The Andes of Southern Peru / Geographical Reconnaissance along the Seventy-Third Meridian

Combined with these topographic features are certain climatic features of equal precision. Between 7,000 and 13,000 feet is a zone of clouds oftentimes marked out as distinctly as the belt of fog on the Peruvian coast.[26] Rarely does it extend across the valley. Generally it hangs as a white belt on the opposite walls. When the up-valley winds of day begin to blow it drifts up-valley, oftentimes to be dissolved as it strikes the warmer slopes of the upper valley, just as its settling under surface is constantly being dissolved in the warm dry air of the valley floor. Where the precipitation is heaviest there is a belt of woodland—dark, twisted trees, moss-draped, wet—a Druid forest. Below and above the woodland are grassy slopes. At Incahuasi a spur runs out and down until at last it terminates between two deep canyons. No ordinary wells could be successful. The ground water must be a thousand feet down, so a canal, a tiny thing only a few inches wide and deep, has been cut away up to a woodland stream. Thence the water is carried down by a contour-like course out of the woodland into the pasture, and so down to the narrow part of the spur where there is pasture but no springs or streams.

Corn fields surround the few scattered habitations that have been built just above the break or shoulder on the valley wall where the woodland terminates, and there are fine grazing lands. The trails follow the upper slopes whose gentler contours permit a certain liberty of movement. Then the way plunges downward over a staircase trail, over steep boulder-strewn slopes to the arid floor of a tributary where nature has built a graded route. And so to the still more arid floor of the main valley, where the ample and moderate slopes of the alluvial fans with their mountain streams permit plantation agriculture again to come in.

To these three climates, the western border type, the eastern border type, and the inter-Andean type, we have given chief attention because they have the most important human relations. The statistical records of the expedition as shown in the curves and the discussion that accompanies them give attention to those climatic features that are of theoretical rather than practical interest, and are largely concerned with the conventional expression of the facts of weather and climate. They are therefore combined in the following chapter which is devoted chiefly to a technical discussion of the meteorology as distinguished from the climatology of the Peruvian Andes.

CHAPTER X

METEOROLOGICAL RECORDS FROM THE PERUVIAN ANDES

A further word to the reader seems necessary before he examines the following curves and tables. It would be somewhat audacious to assume that these short-term records have far-reaching importance. Much of their value lies in their organization with respect to the data already published on the climate of Peru. But since this would require a delay of several years in their publication it seems better to present them now in their simplest form. After all, the professional climatologist, to whom they are chiefly of interest, scarcely needs to have such organization supplied to him. Then, too, we hope that there will become available in the next ten or fifteen years a vastly larger body of climatological facts from this region. When these have been collected we may look forward to a volume or a series of volumes on the “Climate of Peru,” with full statistical tables and a complete discussion of them. That would seem to be the best time for the reproduction of the detailed statistics now on hand. It is only necessary that there shall be sufficient analysis of the data from time to time to give a general idea of their character and to indicate in what way the scope of the observations might profitably be extended. I have, therefore, taken from the available facts only such as seem to me of the most importance because of their unusual character or their special relations to the boundaries of plant provinces or of the so-called “natural regions” of geography.

Machu Picchu [27]

The following observations are of special interest in that they illustrate the weather during the southern winter and spring at the famous ruins of Machu Picchu in the Canyon of Torontoy. The elevation is 8,500 feet. The period they cover is too short to give more than a hint of the climate or of the weather for the year. It extends from August 20, 1912, to November 6, 1912 (79 days).

The high percentage of northwest winds during afternoon hours is due to the up-valley movement of the air common to almost all mountain borders. The air over a mountain slope is heated more than the free air at the same elevation over the plains (or lower valley); hence a barometric gradient towards the mountain becomes established. At Machu Picchu the Canyon of Torontoy trends northwest, making there a sharp turn from an equally sharp northeast bend directly upstream. The easterly components are unrelated to the topography. They represent the trades. If a wind rose were made for still earlier morning hours these winds would be more faithfully represented. That an easterly and northeasterly rather than a southeasterly direction should be assumed by the trades is not difficult to believe when we consider the trend of the Cordillera—southeast to northwest. The observations from here down to the plains all show that there is a distinct change in wind direction in sympathy with the larger features of the topography, especially the deep valleys and canyons, the trades coming in from the northeast.

Santa Lucia [29]

Santa Lucia is a mining center in the province of Puno (16° S.), at the head of a valley here running northeast towards Lake Titicaca. Its elevation, 15,500 feet above sea level, confers on it unusual interest as a meteorological station. A thermograph has been installed which enables a closer study of the temperature to be made than in the case of the other stations. It is unfortunate, however, that the observations upon clouds, wind directions, etc., should not have been taken at regular hours. The time ranges from 8.30 to 11.30 for morning hours and from 2.30 to 5.30 for afternoon. The observations cover portions of the years 1913 and 1914.

TEMPERATURE

Perhaps the most striking features of the weather of Santa Lucia are the highly regular changes of temperature from night to day or the uniformly great diurnal range and the small differences of temperature from day to day or the low diurnal variability. For the whole period of nearly a year the diurnal variability never exceeds 9.5° F. (5.3° C.) and for days at a time it does not exceed 2-3° F. (1.1°-1.7° C.). The most frequent variation, occurring on 71 per cent of the total number of days, is from 0-3° F., and the mean for the year gives the low variability of 1.9° F. (1.06° C.). These facts, illustrative of a type of weather comparable in uniformity with low stations on the Amazon plains, are shown in the table following as well as in the accompanying curves.

If we take the means of the diurnal variations by months we have a still more striking curve showing how little change there is between successive days. June and December are marked by humps in the curve. They are the months of extreme weather when for several weeks the temperatures drop to their lowest or climb to their highest levels. Moreover, there is at these lofty stations no pronounced lag of the maximum and minimum temperatures for the year behind the times of greatest and least heating such as we have at lower levels in the temperate zone. Thus we have the highest temperature for the year on December 2, 70.4° F. (21.3° C.), the lowest on June 3, 0.2° F. (—17.7° C.). The daily maxima and minima have the same characteristic. Radiation is active in the thin air of high stations and there is a very direct relation between the times of greatest heat received and greatest heat contained. The process is seen at its best immediately after the sun is obscured by clouds. In five minutes I have observed the temperature drop 20° F. (11.1° C.) at 16,000 feet (4,877 m.); and a drop of 10° F. (5.6° C.) is common anywhere above 14,000 feet (4,267 m.). In the curves of daily maximum and minimum temperatures we have clearly brought out the uniformity with which the maxima of high-level stations rise to a mean level during the winter months (May-August). Only at long intervals is there a short series of cloudy days when the maximum is 10°-12° F. (5.6°-6.7° C.) below the normal and the minimum stands at abnormally high levels. Since clouds form at night in quite variable amounts—in contrast to the nearly cloudless days—there is a far greater variability among the minimum temperatures. Indeed the variability of the winter minima is greater than that of the summer minima, for at the latter season the nightly cloud cover imposes much more stable atmospheric temperatures. The summer maxima have a greater degree of variability. Several clear days in succession allow the temperature to rise from 5°-10° F. (2.8°-5.6° C.) above the winter maxima. But such extremes are rather strictly confined to the height of the summer season—December and January. For the rest of the summer the maxima rise only a few degrees above those of the winter. This feature of the climate combines with a December maximum of rainfall to limit the period of most rapid plant growth to two months. Barley sown in late November could scarcely mature by the end of January, even if growing on the Argentine plains and much less at an elevation which carries the night temperatures below freezing at least once a week and where the mean temperature hovers about 47° F. (8.3° C.). The proper conditions for barley growing are not encountered above 13,000 to 13,500 feet and the farmer cannot be certain that it will ripen above 12,500 feet in the latitude of Santa Lucia.

The rainfall of Santa Lucia for this first year of record approximates closely to the yearly mean of 23.8 inches for the station of Caylloma in the adjacent province of that name. Caylloma is the center of a mining district essentially similar to Santa Lucia though the elevation of its meteorological station, 14,196 feet (4,330 m.), is lower. It is one of the few Peruvian stations for which a comparatively long series of records is available. The Boletín de la Sociedad Geográfica de Lima[32] contains a résumé of rainfall and temperature for seven years, 1896-7 to 1902-3. Later data may be found in subsequent volumes of the same publication but they have not been summarized or in any way prepared for analysis and they contain several typographical errors. A graphic representation of the monthly rainfall for the earlier period is here reproduced from the Boletín de minas del Perú.[33] The amount of precipitation fluctuates considerably from year to year. For the earlier period, with a mean of 23.8 inches the minimum (1896-7) was 8 inches and the maximum (1898-9) 36 inches. For the later period, 1903-4 to 1910-11, with a mean of 29.5 inches the minimum (1904-5) was 17.5 inches and the maximum (1906-7) was 43 inches.

WIND

An analysis of the wind at Santa Lucia shows an excess of north and south winds over those of all other directions. The wind-rose for the entire period of observation (Fig. 104) clearly expresses this fact. When this element is removed we observe a strongly seasonal distribution of the wind. The winter is the time of north and south winds. In summer the winds are chiefly from the northeast or the southwest. Among single months, August and February show this fact clearly as well as the less decisive character of the summer (February) wind.

The mean wind velocity for the month of February was 540 meters per minute for the morning and 470 meters per minute for the afternoon. The higher morning rate, an unusual feature of the weather of high stations, or indeed of wind-phenomena in general, is due, however, to exceptional changes in wind strength on two days of the month, the 16th and 25th, when the velocity decreased from a little less than a thousand meters per minute in the morning to 4 and 152 meters respectively in the afternoon. More typical is the March record for 1914 at Santa Lucia, when the wind was always stronger in the afternoon than in the morning, their ratios being 550 to 510.

CLOUD

The greater strength of the afternoon wind would lead us to suppose that the cloudiness, which in the trade-wind belt, is to so great an extent dependent on the wind, is greatest in the afternoon. The diagrams bring out this fact. Barely is the sky quite clear after the noon hour. Still more striking is the contrast between the morning and afternoon if we combine the two densest shadings of the figures. Light, high-lying cirrus clouds are most characteristic of early morning hours. They produce some very striking sky effects just before sunrise as they catch the sun’s rays aloft. An hour or two after sunrise they disappear and small cumulus clouds begin to form. These grow rapidly as the winds begin and by afternoon become bulky and numerous. In the wet season they grow into the nimbus and stratus types that precede a sudden downpour of water or a furious hailstorm. This is best seen from the base of a mountain range looking towards the crest, where the cloud-and rain-making processes of this type are most active.

The following abstracts are selected because they give some important features of the weather not included in the preceding tables and graphs. Of special interest are the strong contrasts between the comparatively high temperatures of midday and the sudden “tempests” accompanied by rain or hail that follow the strong convectional movements dependent upon rapid and unequal heating. The furious winds drive the particles of hail like shot. It is sometimes impossible to face them and the pack train must be halted until the storm has passed. Frequently they leave the ground white with hailstones. We encountered one after another of these “tempestades” on the divide between Lambrama and Antabamba in 1911. They are among the most impetuous little storms I have ever experienced. The longest of them raged on the divide from two-o’clock until dark, though in the valleys the sun was shining. Fortunately, in this latitude they do not turn into heavy snowstorms as in the Cordillera of northwestern Argentina, where the passes are now and then blocked for weeks at a time and loss of human life is no infrequent occurrence.[35] They do, however, drive the shepherds down from the highest slopes to the mid-valley pastures and make travel uncomfortable if not unsafe.

The meteorological data for 1908-09 were obtained from records kept by the Morococha Mining Company for use in a projected hydro-electric installation. Other data covering the years 1906-11 have appeared in the bulletins of the Sociedad Geográfica de Lima. These are not complete but they have supplied rainfall data for the years 1910-11;[36] those for 1906 and 1907 have been obtained from the Boletín de Minas.[37]

Temperature

The most striking facts expressed by the various temperature curves are the shortness of the true winter season—its restriction to June and July—and its abrupt beginning and end. This is well known to anyone who has lived from April to October or November at high elevations in the Central Andes. Winter comes on suddenly and with surprising regularity from year to year during the last few days of May and early June. In the last week of July or the first week of August the temperatures make an equally sudden rise. During 1908 and 1909 the mean temperature reached the freezing point but once each year—July 24 and July 12 respectively. The absolute minimum for the two years was -22° C. July of 1908 and June of 1909 are also the months of smallest diurnal variability, showing that the winter temperatures are maintained with great regularity. Like all tropical high-level stations, Morococha exhibits winter maxima that are very high as compared with the winter maxima of the temperate zone. In both June and July of 1908 and 1909 the maximum was maintained for about a week above 55° F. (12.8° C.), and in 1909 above 60° F. (15.6° C.), the mean maximum for the year being only 4.7° F. higher. For equal periods, however, the maxima fell to levels about 10° F. below those for the period from December to May, 1908.