Division of Agricultural Sciences UNIVERSITY O F C A L I F O R N *™Q mmmmB: WMmMim ULLAGE AND NITROGEN FOR DRYLAND GRAIN '■m¥w0M:. '-";>: '<:?:>■?. /S >?■€*-• r^^M"^ '^iM;'s%: : -$PSiM^^ ■ « I • IN A WINTER RAINFALL CLIMATE CALIFORNIA AGRICULTURAL EXPERIMENT STATION BULLETIN 805 This bulletin, written as a reference for agronomists and soil scientists, • describes climate, soil, and cropping systems for dryland agriculture in a winter rainfall climate; # reports experimental results showing major tillage effects on soil moisture storage, soil nitrification, plant growth, grain quality, and yield ; * reports experimental results showing the wide range in effects of nitrogen fertilization and legume cropping on dryland grain pro- duction; and * discusses soil moisture-nitrogen relations as they are affected by till- age and nitrogen application. The research reported in this bulletin has been a cooperative effort by the South- west Branch, Soil and Water Conservation Research Division, Agricultural Research Service, USDA, and the California Agricultural Experiment Station. AUGUST, 1964 The Authors: R. E. Luebs is Associate in the Experiment Station, Department of Soils and Plant Nutrition, Riverside, and Research Soil Scientist, USDA; A. E. Laag is Principal Laboratory Technician, Department of Soils and Plant Nutrition, River- side. Much of the research reported here was initiated by Maurice Donnelly, formerly Soil Scientist, USDA. CONTENTS THE FINDINGS 5 Tillage 5 Nitrogen 6 DRYLAND FARMING IN CALIFORNIA 7 Climate 7 Crops and cropping systems 8 Soils 10 TILLAGE 11 Grain yield 12 Grain quality 15 Straw production 16 Nitrogen content of straw 17 Forage yield of volunteer barley 17 Soil nitrification 18 Reduction of surface crop residue 19 Soil moisture storage 19 Nitrogen content of soil 20 Discussion of tillage effects 20 NITROGEN 24 Frequency and magnitude of grain yield response 25 Rate of application 28 Fertilizer sources 28 Grain quality 29 Straw yield 29 Residual effect 30 Response to applied phosphorus 30 Legume cover crops 30 Dryland crop response to nitrogen 31 LITERATURE CITED 33 [3] TILLAGE AND NITROGEN FOR DRYLAND GRAIN THE FINDINGS Experiments were conducted in southern California over a 15-year period to eval- uate tillage and nitrogen fertility in dryland grain production. Field studies involved the prevailing fallow-grain crop- ping system. Tillage experiments included a study of methods and times of primary fallow tillage. One of the specific objec- tives was to compare subsurface or stub- ble-mulch tillage with the more intensive methods, such as moldboard plowing and disking. The duration of the tillage ex- periments ranged from 6 to 15 years. Nitrogen fertility studies included rate and source variables. Thirty-two nitrogen fertilizer experiments were conducted. Average rainfall over the 15-year pe- riod was approximately 80 per cent of the long-time average, ranging between 9 to 15 inches for the different sites. Most of the soils were formed from granitic alluvium and contained less than 0.1 per cent total nitrogen. In some of these soils (mainly the Placentia and Arlington 2 series) , moisture movement was undoubt- edly restricted by layers of high bulk density. Tillage Grain yield. During a period of less than 10 inches annual rainfall, tillage methods had little or no effect on grain yield in three of four years on an Arling- ton sandy loam. Moldboard plowing out- yielded chiseling and subtilling by nearly 500 pounds of grain per acre in one year. Fall or winter fallow tillage was fre- quently superior to spring tillage. Appli- cation of up to 2 tons of straw per acre on the soil surface when barley was in the 1 Submitted for publication November 13, 1963 2 Tentative soil series. seedling stage did not affect yield in two of three years. The 2-ton rate decreased yield one year. With annual average rainfall of 13 to 14 inches, moldboard plowing averaged 400 and 500 pounds per acre more grain than did disking or subtilling, respec- tively, on Ramona and Placentia sandy loams. Yields with plowing were also sig- nificantly greater when 30 pounds of nitrogen per acre were applied. Average yield increases from nitrogen were larger on disked and subtilled plots, but sub- tillage with nitrogen averaged more than 300 pounds less than plowing without nitrogen, indicating a greater need for nitrogen with subtillage. Fallow tillage had little or no effect on grain yield where production was low (average of 1,000 pounds per acre) on an eroded Placentia sandy loam. Nitrogen was not limiting for grain production at this site, although rainfall averaged about 15 inches over a 12-year period. Fall chiseling resulted in significant yield in- creases over midwinter disking on this soil. Test weight of grain. Fallow tillage methods — moldboard plowing, disking, and subtilling — usually had little effect on the test weight of grain. Exceptions in two years were lower test weights with plowing and disking. Crude protein content of grain. Higher crude protein content of grain was obtained with moldboard plowing. An average increase of 3 per cent over other tillage methods was measured for four cropping seasons. Lowest protein levels were obtained when chiseling, rod- weeding, or subtilling were the primary fallow tillage methods. Yield and nitrogen content of straw. On Ramona sandy loam soil with [5] 15 inches average annual rainfall, straw production with plowing was 500 and 600 pounds per acre greater than with disk- ing and subtilling, respectively. Highest nitrogen content of straw usually oc- curred with plowing. In an experiment where grain yields were not affected by tillage, both yield and nitrogen content of straw were highest with plowing. Soil nitrates. Soil nitrate accumula- tion over the fallow period was greatest with moldboard plowing. Increases up to 80 per cent over subtilling and disking were determined in some years. Higher grain yields in some experiments and generally higher crude protein content of grain, higher straw yields, and higher nitrogen contents of straw undoubtedly reflect the greater nitrogen availability with plowing. Residual effect. Forage production of volunteer barley was significantly greater on plowed than on disked or sub- tilled plots two years after tillage opera- tions were performed. Residue reduction. Six months after tillage, 60 per cent of the crop residue remained on the soil surface with sub- surface tillage (stubble mulch). Residue remaining on plowed, chiseled, and disked plots amounted to 8, 38, and 46 per cent, respectively. Soil moisture. Soil moisture storage over the fallow period was 1.56 inches greater with subtilling than with plowing one of four years. No significant differ- ence was measured in other years, and there was no reflection in yield in any of these years. Nitrogen Crop response. On low nitrogen soils the practice of fallowing frequently does not insure adequate available nitrogen for dryland grain even under very limited rainfall. Yield increases with applied nitrogen were obtained in 18 of 32 experiments over a 15-year period. Increases exceeded 200 pounds in one- third of the experiments. Yield decreases with the application of nitrogen occurred in a few experiments. Observations of the crop indicated greater moisture stress toward the end of the growing season on plots with more available nitrogen. Rate of application. Where nitrogen was needed, 30 to 40 pounds per acre (applied before planting) were optimum for dryland grain after fallow. Lower rates reduced the probability of yield increase and higher rates increased the probability of yield decreases. Crude protein and test weight. Applied nitrogen usually increased the crude protein content of grain, but soil type and growing season were modifying factors. Nitrogen reduced test weight in a few instances. Straw production. Consistent and frequentlv large increases in straw yield (up to 800 pounds per acre) were ob- tained from applying 30 to 40 pounds of nitrogen. Increases in straw yield occurred even when grain yields were decreased or not affected. Residual effect. Air-dry forage yields of barley were increased over 1,000 pounds per acre by the residual effect of 30 pounds of nitrogen applied the rela- tive^ dry previous year for a grain crop. Phosphorus. Grain yield response to nhosphate fertilizer was obtained in 2 of 14 fieM experiments. In both cases, yields wer~ increased only when nitrogen fertilizer was also applied. Legume cover crop. Over an 11- year period, 30 pounds of nitrogen usually produced higher grain yields than a purple vetch cover crop between barley crops. Under the low rainfall of southern California, nitrogen in the tops of winter legumes frequently has not exceeded 50 pounds per acre. [6] DRYLAND FARMING IN CALIFORNIA Dryland farming is generally character- ized as crop production in areas where lack of moisture is the major limiting factor. Nonirrigated, arable lands re- ceiving an average of 8 to 20 inches of precipitation are considered to be dry- farmed. More than 3 million acres of cropland in California are in dryland agriculture. Approximately 400,000 acres are dry-farmed in southern California. A major portion of this land is unsuitable for irrigation because of rolling topog- raphy, shallow soil, or no readily acces- sible source of water for irrigation. Climate Annual rainfall distribution and other climatic factors, principally temperature and solar radiation, determine crop- growing conditions in a particular area. The climate of California is frequently described as a dry, mild winter or Medi- terranean type. Taylor (1960) has classi- fied dryland climates into six types. His climatic Type I most nearly represents the California climate: "Warm, dry growing weather with precipitation occurring mostly as snow and rain during a mild winter season such as occurs in the intermountain region of the western PER CENT AVERAGE ANNUAL RAINFALL Figure 1. Monthly Rainfall as a Percentage of the Average Annual Amount for an 80- year Period (1880-1960), Riverside, California, [7] United States." Except for the northern- most counties, snowfall is not significant in California dryland agriculture, and effective rainfall is received during a rela- tively short period of the year, in contrast to other intermountain areas. Significant amounts of rainfall usually occur only in the winter and early spring months. At Riverside, more than 72 per cent of the annual average has been recorded during the months of December, January, February, and March (figure 1). Negli- gible amounts of moisture are received between June and September. An analysis of 65 years of Los Angeles rainfall records by Donnelly (1943) shows that years of below-average rainfall exceed those of above-average. Mean rainfall was 1.6 inches greater than the median rain- fall. Field data on dryland soil management discussed in this bulletin were obtained from 1945 through 1960, a period of below-average rainfall. The long-time average of 11.06 inches at Riverside (determined by the U. S. Weather Bureau for the 1880-1960 period) was exceeded in only three years, and the average for 1945-60 was 82 per cent of the 80-year average (table 1). A review of rainfall data from several locations in southern California revealed that while different locations consistently receive more or less amounts than others, seasonal distribu- tion is similar. Evaporation from a free-water surface is directly related to moisture use by plants. Over the 35-year period of 1925- 60 at Riverside, an average of nearly 65 inches evaporated annually (table 2). The amounts for November, December, January, and February are relatively low. Evaporative potential increases in March and continues to increase, reaching a peak in July, when an average of 9.21 inches evaporated from a free- water sur- face. Beginning with September, the potential for evapotranspiration markedly decreases during the remainder of the year. i The growing season for dryland crops in a Mediterranean-type climate is deter- mined by the yearly rainfall distribution. Frosts rarely affect the length of growing seasons, in contrast to the continental dry- land climate which prevails in the Great Plains. The length of any one season in winter rainfall areas is largely limited by the earliness and lateness of effective rain- fall. Depending on local conditions, dry- land cereals are usually planted between November 15 and January 15 and do not mature until late June. During the last 12 weeks of the growing season, the proba- bility of rainfall decreases and the proba- bility of greater moisture use through evapotranspiration increases. Crops fre- quently show moisture stress during this period, the level of grain yield being determined by the amount of late rain- fall or available moisture stored in the soil During the recent 15-year dry period, poor distribution as well as lower amounts of rainfall have decreased production. Except for the 1951-52, 1953-54, and 1957-58 seasons, little or no significant moisture has been received in the spring (table 1). The data suggest that low- rainfall years have shorter seasons of ef- fective rainfall. Higher average evapora- tion from a free-water surface is also associated with the low-rainfall period (table 2). Crops and Cropping Systems Small grains, because of their charac- teristic growing season and drouth toler- ance, are the best adapted crops for dry- land agriculture. Spring varieties are grown under the mild winter climatic conditions. The dryland acreage of small grain exceeds 2 million acres in Califor- nia each year. Barley and wheat pre- dominate, with barley occupying twice the acreage of wheat. Oats are grown on a relatively small dryland acreage. Small grains contribute markedly to the Cali- fornia agricultural economy through the feed industry, brewing industry, and the CO N CD »»!6CfjNIONO! ro co O co O x. i^ o co »-« t~ o> 10 ■«*<' t~- o in o o ocooHooHooo— •oor -1 ^ 1Q UJ fj to n i- n .-> -H lO O C^J Q S3 « 03 O Plh c3 P . m O COOCOOOOTf'f'fOt-- NMNOCCOCMCNOOO <« OO ffl lO CO -* CM ^< OS O CO o «o <«J< C-J o o o ■<*< 1-H ^h c^ ^H ^H CO ctj r ^ u; *— i "^T" ^T *-^ l^- >-*J OPOrnrfOOINNCO t^ lO CO O O CO OO O -H o O O O 1-* §h8§§ShShh§§ Csl >o o o o o o o o o o o 2§ ) t> oc — c e> ec irt tC X a o M 1 - I I I I W ffl h OO i s Table 2. Average Monthly and Annual Evaporation from a Free-Water Surface, University of California, Riverside Period Evaporation, inches Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Annual 1925-60 1946-60 2.58 2.43 2.86 2.94 4.39 4.48 5.36 5.42 6.85 6.65 7.69 7.96 9.21 9.35 8.58 8.58 6.70 7.00 4.75 4.84 3.47 3.32 2.44 2.63 64.9 65.6 baking and milling industries. Oil crops, mainly safflower, and some forage crops are now being more widely grown on nonirrigated land. The prevailing cropping system for dryland grains alternates fallow with cropping in a two-year cycle. The chief purpose of fallow has been moisture con- servation. Mathews (1951) considers fallowing necessary for crop production when annual precipitation is less than 15 inches. The practice may be profitably used in certain areas where rainfall ex- ceeds 20 inches. Another important purpose of fallow is nitrate accumulation in the soil. Little- john (1956) found that fallow increased soluble nitrogen five to six times over recent cropping in a winter rainfall area. Staple (1960) recently reported that fallow grain crops have not responded to nitrogen fertilizer in Canada and the United States, presumably because fallow makes sufficient nitrates available. Soils A large part of the field research was conducted on sandy loam soils developed from granitic alluvium. Major soil series were Ramona, Placentia, and Arlington. The Ramona sandy loam is a well-drained Noncalcic Brown soil occurring on gentle slopes. It has a neutral to slightly acid A-horizon extending to a depth of about 15 inches. The B-horizon has a sandy clay loam texture and is neutral in reaction. The mildly alkaline, sandy loam C-horizon occurs at about 50 inches. The chief difference between the Ramona and Placentia sandy loams is the strongly developed clayey layer in the B-horizon of the latter. This compact layer restricts root development and internal drainage. Tillage experiments at Riverside were located on a soil which has a weakly cemented layer in the profile. This layer occurs at a depth of from 24 to 42 inches and varies in thickness from 20 to 36 inches. The soil has less clay in the B-horizon than does the Ramona and was formerly mapped as Greenfield. The Soil Conservation Service is currently con- sidering the establishment of a new series for this soil, which is tentatively desig- nated as Arlington. Most of the California dryland soils have been formed under semiarid condi- tions. Soil organic matter contents are characteristically low. Soil samples from several experimental sites where dryland cropping has been practiced were analyzed for organic carbon content by the chromic acid wet digestion method. Samples of uncropped or virgin soil adjacent to the cropped areas were ob- tained at two locations. All cropped soils contained less than 1 per cent organic matter in the surface 6 inches, and less than 0.65 per cent at the 18- to 24-inch depth (table 3). The data indicate that after from 50 to 70 years of grain cropping, mostly in the alternate fallow-grain system, from 30 to 40 per cent of the organic matter has been lost in the surface 6 inches. Smaller losses occurred below the 6-inch depth. Much dryland agriculture in California is located on lands with relatively steep slopes and on soils with moderate to low infiltration rates. As a consequence, soil erosion and runoff have occurred with individual storms even though annual [10 Table 3. Organic Matter Content of Soils Cropped under Dryland Conditions in Southern California, 1952 Location Soil type Slope History Organic matter* 0-6" 6-12" 12-18" 18-24" Ramona sandy loam Placentia sandy loam Arlington! sandy loam Placentia sandy loam per cent 3 9 3 4 Cropped 70 yr Virgin Cropped 70 yr Virgin Cropped 50 yr Cropped 75 yr per cent Beaumont Beaumont Riverside Beaumont 0.91 1.31 0.87 1.48 0.90 0.96 0.76 0.99 0.68 0.82 0.51 0.84 0.69 0.82 49 0.54 34 0.72 0.64 0.76 0.45 0.42 0.36 0.65 * Organic carbon X 1.7; average of 54 soil samples on cropped soils; minimum of 6 samples on virgin soil. t Tentative soil series. rainfall has been low. Nearly every winter California. In addition to the loss of top- the effect of erosion by water is evident soil, the abrasive action of drifting soil on bare fallow or seeded cropland. particles may have a damaging effect on Sustained high wind velocities have crop plants, caused soil erosion in certain areas of tillage Field research in other areas of the United States has shown that crop resi- dues on the soil surface control runoff, water erosion, and wind erosion. In eastern Nebraska, Duley and Russel (1939) reported that 2 tons of straw per acre on the surface conserved 3.6 inches more water in the soil than did plowing the straw in. In their summary of stubble- mulch research in the western states, Zingg and Whitfield (1957) referred to soil losses on plowed and mulched land at Lincoln, Nebraska, and Pullman, Washington. At Lincoln, average annual amounts of erosion were 1.26 and 6.02 tons of soil per acre for the mulched and plowed surfaces, respectively. At Pull- man, 3.63 and 17.93 tons per acre were lost with mulching and plowing, respec- tively. Approximately 1 ton of straw per acre comprised the mulch at both locations. In early wind erosion studies, Chepil (1944) observed that crop residues on the soil surface markedly reduced wind [ velocity and soil erosion. More recent work in Kansas (Zingg and Whitfield, 1957) indicated that from 500 to 2,000 pounds of anchored surface residues, depending on soil properties, would stabilize the surface against wind erosion. The major objectives of tillage experi- ments reported here were (1) to deter- mine the magnitude of tillage effects on crop yield, and (2) to ascertain the soil factors which may be responsible for the tillage effects obtained. Nitrogen analyses of soil and crop were included. Moisture conditions have been characterized by monthly rainfall data and some soil mois- ture data. Surface residue reduction was measured for certain tillage methods. Tillage experiments were conducted at several locations in southern California from 1945 to 1960. The use of three im- plements for the initial and major fallow tillage operation was common to all ex- periments. These implements and the approximate depth of operation were moldboard plow, 8 to 10 inches; disk, 4 11 to 6 inches; and subtiller, 4 to 5 inches. The latter implement is frequently referred to as a sweeps machine. In these experiments, the Noble blade was chosen to represent subsurface tillage equipment. Throughout the tillage studies, all opera- tions were completed under the direction of one of the authors and with equipment maintained only for this purpose. Opera- tions necessary for weed control and seed- bed preparation could be considered minimal when compared to the number of operations commonly used in dry-farmed areas. Plot sizes ranged from .05 to .5 acre, depending on the experiment. Approxi- mately 70 pounds of barley and 50 pounds of wheat per acre were drilled in a 7-inch spacing. Entire plots were har- vested for yield. Crop varieties used in field studies were Club Mariout barley, Atlas barley, and Ramona wheat. Grain Yield Experiment No, 1. Twelve different tillage methods were compared on an Arlington sandy loam soil in an experi- ment initiated in 1949 at Riverside. Plots were .05 acre in size. Treatments were replicated four times and alternated between two adjacent plot areas to pro- vide for yields each year under the fallow- crop system. Four tillage methods were used in the fall prior to the fallow rain- fall season, one method during the fallow winter period, and six methods in the spring. Depths of tillage were as follows : subsoiler, 12 to 18 inches; chisel and moldboard plow, 6 to 8 inches ; disk and disk plow, 4 to 6 inches; and subtiller and rodweeder, 3 to 4 inches. Except for the nontillage plots, where weeds were sprayed, uniform shallow tillage was per- formed with the rodweeder for sub- sequent weed control. Statistical analysis of the yield data Included use of the Duncan's multiple range test for determining the signifi- cance of differences. Over the four-year < o O 3 ml 5 < n«2 o < II Q a ^ a o 3 O M l« O) OO H r- >0 rH O) t~ CD cc O OO ■* QO 00 OO O *< OO cm 10 OS CO 00 << ■*' B M "3 O © ■£ T3 a T3 IO CO CO CM m t^ O lO •"■* C> U) OO £j O ti Oi 10 to O) CO VJ3 N O (D 00 X! os ■*' 00 2 «} .2 s OO CM CO 10 OO CO IO CO IS Hi O! 115 0O 3 (4 M Tf 0O rt -*fl CO * OO 0O £ Q O *ei 02 CM GO 00 CO CO CO 10 8 s O CO OS ~ a> CM 00 CO 10 ■^ CO W X) 00 a> rQ OJ M ffl Jf 3 aj ci fe 01 IS O lO CO CO CO r- © © as <3> Tft O: O S-. a> '0 03 CM © < 200 lb decrease 100-250 lb 1946.... 1 1 1947.... 1 1 1948.... 1 1 1949.... 1 1 1950.... 3 1 l 1 1951.... 1 1 1952.... 3 2 1 1953.... 3 2 1 1954... 3 1 2 1955.... 1 1 1956.... 4 1 2 1 1957.... 2 1 1 1958... 2 2 1959... 2 1 1 I960.... 4 1 1 1 1 — ■ - — , — Total. 32 10 7 11 4 * Average of 6 to 10 samples. t Tentative soil series. * Yields were average of 2 to 4 replications. Table 26. Barley Yield Response to Applied Nitrogen under Different Moisture Conditions, Western Riverside County, I960 Available* soil moisture at planting Rainfall after planting Soil N0 3 -N in 0-6" depth at planting Barley yield t Location Check 30 lb N/A inches inches lb/A lb/A lb/A Dysart Ranch 6.2 1.0 3.4 7.0 6.3 2.0 3.3 4.9 22 4 6 8 1440 290 350 1060 Christensen Ranch 1210 McSweeney Ranch 290 Thompson Ranch J 460 1600 + Moisture m excess of that held at 15 atmospheres tension, t Average of 3 replications. I Wheat. 26 same when nitrogen is applied. Unfavor- able rainfall distribution toward the end of the growing season produced some drouth stress at all locations. Phosphate was applied at all locations and had no effect on grain yield. At the Dysart Ranch, with favorable moisture until heading and initially high soil nitrates, added nitrogen resulted in a significant yield decrease of 230 pounds per acre. A severe lack of moisture throughout the growing season at the Christensen Ranch precluded any nitro- gen effect. With initially low soil nitrates and relatively moderate moisture, ap- plied nitrogen increased average barley yield 110 pounds per acre at the Mc- Sweeney Ranch. This increase was sig- nificant at the 5 per cent probability level. Greatest positive response from nitrogen occurred where soil nitrates were low at planting and soil moisture was favorable. Under these conditions at the Thompson Ranch, a highly significant increase of 540 pounds of wheat per acre was ob- tained. Moisture stress from heading to har- vest was evident in the Dysart experi- ment. A shriveled appearance of the grain and an average test weight of 41 pounds per bushel verified this condition. Test weights of grain and straw yields were similar for all treatments. It is sug- gested that the excess of available nitro- gen (provided by fertilizing) increased early growth to such an extent that mois- Table 27. Effect of Rate of Nitrogen Application on Grain Yields in the Alternate Fallow — Grain Cropping System * Western Riverside County Year Location Grain yields Check 15-20 lb N/A 30-40 lb N/A 60-80 lb N/A Experiment Station, Riverside Experiment Station, Riverside Experiment Station, Riverside Experiment Station, Riverside Yoder Ranch, Murrieta pounds per acre 1950 1952 1953 1954 1956 710 1500 880 1040 1030 1040 770 1220 460 980 1041 760 1250 490 1190 590 1360 550 1300 1709 1091 930 1040 440 1440 530 2020 630 1170 717 1027 870 1960 Dysart Ranch, Banning McSweeney Ranch, Hemet 980 430 Thompson Ranch, Murrieta 1940 Average yield from 3 or 4 replications. Table 28. Comparison of Nitrogen Carriers at 30 Pounds of Nitrogen per Acre for Barley Production at Six Locations in Western Riverside County* Year Barley yields Check (NH4) 2 SO-4 NH4NO3 Ca(N0 3 ) 2 CO(NH 2 )2 NH 3 pounds per acre 1952 1100 940 1420 1030 630 940 1600 1240 1440 1710 810 1230 1600 1550 1700 850 1040 1680 1540 1780 760 1380 1330 1954 1480 1954 1400 1954 1956 1956 Average of 3 or 4 replications. [27] Table 29. Effect of A 30- to 40-Pound Nitrogen Application on Crude Protein Content of Grain* Location Soil type Crude protein t Year Check Nitrogen Beaumont Beaumont Beaumont Riverside Murrieta Murrieta Wildomar Murrieta Beaumont Beaumont Ramona sandy loam Placentia sandy loam Placentia sandy loam Placentia sandy loam Placentia clay loam Placentia clay loam Hanford sandy loam Las Posas|| loam Placentia sandy loam Placentia sandy loam per cent 1950 8.0 6.8 13.1 13.4 9.4 12.9 9.9 11.9 13.1 11.2 8.7 1950 7.2 19531 13.8 1954 14.1 1954§ 9.8 1956§ 15.4 1956 15.9 1956 12.7 1957J 14.1 1959J ' 12.7 Avg 11.0 12.4 * Plow or disk tillage; ammonium sulfate or ammonium nitrate fertilizer; alternate fallow — crop system. t Per cent nitrogen X 6.25. J Same experimental site. § Same experimental site. || Tentative soil series. ture use was increased and the subse- quent exhaustion of soil moisture oc- curred first in the fertilized plots. Rate of Application Nitrogen application rates were com- pared in ten field experiments from 1950 to 1960, inclusive. Rates ranged from 15 to 80 pounds per acre. The method of application usually consisted of broad- casting, followed by shallow disking be- fore planting. Only in one of five experiments was a marked yield increase obtained from a rate in the 15- to 20-pound range (table 27). Rates between 30 to 40 pounds per acre increased yields in 4 of 10 experi- ments. Significantly greater yield in- creases from 60 to 80 pounds of nitrogen per acre occurred in two of the experi- ments. In three experiments, significant decreases in yield were caused by appli- cation of these relatively high rates of nitrogen. Where nitrogen is broadcast for dryland grain production with the alternate fallow-crop system, rates of from 30 to 40 pounds per acre are prob- ably near optimum on southern Cali- fornia soils. Preliminary studies with nitrogen placement for dryland grain indicate that placement with the seed at planting or below the seed row results in slightly more efficient use than does broadcasting before drilling or after emergence. Plac- ing more than 30 pounds of nitrogen (as ammonium nitrate or urea) with the seed is hazardous under dryland conditions because of frequently observed decreases in emergence of the crop. Fertilizer Sources The following nitrogen fertilizers at 30 pounds of nitrogen per acre were Table 30. Effect of Applied Nitrogen on Barley Straw Production, Arlington Sandy Loam, Riverside Year Rainfall Nitrogen applied (lb/A) None 40 80 inches pounds per acre 1952 17.47 9.58 11.56 1609 1766 1641 1969 1453 2516 3188 1953 1906 1954 3328 Avg 12.87 1672 1979 2807 [28 compared at several locations: am- monium sulfate, ammonium nitrate, calcium nitrate, urea, and ammonia. The source of nitrogen applied usually either had little or no effect on barley grain yield (table 28) . In one of the 1954 experiments where nitrogen markedly increased yield, ammonia was clearly superior to ammonium sulfate. Grain Quality Crude protein content is a criterion for quality of feed grains such as barley. Inasmuch as it is determined by the nitrogen content of the grain, nitrogen fertilizer may be expected to affect the level present. Thirty to forty pounds of nitrogen per acre increased the average crude protein content of barley grain 1.4 per cent in ten field experiments (table 29) . Increases in individual experiments ranged from less than 1 to 6 per cent. Both the soil type and the growing season had a marked effect on the level of crude protein in the grain. The largest increase with nitrogen occurred on the Hanford sandy loam soil. Test weight, a universal criterion for quality in the marketing of grains, was determined in several experiments. Ex- cept in a few instances, the application of 30 or 40 pounds of nitrogen per acre did not significantly affect test weight of the grain. In one experiment, 30 pounds of nitrogen significantly decreased test weight 2.6 pounds per bushel. Similar trends were noted in other experiments, but differences were small and not statis- tically significant. Straw Yield When needed, relatively low rates of nitrogen have previously been shown in this report to be optimum for dryland grain production. Increasing the rate to 80 pounds per acre usually either in- creased yield very little or not at all and, in some cases, resulted in decreases. Higher rates of nitrogen, however, fre- quently produced marked additional in- creases in straw yield (table 30). In two of three seasons at Riverside, 80 pounds of nitrogen doubled straw yields and markedly increased yields over those ob- tained with 40 pounds every year. These results confirm many observations of in- creased vegetative growth with added nitrogen. Data presented in table 25 show no grain yield response to applied nitrogen in about one-third of the experiments over a period of lower-than-average rain- fall. Straw yields, however, were in- creased in several of these experiments (table 31). Consistent and frequently large increases in straw production were obtained with 30 or 40 pounds of nitro- gen even though grain yields were de- creased or not appreciably affected. Straw increases ranged from 370 to 780 pounds per acre. The supply of moisture in the latter part of the growing season fre- Table 31. Grain and Straw Production With Applied Nitrogen Under Limiting Moisture, Western Riverside County Year Grain yield Straw yield Location Check 30-40 lb N/A Check 30-40 lb N/A 1952 1954 1956 1956 1960 pounds per acre Experiment Station, Riverside Experiment Station, Riverside Yoder Ranch 1500 1420 630 1040 410 1360 1440 810 1110 470 1600 1S10 790 1560 810 O Oi ■■£> 3 S O Flvnn Ranch 2340 McSweeney Ranch 1320 [29 quently limits grain yield and a positive response to nitrogen. Residual Effect Under the variable and deficient mois- ture supply encountered with dryland farming, it is reasonable to expect incom- plete use of nitrogen fertilizer applied to the soil. Nitrogen not removed from the soil by the crop could remain in the root zone, be leached from the root zone, and/ or be volatilized and lost to the atmos- phere. Under conditions of low effective rainfall, as in the Pacific Southwest, leaching would likely be less of a factor than in high-rainfall areas. In two experiments near Beaumont, the volunteer barley forage production was measured the year following fertiliza- tion with 30 pounds of nitrogen for the grain crop. A marked residual effect was noted in every case (table 32). Forage yields were increased as much as 1,000 pounds per acre by the residual effect of this relatively low rate of applied nitrogen. Continuous annual grain cropping, practiced in one experiment, revealed a residual effect of nitrogen on grain yield. A 60 per cent yield increase was obtained from the residual effect of 40 pounds of nitrogen applied the previous year, com- pared to a 90 per cent increase for the same rate applied the season of meas- urement. Response to Applied Phosphorus Fourteen field experiments conducted in western Riverside County between 1950 and 1963 included the application of phosphate fertilizer alone and with nitrogen. In only two of these experiments did the application of phosphate increase grain yields. The phosphate was applied with nitrogen and the yield increase over the average of the nitrogen plots was from 200 to 400 pounds of grain per acre. Phosphorus alone had no effect on yield. Table 32. Residual Effect of Nitrogen Fertilizer on Annual Forage* Production, Western Riverside County Year Forage yield Location Check 30 lb N/A applied previous year 1951 1951 1952 1956 1958 pounds per acre Haskell Ranch, No. 1 Haskell Ranch, No. 2 Houston Ranch Houston Ranch Houston Ranch 1570 1730 920 770 3180 1880 1800 1250 1720 4250 * Volunteer barley; air-dry weights. Legume Cover Crops A legume cover crop alternating with grain not only should provide adequate protection for the soil surface, but also should (if turned under as a green ma- nure crop) increase the available nitro- gen for the subsequent grain crop. A purple vetch cover crop was compared to the application of 30 pounds of nitrogen for the grain crop. Considerable data over an 11-year period show that the nitrogen application produced the highest yields in the majority of comparisons (table 33). Only in one case on the higher-producing Placentia soil was vetch markedly better than 30 pounds of nitro- gen. At the Ramona site, yield increases from fertilizer over vetch cropping were statistically significant. Differences on the lower-producing Placentia soil where nitrogen was not limiting were small and usually not significant. Two of the best adapted winter legumes for dryland cropping when this work was initiated were purple vetch and blue lupine. Each of these legumes was grown in alternate years with barley over an eight-year period on a Ramona sandy loam soil. Volunteer barley was permitted to grow on check plots. All cover crop growth was killed between March 15 and April 7. No significant difference between legumes was measured in the yield of the [30 following barley crop (table 34) . Legume cover cropping did result in small but consistently higher barley yields than vol- unteer barley. As previously stated, a cover cropping with legumes has the added advantage of increasing the nitrogen supply in the soil. A major part of this increase results from returning the crop to the soil. Amounts of nitrogen in above-ground parts of several cover crops are shown in table 35. As ex- pected, the legume vegetation contained more nitrogen than the barley. From 5 to 35 pounds more nitrogen per acre were contained in the purple vetch and blue lupine crops. Dryland Crop Response To Nitrogen Fallowing land increases the accumu- lation of soil nitrates between crops. Under dryland conditions in California soils, however, the total supply of avail- able nitrogen after fallow is not always adequate for grain production. Grain yields were increased by nitrogen appli- cation in one-half of the field experiments conducted in southern California over a 15-year period of below-normal rainfall. Martin and Mikkelsen (1960) report positive response to nitrogen in 49 per cent of their nonirrigated grain tests, the majority of which were in central and northern California. Widespread re- sponse of wheat to nitrogen after fallow on dryland areas of Utah has been shown by Peterson (1952). These results con- trast with those obtained in the Great Plains, where fallow grain crops have not generally responded to nitrogen (Staple, 1960). With much higher nitrogen con- tents, Great Plains soils apparently pro- vide sufficient available nitrogen after fallow. Under dryland conditions, application of nitrogen fertilizer without regard to Table 33. Comparison of the Effect of 30 Pounds of Nitrogen per Acre and a Vetch Cover Crop on Barley Yield,* Beaumont Barley yield Soil Year Check! Nitrogen fertilizer! Vetch 1948 pounds per acre Ramona sandy loamf 1236 1646 1393 1950 2424 3004 2746 1952 900 1456 1256 1523 2035 1798 Placentia sandy loamt. . . 1948 1018 1420 1528 1950 2337 2580 3256 1952 935 1344 1096 Avg 1430 1781 1960 Placentia sandy loam§. . . 1949 516 446 556 1951 887 699 889 1953 996 1014 725 1955 1356 1354 1260 1957 380 492 304 1959 1320 1256 1140 Avg 909 877 S12 * Ammonium sulfate fertilizer applied before planting grain crop; purple vetch planted in alternate years with grain crop; heavy disking was primary tillage method. t Volunteer barley crop. t Haskell Ranch. § Houston Ranch. [31] Table 34. Effect of Different Cover Crops on Subsequent Barley Yields, Ramona Sandy Loa.m Soil, Beaumont Barley yields* Cover crop 1953 1955 1957 1959 Avg. pounds per acre 1000 1110 1070 1060 1220 1310 560 590 700 1330 1560 1650 980 1120 Blue lupine 1180 * Average of 4 replications. the available moisture supply or the ca- pacity of the soil to make nitrogen avail- able may result in no response or a decreased grain yield. Grain crops fre- quently show a vegetative color and growth response to added nitrogen. Straw yields reflect this increased vegetative growth. If rainfall is deficient in the spring and the stored soil moisture is exhausted before the grain crop matures, either no response or a decrease in grain yield may result. Ramig and Rhoades (1963) found that when available mois- ture was low, nitrogen increased straw more than grain production. Under seri- ously limiting moisture conditions in Oregon, Hunter and co-workers (1961) reduced wheat yields by applying rela- tively low rates of nitrogen. Fall applica- tions were more effective in reducing yields. The evidence suggests that de- creases in dryland wheat yields from applied nitrogen result from earlier ex- haustion of available moisture, which may be caused by additional early growth, and/or generally taller and more luxuriant vegetative growth throughout the season. A higher probability for re- sponse of dryland grain crops to nitrogen undoubtedly could be achieved by con- sidering the amount of available moisture in the soil; time of application with re- spect to the current rainfall season; and amount of soil nitrates or the nitrifying capacity of the soil. An application of nitrogen appears to be required for dryland grain in southern California before a response to phosphate fertilizer will be obtained. The limited response to phosphorus indicates an ade- quate supply in several of the soils studied under the prevailing low moisture con- ditions. Marked positive residual effect of ni- trogen fertilizer on forage production the year following a grain crop indicates that under a fallow-grain-annual pasture crop- ping system there is some compensation Table 35. Nitrogen Content of Different Leguminous Cover Crops, Western Riverside County* Year Cover crops Location Volunteer barley Purple vetch Blue lupine Bur clover Bitter clover Bicolor lupine 1953 1953 1954 1956 pounds nitrogen per acre Murrieta 19.7 15.9 12.2 25.5 29.3 40.2 17.6 27.0 18.2 51.3 27.8 32.1 21.2 Murrieta 31.8 Beaumont Average of a minimum of seven J^-milacre sa mples per plot taken in late March or early April. [32] for low efficiency of nitrogen use on grain in dry years. Legume cover crops, such as purple vetch and blue lupine, increase the nitro- gen content of the soil for a short period after they are incorporated. Under low rainfall in southern California, the amount of nitrogen in the tops of these legumes frequently has not exceeded 50 pounds per acre. Both purple vetch as a cover crop and 30 pounds of nitrogen as ammonium sulfate increased barley yields significantly. Higher yields were more frequently obtained with the fertilizer. In the low rainfall areas for dryland crop- ping, annual legumes frequently are not a dependable source of nitrogen. Defi- cient rainfall in the fall or early winter prevents emergence and results in limited growth. A cover crop cannot be permitted to grow beyond the early spring months because depletion of stored soil moisture probably will decrease the subsequent grain yield. Judicious use of nitrogen fertilizer can undoubtedly increase the efficiency of moisture use in dryland grain production. Recommendations for a nitrogen fertili- zation program are complicated because of the variability in moisture supply, both within and between cropping sea- sons. Consideration can be given to nitro- gen application when the crop is in the seedling stage if more moisture suddenly becomes available. Results reported here indicate that rates in excess of 40 pounds of nitrogen per acre are rarely justified in southern California. This is in agreement with the findings of Martin and Mikkelsen (1960) for other areas of the State. LITERATURE CITED Bennett, W. H., D. W. Pittman, D. C. Tingey, D. R. McAllister, H. B. Peterson, and I. G. Sampson 1954. Fifty years of dryland research at the Nephi Field Station. Utah Agr. Expt. Sta. Bull. 371:1-81. Bracken, A. F., and J. E. Greaves 1940. Losses of nitrogen and organic matter from dry-farm soils. Soil Sci. 51 ; 1-15. Brady, H. A. 1960. An evaluation of tillage practices for winter wheat in the semiarid region of southwestern Kansas. Soil Sci. Soc. Amer. Proc. 24: 515-518. Chepil, W. S. 1944. Utilization of crop residues for wind erosion control. Sci. Agr. 24; 307-319. Donnelly, M. 1943. Monthly rainfall distribution in southern California with special reference to soil-erosion problems. Trans. Amer. Geophys. Union (1942) 23; 144-148. Duley, F. L., and J. C. Russel 1939. The use of crop residues for soil and moisture conservation. Agron. J. 31: 703-709. Haas, H. J., C. E. Evans, and E. F. Miles 1957. Nitrogen and carbon changes in Great Plains soils as influenced by cropping and soil treat- ments. U. S. Dept. Agr. Tech. Bull. 1164 : 1-1 11. Harris, W. W. 1963. Effects of residue management, rotations, and nitrogen fertilizer on small grain production in northwestern Kansas. Agron. J. 55: 281-284. Hobbs, J. A., and P. L. Brown 1957. Nitrogen and organic carbon changes in cultivated western Kansas soils. Kansas Agr. Expt. Sta. Tech. Bull. 89:1-48. Hunter, A. S., L. A. Alban, C. J. Gerard, W. E. Hall, H. E. Cushman, and R. G. Petersi n 1961. Fertilizer needs of wheat in the Columbia basin dryland wheat area of Oregon. Oregon Agr. Expt. Sta. Tech. Bull. 57: 1-59. Johnson, W. C. 1950. Stubble-mulch farming on wheatlands of the southern high plains. U. S. Dept. Agr. Circ. 860; 1-18. 33 LlTTLEJOHN, L. 1956. Some aspects of soil fertility in Cyprus. Emp. J. Expt. Agr. 14; 123-134. Martin, W. E., and D. S. Mikkelsen 1960. Grain fertilization in California. California Agr. Expt. Sta. Bull. 775: 1-39. Mathews, 0. R. 1951. Place of summer fallow in the agriculture of the Western States. U. S. Dept. Agr. Circ. 886: 1-17. McCalla, T. M., and T. J. Army 1961. Stubble mulch farming. Advances in Agronomy 13: 125-196. Edited by A. G. Norman. Academic Press, Inc., New York. Mehring, A. L. 1945. Fertilizer nitrogen consumption. Ind. Eng. Chem. 37 ; 289-295. Peterson, H. B. 1952. Effect of nitrogen fertilizer on yield and protein content of winter wheat in Utah. Utah Agr. Expt. Sta. Bull. 353 : 1-29. Ramig, R. E., and H. F. Rhoades 1963. Interrelationships of soil moisture level at planting time and nitrogen fertilization on winter wheat production. Agron. J. 55; 123-127. Siddoway, F. H., H. C. McKay, and K. H. Klages 1956. Dryland tillage, methods and implements. Idaho Agr. Expt. Sta. Bull. 252: 1-46. Staple, W. J. 1960. Significance of fallow as a management technique in continental and winter rainfall climates. UNESCO Arid Zone Research 15:205-213. Taylor, S. A. 1960. Principles of dryland crop management in arid and semi-arid zones. UNESCO Arid Zone Research 15; 191-203. Zingg, A. W., and C. J. Whitfield 1957. Stubble-mulch farming in the Western States. U. S. Dept. Agr. Tech. Bull. 1166: 1-56. 7%m-8,'64 (5908) J.F. KNOWLEDGE GAINED BY RESEARCH CAN HELP CONSERVE CALIFORNIA'S WILDLAND RESOURCES CALI THE FORNIA WILDLANDS... 65 million acres of mountains, foothills, canyons, rivers, lakes, and sea coasts. a giant "farm" for timber and forage. a vital source of California's water supply. an "outdoor playground" for millions of vacationers. THREAT: the onslaught of... population growth. urban and industrial expansion. increasing demand for water, lumber, forage. wildfires. • insects and plant and animal diseases. • waste. '/ THE SOLUTION: coordinated research on using wildland resources to realize their full potential . . . present rate of timber growth could be doubled. usefulness of timber cut could be doubled by new products made from current waste. forage production for livestock and game could be tripled, watersheds could be made to yield more usable water and cause fewer floods, tens of millions of dollars lost to fire, insects, diseases could be saved. - timber, forage, and recreation uses need not exclude each other. THE WILDLAND RESEARCH CENTER at the University of California was established to help conserve California wildland resources through research. It operates within the University's state-wide Agricultural Experiment Station, with administrative headquarters on the Berkeley Campus. THE CENTER... • coordinates and supports research in more than a dozen fields. • integrates studies of complex wildland problems. • strengthens cooperation between University and other research workers. • promotes the exchange of information between research workers and wildland managers and policy makers. • collects and disseminates scientific data on wildland studies. TO KNOW IS TO LIVE IN ABUNDANCE...