From the description of the line as given, it is evident that much curvature must be found in spite of the most judicious efforts to avoid it, and the table indicates that condition. Yet the amount of curvature may be considered moderate for a location through such a country as Nicaragua. The smallest radius is seen to be a little over 4000 feet. The result may be considered satisfactory for such a difficult canal country, although the total amount of curvature is rather formidable.

339. The Conchuda Dam and Wasteway.—The most important single engineering feature of the whole plan is the dam at Conchuda. The ordinary low-water elevation in the river at the dam site may be taken at about 55 feet above the sea. Inasmuch as the greatest elevation of the water in the lake is supposed to be about 110 feet, it will be seen that its surface will be but 55 feet above the present elevation, making its maximum depth at that point about 105 feet if there should be no fill on the up-stream side of the dam, inasmuch as the present depth of water in the river at the stage assumed is about 50 feet.

This dam would be a structure of concrete masonry with cut-stone facing only at a few points where it would be advisable to use that material. A large part of the flood discharge, or the discharge of other surplus water, would be made over a properly designed crest of the dam; hence its outline would be that shown in the accompanying figure, shaped so as to prevent the overflowing sheet of water from damaging the structure. This dam will be founded upon pneumatic caissons, and the borings made by the commission show that the deepest of them would reach satisfactory bed-rock at no greater depth than 25 feet below sea-level, or about 80 feet below the ordinary stage of water in the river. The construction of this dam therefore would involve no unusual operations, but it would all be performed within the more usual and easy limits of the pneumatic process of constructing foundations. The masonry crest of this dam would be finished at the elevation of 97 feet above sea-level, or about 13 feet below the highest elevation of water in the lake. The length of that part of this masonry dam, located on pneumatic caissons, would be 731 feet, but the total length of the entire masonry structure would be 1310 feet. The total length of crest, including the masonry piers on it, over which the surplus waters would flow, would be 810 feet, but there are twenty piers 9 feet thick, so that the net length of crest available for overflow of waste-waters would be about 630 feet. The piers to which reference is made are those required for the support of the movable gates of the Stoney type which would be employed to regulate the discharge over the dam. The maximum elevation of the tops of these piers required for the support and operation of the Stoney gates is 132 feet above sea-level. The masonry dam thus furnished with movable gates can be used in times of flood to prevent the water of the lake rising above about 110 feet above sea-level. In times of low rainfall or during the dry season the gates would prevent the escape of water needed for storage.

The total available length of crest on this masonry dam is not sufficient to exercise all the control that is needed to keep the lake within desired limits, and the commission was obliged to avail itself of a low depression or saddle between the hills less than a half-mile easterly of the dam site. The depression affords an additional total length of crest of 1239 feet, or, taking out thirty-one piers, each 9 feet wide, a net available length of 960 feet, making in combination with the crest of the main dam a total net available length of 1590 feet. The total wastage over these two structures, i.e., the main dam at Conchuda and the Conchuda wasteway on the Costa Rican side of the river, may be at the rate of 100,000 cubic feet per second, with a maximum depth over the crest of 7 feet, which is sufficient to meet the demands of the heaviest rainfall in the lake basin.

The plans and elevations on pages 421, 423, and 424 show all the main features of both the Conchuda dam and wasteway as designed by the commission.

340. Regulation of the Lake Level.—One of the most important engineering questions connected with the consideration of the Nicaragua route is that of the regulation or control of the surface of the water in Lake Nicaragua constituting the summit level of the canal.

As has already been stated, the drainage-basin of the lake, about 12,000 square miles in area, is subjected to an annual wet season extending from about the middle of May to the middle of December, the dry season extending over the remaining portion of the year. The average annual rainfall over the entire lake basin is not accurately known, although the Isthmian Canal Commission maintained rainfall records at several points on the lake shore and at other points in the basin during periods of 1½ to 2 years, and records running back over periods of perhaps 12 to 15 years are available from Rivas, Granada, and Masaya. Fortunately, also, both the Nicaragua and the Isthmian Canal Commissions maintained gauging-stations at various points on the San Juan throughout the periods of service of these commissions, so that the discharges of the river could be known from accurate measures at various seasons for at least two or three years. These observations, although not as extended as could be desired, yield sufficient data for a comparatively thorough treatment of the subject of lake-surface control.

Obviously throughout the rainy season of the year, except during years of low rainfall, some water would necessarily be wasted from the lake because its retention would raise the surface of the lake too high, causing damage, floods, or injurious overflows at various places around the lake shore. On the other hand, unless some water were stored from the rainy periods or wet seasons there would not be sufficient in the lake to supply during the dry period of the year, or during low rainfall years, the requisite quantity for the wastage of evaporation from its surface and for the operation of the canal, and at the same time maintain the minimum depth of water of 35 feet required in the canal. It was necessary, therefore, to design at least the general features of such regulating-works as would prevent the lake from rising too high in wet periods, and from falling too low in dry periods or low rainfall years.

341. Evaporation and Lockage.—The observations of both commissions show conclusively that the average evaporation from the surface of Lake Nicaragua is about 60 inches or 5 feet per year, varying from perhaps a maximum of 6 inches per month to a minimum of possibly about 4 inches per month. Furthermore, careful estimates of the quantity of water required for the purposes of the canal, on the supposition that about 10,000,000 tons of traffic would pass through it annually, including lockage, leakage through the gates of the locks, evaporation, power purposes, and other incidentals, show that about 1000 cubic feet of water per second must be provided. Whatever may be the character of the season, therefore, there must be at least sufficient water stored in the lake to provide for the wastage of evaporation from the lake and canal surfaces and for the proper operation of all the locks throughout the length of the canal. The superficial area of Lake Nicaragua is but little less than 3000 square miles. The quantity of water required for the operation of the canal, amounting to 1000 cubic feet per second, would, for the entire year, make a layer of water over the lake surface of less than 5 inches in thickness. In other words, the operation of the canal, for a traffic of about 10,000,000 tons annually, requires an amount of water less than one twelfth of that which would be evaporated from the lake surface during the same period.

342. The Required Slope of the Canalized River Surface.—The dam located at Conchuda and fitted with suitable movable gates affords means of accomplishing the entire lake-surface control. That dam is located, however, nearly 53 miles from the lake, and in order that the requisite discharge may take place over it during the rainy season there must be considerable slope of the water surface in the canalized river from the lake down to the dam. It was necessary, therefore, to compute that slope, from data secured by the commission, with the lake surface at various elevations between the minimum and maximum permitted. These slopes were found to be such that the difference in elevations of the surface of the water at the dam and in the lake might vary from about 6 to 9 feet, those figures representing the total fall for the distance of 53 miles.

343. All Surplus Water to be Discharged over the Conchuda Dam.—The Nicaragua Commission contemplated the construction of dams not only on the San Juan River at Boca San Carlos, about 6 miles below Conchuda, but also another a few miles west of the lake at La Flor, so as to discharge the surplus waters at both points, but by far the largest part over the dam at Boca San Carlos. The Isthmian Canal Commission, however, decided to build no dam on the west side of the lake, but to discharge all the surplus waters over the dam at Conchuda.

344. Control of the Surface Elevation of the Lake.—The rainfall records in the lake basin have shown that a dry season beginning as early as November may be followed by an extremely low rainfall period, which in turn would be followed by a dry season in natural sequence, lasting as late as June. It may happen, therefore, that from November until a year from the succeeding June, constituting a period of nineteen months, there will be a very meagre rainfall in the lake basin, during which the precipitation of the seven low rainfall wet months may not be sufficient even to make good the depletion of evaporation alone during the same period. It would be necessary, then, at the end of any wet season whatever, i.e., during the first half of any December, or in November, to make sure of sufficient storage in the lake to meet the requirements of the driest nineteen months that can be anticipated. That condition was assumed by the commission, and the elements of control of the lake surface, in its plans, are such as to afford resources to meet precisely those low-water conditions.

The commission’s study of these features of the Nicaragua Canal problem resulted in plans of works to prevent the surface of the lake ever falling below 104 feet above sea-level, or rarely if ever rising higher than the elevation of 110 feet above the same level, thus making the possible range of the lake surface about 6 feet between its lowest and its highest position.

Obviously at the end of a dry season the gates at the dam will always be found closed, and there will be no water escaping from the lake except by evaporation and to supply the needs of canal operation. It is equally evident that the gates will also remain closed so as to permit no wastage during the early part of the wet season. As the wet season proceeds the surface of the lake will rise toward, and generally quite to its maximum elevation; the operation of wasting over the weirs will then commence. The time of beginning of this wastage will depend upon the amount and distribution of the rainfall during the wet period. Indeed no wastage whatever would be permitted during such a low-water wet season as that shown by the records of 1890, which was almost phenomenal in its low precipitation. The rainfall for the entire drainage-basin would be impounded in the lake in that case, and it would then fall short of restoring the depletion resulting from evaporation and requirements of the canal. On the other hand, during such a wet season as that of 1897 wastage would begin at an early date. In general it may be said that neither the rate nor the law of the rise of water surface in the lake can be predicted. There will be years when no wastage will be permitted, but generally considerable wastage will be necessary in order to prevent the lake rising above the permissible highest stage.

Detailed computations based upon the statistics of actual rainfall records in the basin of Lake Nicaragua may be found by referring to pages 147 to 152 of the Report of the Isthmian Canal Commission, and they need not be repeated here. Those computations show among other things that October is often a month of excessive rainfall, and that the greatest elevation of the lake surface is likely to follow the precipitation of that month. Hence the greatest discharge of surplus waters over the Conchuda dam may be expected in consequence of the resulting run-off or inflow into the lake. Those computations also show that at long intervals of time the lake surface might reach an elevation of nearly 112 feet above sea-level for short periods, causing the discharge in the canalized river or over the Conchuda dam to reach possibly 76,000 cubic feet per second, the elevation of the water at the dam being 104 feet above sea-level. Furthermore, the Sabalos River and one or two other small streams, emptying into the San Juan above the dam, might concurrently be in flood for at least a few hours and augment the discharge over the dam to 100,000 cubic feet per second. The regulating-works at the dam, consisting of the movable (Stoney) gates, were devised by the commission to afford that rate of discharge, an aggregate net or available length of overflow crest at the dam and wasteway of 1590 feet being necessary for that purpose with a depth of water on the crest not exceeding 7 feet.

The commission states on page 156 of its report:

“While, therefore, no detailed instructions can be set forth regarding the condition of the sluices at the wasteway on specified dates, the general lines of their operation should be stated below, viz.:

It is thus seen that while the measures for control and regulation are entirely feasible, they are not sharply defined, nor so simple that some experience in their operation might not be needful for the most satisfactory results.

345. Greatest Velocities in Canalized River.—It is necessary to ascertain whether the velocities induced in the canalized portions of the San Juan River would not be too high for the convenience of traffic during the highest rainfall season. The following table and the succeeding paragraph, taken from the commission’s report, show that no sensible difficulty of this kind would exist.

“The discharge of the river corresponding to the velocity of 2.7 miles per hour is 63,200 cubic feet per second; while that corresponding to 3.3 miles per hour is 77,000 cubic feet per second. These estimated high velocities will occur but rarely, and they will not sensibly inconvenience navigation. In reality they are too high, for the reason that while the overflow at the minimum river section materially increases the areas of those sections, it has been neglected in this discussion.”

Brito, at the Pacific Terminus of the Nicaragua Route,
showing the mouth of the Rio Grande on the left and
the easterly side of Brito Head.

346. Wasteways or Overflows.—At a number of places on the route there are some small streams which must be taken into the canal, and which when in flood require that certain wasteways or overflows from the canal prism should be provided at or near where such streams are received. These wasteways are simply overfall-weirs with the crests at the elevation of the lowest water surface in the canal prism. The principal works of this kind are on the east side of the lake and involve a total drainage area or area of watershed of about 107 square miles. Ample provision has been made by the commission for all such structural features.

347. Temporary Harbors and Service Railroad.—Before actual work could be begun at either end of the Nicaragua route temporary harbors would have to be constructed both at Greytown and at Brito to enable contractors to land plant and supplies or other material. These temporary harbors would probably require no greater depth of water than 18 feet, but they would be works of considerable magnitude, and provision was made for them in the commission’s estimate of cost. Again, a service railroad of substantial character would have to be built from Greytown up to Sabalos, approximately half-way between the Conchuda dam and Fort San Carlos, as well as from the west shore of the lake to Brito, making a total line of about 100 miles. The commission estimated the cost of this railroad and its rolling stock at $75,000 per mile.

348. Itemized Statement of Length and Cost.—The following table gives the lengths of the various portions of the canal and the principal items of its cost, so arranged as to show the classification of the various items of the total sum to be expended for all purposes during the construction of the entire work.

The commission estimated the total time required in preparing for and performing the actual construction of the work at eight years, but the writer believes that at least two years more should be allowed for the work.