY IS"CE: PRINT "" GOTO 210 REM DR = DR + 5: IF DR > MR GOTO 585 GOSUB 650 LS = (100 - DR) / 100:DE = O: GOTO 210 GOTO 210 IF X$ = "N" GOTO 585 PRINT "A HIGHER RELIABILITY IS UNLIKELY": GOTO 585 REM REM REM IF SR$ = "Y" GOTO 690 PRINT "RERUN THE PROGRAM?" GET R$: IF R$ = "" THEN 595 IF R$ = "N" GOTO 615 DE = O:COUNT = O:CX = 99 GOTO 165 END REM PRINT "SPILL="SPILL: IF SPILL = O GOTO 565 IF CX = 99 THEN COUNT = O: GOTO 640 IF COUNT > 10 GOTO 565 PRINT "Y="Y"WK="WK"DE="DE"RS="RS"SPILL="SPILL DE = O:RS = RS + 0.5: GOTO 210 REM INCREMENTING 'RS' BASED ON PAST TRENDS DI = RS - PRS: IF DI > 0.5 GOTO 665 PRS = TPRS: GOTO 670 RS = RS + (RS - PRS):PRS = TPRS RETURN INPUT "MINIMUM REQUIRED DEMAND";DM INPUT "MAXIMUM REQUIRED DEMAND";MAXD GOTO 160 IF DM > = MAXD GOTO 615 DM = DM + 50 DR = RR:LS = (100 - DR) / 100: REM RESETTING RELIABILITY RS = LRS: REM RESETTING TO LEAST FEASIBLE REQUIRED STORAGE DE = O:COUNT = O:CX = 99 PRINT :; PRINT "DEMAND IS NOW"DM: PRINT IF GG = O GOTO 735 REM REM GOTO 210 ~AP 15-