Maataloustieteellinen A ikakauskirja Vol. 60: 631—660, 1988 Testing of a Danish growth model for barley, turnip rape and timothy in Finnish conditions ARI ILOLA 1 , ESKO ELOMAA2 and SEPPO PULLI 3 ' Department of Crop Science, Agricultural Research Centre SF-31600 Jokioinen, Finland 2 Technical Department, Finnish Meteorological Institute P.O. Box 503 SF-OOWI Helsinki, Finland 3 Department of Plant Breeding, Agricultural Research Centre SF-31600 Jokioinen, Finland Abstract. The biological and meteorological data were collected at Jokioinen in 1982—87. Potential and actual (water limited) production of dry matter were simulated using a Danish WATCROS model for spring barley, spring turnip rape and timothy grass. The most important data of the biological programme comprised weekly measurements of the crop surface (GAI), dry matter yield, root growth, soil water content and yield analyses of the harvest. All these measurements were performed for both irrigated and non-irrigated plots. The needed meteorological parameters for the daily simulation of the dry matter yield were global radiation, air temperature and precipitation. The simulated dry matter production results with the WATCROS model were generally higher than those measured. In order to obtain a better fit into the Finnish climatic and soil conditions, the Finnish model should take soil water conditions and efficient use of photosyn- thetically active radiation into consideration. Index words: Finland, crop growth, crop production, simulation, barley, turnip rape, timothy Contents Abstract 631 Symbols 632 I. INTRODUCTION 633 2. THE EXPERIMENTAL FIELD 633 2.1. Layout of the field 633 2.2. Soil properties 633 3. CLIMATE 634 3.1. General description of the meteorological measurements 634 631 JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND https://www.c-info.fi/en/info/?token=PwtrpGhZQZqtgJ6n.pmRM8yiPjQXxRTWgQst0yQ.2Q7zs9CkSTCVccZ0WuvnvBb4KvEnU_YhMBqYmq6SRjjMeaX-ts8YGN35gSE9n8tmkd5RLxPEoA6Rw63Fnada4fK1pCK0h5sO21M_uHdGcBzAmQEDkaVUktJ23c0BlqafgjeDMthOzReLAyCXr9nFh13yO_b__IVYnhfVXGYHXmaGLS-ISTZDAv4lt2_U7t3cvYwgHJ8 3.2. Solar radiation 3.3. Air and soil temperature 637 3.4. Air humidity 639 3.5. Wind speed and direction 639 3.6. Potential evapotranspiration 639 3.7. Precipitation 640 3.8. Precipitation deficit 640 4. THE BIOLOGICAL PROGRAMME 640 5. PLANT GROWTH AND DEVELOPMENT 641 5.1 Crop surface 641 5.2. Root growth 641 5.3. Dry matter production 643 6. RESULTS OF THE END HARVESTINGS 644 6.1. Barley and turnip rape 644 6.2. Timothy 644 7.1. Simulation of the crop area index 645 7.2. Actual evapotranspiration 646 7.3. Potential gross production 650 7.4. Respiration and net plant production .. 651 7.5. Water limited plant production 653 8. RESULTS AND DISCUSSION 653 8.1. Barley and turnip rape 653 8.2. Timothy 656 9. CONCLUSIONS 658 ACKNOWLEDGEMENTS 658 REFERENCES 659 SELOSTUS 660 637 7. THE MODEL 645 Symbols Symbols in the text marked with an asterisk denote ca- pacities or potential values. A Gross C02 single leaf assimilation, photosynthesis A m Gross C02 single leaf assimilation, photosynthesis, at light saturation Av Albedo visible radiation (400 —700 nm), light c Factor converting stored energy into structural plant dry matter C Crop area index dr Maximum effective root depth D Slope of the saturation vapour pressure curve ver- sus temperature D p Precipitation deficit e Vapour pressure e. Saturation water vapour pressure esa Saturation vapour pressure, at dry-air temperature e,„ Saturation vapour pressure, at wet-bulb tempera- ture E Evapotranspiration E c Evapotranspiration from the crop Ec 8 Evapotranspiration from the crop, green active area Ec Evaporation from the crop, yellow inactive area E, Evaporation from the soil Et Transpiration from the crop G Crop area index, green active area Gg Ground heat flux Gm Maximum green area index H Harvest index I Irrigation k Extinction coefficient of PAR K Extinction coefficient of net radiation L Latent heat of the vaporization of water m Constant p Gross photosynthetic efficiency P Precipitation Pg Gross production P„ Net production rg Growth respiration coefficient r m Maintenance respiration coefficient R g Growth respiration R m Maintenance respiration R„ Net radiation above grass S Radiative flux density below the downward accumu- lated GAI (Green area index) Sabs Absorption of photosynthetic active radiation (PAR) S, Global radiation S„ Radiative flux above the canopy Sv Visible radiation (400—700 nm) fraction of the global radiation (300 —2500 nm) S, Topsoil water capasity S, Root zone water capacity S, Storage, intercepted water S, g Storage, intercepted water, green active area S, y Storage, intercepted water, yellow inactive area t time ta Dry-air temperature, °C t„ Wet-bulb temperature, °C ts Soil temperature, °C v Wind speed W Total dry matter in the field W h Harvested dry matter yield W, Non-harvested dry matter (stubble, root, etc. mass loss) Y Yellow area index Yp Psychrometric constant 632 I. Introduction The joint Nordic project on the effect of climatological factors on crop growth and production was started in 1982. The research programme was planned by a working group of the agricultural meteorology of Section I of the Association of the Agricultural Scien- tists of Scandinavia. It was carried out in Den- mark in 1982—85, in Norway in 1982—86 and in Finland in 1982—87. It was funded by the national authorities (in Finland by the Acade- my of Finland). Field experiments of climatic field (the ex- perimental field) were carried out in 1982—87 at Jokioinen, in SW- Finland, to test Danish growing models for various crops. The main aim of the project was to calculate potential and water-limited crop growth and produc- tion. Danish models constructed by Aslyng and Hansen (1982), in modified forms, were used as the basis of calculations. The models are based on experimental results of the week- ly dry matter and crop area index (CAI) mea- surements and daily measurements of the climatological and hydrological factors of the experimental field. The models are simple enough to be used for routine monitoring of changes during the growing season and of the production of various crops. Another aim of the project was to test the Danish models. After six years of experimental work and one year of research work, results can be giv- en for spring barley (barley), spring turnip rape (turnip rape) and timothy grass. Some de- tails concerning the project have been pub- lished previously (Elomaa and Pulli 1985, Saarinen et al. 1986, Elomaa et ai. 1986, Elomaa 1987). 2. The experimental field 2.1. Layout of the field Three species, barley, turnip rape and perennial grass timothy, were tested in irrigat- ed and non-irrigated plots (Fig. 3.1.). Detailed crop and soil observations were made for each of six plots from 1982 to 1987. 2.2. Soil properties The topsoil (0—20 cm) was classified as heavy clay with 7—ll % organic matter. The subsoil was defined as heavy clay lacking C compounds and phosphorous, but rich in magnesium and calcium (Tables 2.1, 2.2). Chemical analyses of the plots were performed annually since 1983 (Table 2.3). Plots A and C were limed in spring 1986 and 1987. In 1984, the water retention capacity of the soil profiles was studied, and the entire soil moisture retention curve was determined. The water capacity usable by plants was 15—20 percent of volume, depending on the plot and soil depth (Table 2.4). The hydraulic conductivity of the soil was measured with the MSU (Michigan State University) method in 1987 (Saavalainen and Rintanen 1986). Normal reliable (R> 0.95) measurements were 0.15—0.38 (mean 0.22) cm water in one hour. The heterogenei- ty of the plots and the depth of the soil as well as pore holes caused some variation between the measurements (0.02—4.83 cm h-1)- During the growth period, the soil moisture content of each plot was monitored weekly. In 1982 soil moisture was measured with gyp- sum blocks at five soil depths. Irrigated plots 633 Table 2.1. Chemical analysis and physical properties of soil layers in 1985. pH mg/l per cent BulkPlot Depth Density C °r«' N ' g cm - ’ cm Ca K Mg P matter AI 00— 20 5.7 2175 265 20— 40 5.8 2600 310 40— 70 6.7 2750 255 70—100 6.8 2600 305 A 2 00— 20 5.8 2375 225 20— 40 5.9 2375 185 40— 70 6.4 3325 245 70—100 6.9 3000 287 Bl 00— 20 5.6 2250 282 20— 40 5.8 2250 277 40— 70 6.2 2375 225 70—100 6.7 2550 267 B 2 00— 20 5.5 2275 350 20— 40 5.6 1625 187 40— 70 6.4 3000 240 70—100 6.6 2850 265 Cl 00— 20 5.7 2450 385 20— 40 5.7 2275 340 40— 70 6.5 2775 245 70—100 7.0 2625 270 C 2 00— 20 5.7 2450 385 20— 40 5.8 1875 245 40— 70 6.2 2775 237 70—100 7.0 2925 275 KEY: l=Year 1986; Plot 1 = irrigated, 2 =non-irrigated 542 7.0 3.6 7.9 0.20 1.25 1250 1.9 1.8 3.8 0.04 1.31 1775 0.6 0.4 2.5 1.37 1850 0.8 0.3 3.0 1.32 585 5.4 3.3 7.2 0.23 1.34 825 2.1 1.9 3.8 0.09 1.35 1850 1.0 0.6 3.1 1.32 1775 1.4 0.3 2.8 1.32 515 6.6 4.2 8.8 0.26 1.27 877 3.8 3.0 7.2 0.04 1.29 1500 1.0 1.3 3.5 1.32 1825 0.8 0.3 2.8 1.35 475 8.1 4.6 10.4 0.23 1.33 615 1.7 2.0 9.3 0.04 1.30 1800 0.5 0.4 3.6 1.31 1750 0.6 0.3 2.7 1.35 600 7.4 4.6 9.4 0.26 1.17 1200 3.3 2.7 3.4 0.07 1.32 2000 0.7 0.4 3.1 1.31 2025 0.9 0.3 2.8 1.32 610 7.1 4.6 10.0 0.26 1,18 790 2.2 2.7 4.9 0.07 1.27 1925 1.1 0.6 3.3 1.27 2125 1.2 0.3 2.8 1.33 were watered (plots 1, Fig. 3.1) if the soil moisture content available for plants was be- low 50 % (Table 2.5). Gypsum blocks were not reliable after winter, and soil moisture could not be measured in 1983. In 1984 soil water content was measured gravimetrically, taking soil samples from each plot, to a depth of 50 cm. Because 1984 was a rainy year, no irrigation was needed. In 1985 the soil water content was measured both gravimetrically and with the neutron scattering method, BASC depth moisture probe (Table 2.6). 3. Climate 3.1. Genera! description of the meteorological measurements Solar radiation and wind direction were measured at the top of a meteorological mast situated beside the experimental field. Profiles of wind speed and dry-air and wet-bulb tem- peratures were also measured at the mast. Short-wave solar radiation, air and soil tem- perature, air humidity and soil moisture were measured for each experimental plot (Fig. 3.1). In 1984gypsum blocks were removed for soil moisture measurements; they were Table 2.2. Mechanical analysis of the soil layers in the experimental field in 1987. Depth Weight % Clay Sill Fine Coarse Sand Sand 00— 20 63.0 13.8 20.0 3.2 20— 40 71.8 11.1 13.1 4.0 40— 70 75.0 10.1 13.8 1.1 70—100 87.2 3.1 8.6 1.1 634 Table 2.3. Chemical analysis of the tillage layer of the plots. Plot Year pH mg/I Ca K Mg P AI 1983 6.3 2590 310 621 6.3 1984 6.2 2330 325 625 6.0 1985 5.7 2175 265 542 7.0 1986 6.2 2279 262 615 5.0 1987 6.2 2945 292 628 6.7 A 2 1983 6.1 2660 300 631 5.4 1984 6.1 2555 288 683 4.9 1985 5.8 2375 225 585 5.4 1986 5.8 2255 334 498 8.0 1987 6.2 3304 333 752 9.2 Bl 1983 5.9 2390 338 616 5.7 1984 5.9 2165 345 572 6.2 1985 5.6 2250 282 515 6.6 1986 5.7 2190 294 478 8.8 1987 6.1 2787 307 822 5.7 B 2 1983 5.9 2320 326 515 6.0 1984 6.0 2190 329 553 5.2 1985 5.5 2275 350 475 8.1 1986 5.7 2090 285 449 7.9 1987 6.0 2846 296 892 5.6 Cl 1983 6.1 2680 348 727 6.7 1984 5.8 2135 338 683 4.5 1985 5.7 2450 385 600 7.4 1986 5.8 2303 296 714 7.5 1987 6.0 3030 419 714 7.0 C 2 1983 5.9 2480 331 595 6.9 1984 5.9 2255 348 590 5.7 1985 5.7 2450 370 610 7.1 1986 6.2 2734 260 673 6.2 1987 6.2 2962 345 691 7.2 Table 2.4. Soil capasity for available water mm in differ- ent soil layers for I cm, 25 cm and effective root depth. Plot Soil Depth (cm) 0—25 25—50 50—75 0—75 A I cm 1.6 1.6 1.5 25 cm 40 40 37 117 B 1 cm 1.7 1.7 1.6 25 cm 42 42 40 124 C 1 cm 2.0 1.8 1.6 25 cm 50 45 40 135 Table 2.5. Irrigation schedule in 1982—87. Date, irrigation (mm) 1982 1983 1985 1986 1987 Barley 27/7 45 18/7 35 11/6 25 18/6 35 1/6 10 3/8 10 1/8 30 28/6 30 2/7 50 21/7 25 9/7 25 Sum: 55 65 80 85 35 Turnip rape 27/7 45 19/7 35 13/6 30 23/6 30 22/7 25 3/8 10 2/8 30 28/6 30 1/7 50 10/7 20 Sum: 55 65 80 80 25 Timothy 20/7 35 12/6 25 16/6 50 20/7 25 3/8 30 27/6 30 1/7 50 9/7 30 4/8 20 Sum: 65 85 120 25 replaced by pyranometers to measure the reflected short-wave radiation of each plot. A calculating data logger, an Autodata Ten/5 made by Acurex (USA), was used for data-logging. A one-minute scanning interval was used to measure the meteorological vari- ables. Hourly mean values were stored in the C-cassettes of an MFE 2500 tape recorder. The C-cassettes were converted to magnetic tapes for further analysis. The climatological measurement results for the experimental field were compared to the Table 2.6. Measuring depths of soil moisture in 1982—87. Year Management Depth (cm) 1982—1983 Gypsum Block —lO, —2O, —3O, —SO, —IOO 1984— 1987 Gravimetrically —lO, —2O, —3O, —SO 1985 Neutron Scattering —lO, —2O, —3O, —4O, —SO, Method (BASC) —6O, —BO, —IOO KEY: 1 =Year 1987: —ls, —3O, —45, —6O, —75, —9O 635 Fig. 3.4. Cumulative sum of potential evapotranspira- tion at Jokioinen (medians and probability limits). Fig. 3.1. Meteorological observations on the experimental field. S = global radiation u =wind speed d =wind direction Ta =dry-air temperature T w =wet-bulb temperature Rc =reflected short-wave radiation r =relative humidity T =soil temperature Fig. 3.5. Cumulative sum of precipitation deficit at Jokioinen (medians and probability limits). Fig. 3.2. Cumulative sum of global radiation at Jokioinen (medians and probability limits). Fig. 3.3. Cumulative sum of effective growing tempera- ture at Jokioinen (medians and probability limits). 636 637 Table 3.1. Monthly cumulative sums of global radiation (MJ/m!) at Jokioinen Observatory in April—October. Mean 1957—1983 1983 1984 1985 1986 1987 April 391 309 412 419 343 464 May 578 453 601 585 544 436 June 639 581 592 585 680 416 July 573 638 535 556 578 642 August 441 537 455 391 377 356 September 242 240 180 264 247 217 October 105 93 95 123 109 108 Table 3.2. Mean air temperature (°C) at Jokioinen Observatory in April—October. Mean 1957—1984 1983 1984 1985 1986 1987 April 2.0 4.8 4.2 0.5 2.1 2.4 May 8.7 11.0 12.6 8.6 10.5 7.6 June 14.0 13.3 13.1 13.2 16.3 12.1 July 15.6 16.6 14.8 15.3 16.2 14.8 August 14.2 15.0 13.8 15.5 12.9 11.7 September 9.3 11.0 9.2 8.9 6.4 8.4 October 4.4 5.4 6.6 6.4 5.2 6.4 Table 3.3. Mean soil temperature (—lO cm, °C) at Jokioinen Observatory in May—September. Mean 1957—1983 1983 1984 1985 1986 1987 May 7.5 9.1 9.9 4.7 8.5 5.8 June 13.8 13.7 14.3 12.6 14.4 11.7 July 16.0 16.2 16.2 15.2 16.3 15.8 August (15.1) 15.2 15.5 15.6 14.7 13.2 September (10.8) 11.8 10.6 10.3 8.5 9.8 (1957—1970) measurements made by the Meteorological Observatory at Jokioinen, 1 km from the ex- perimental field. 3.2. Solar radiation Short-wave solar radiation was measured with Kipp & Zonen pyranometers, which were calibrated with the pyranometer used at the nearby Meteorological Observatory. So- lar radiation was measured at the top of the mast (global radiation) and inside the stand, at a height of about 5 cm above ground level, with the same type of pyranometers used for estimating the extinction of solar radiation in the stand. In 1985—87, short-wave reflected radiation was also measured above each plot. During the growing periods of 1982—86 the sum of global radiation was rather stable from year to year (Fig. 3.2). The highest values were registered in 1983, the lowest in 1987. 3.3. Air and soil temperature Dry-air and wet-bulb temperatures were measured with Pt-100 sensors; the ventilation 638 Table 3.4. Mean wind speed (m s ') at Jokioinen Observatory in May—October. Mean 1957—1980 1983 1984 1985 1986 1987 May 3.9 3.2 3.4 3.6 4.3 4.4 June 3.8 3.4 3.4 3.4 3.9 4.1 July 3.4 3.5 2.9 3.4 3.5 3.4 August 3.3 3.4 2.7 3.9 3.8 3.7 September 3.8 4.5 3.6 3.8 3.9 3.4 October 4.0 4.4 3.8 4.3 4.5 4.7 Table 3.5. Potential evapotranspiration (PET, mm) at Jokioinen Observatory in May—October. Mean 1929—1986 1983 1984 1985 1986 1987 May 59 56 74 58 63 47 +/ 8 June 107 94 94 92 139 73 + /—lB July 113 114 76 104 120 111 + /—24 August 90 108 80 80 70 62 + /—23 September 41 43 27 45 36 30 + / 7 October 20 22 18 32 21 25 +/ 5 May—October 430 437 369 411 449 348 Table 3.6. Monthly precipitation (mm) at Jokioinen Observatory in April—October. Mean 1957—1983 1983 1984 1985 1986 1987 April 32 22 18 32 38 5 May 40 44 66 43 52 38 June 48 84 113 41 11 81 July 77 41 91 55 65 68 August 79 58 69 119 110 83 September 66 86 77 51 102 120 October 68 62 99 36 74 43 May—October 378 375 515 345 414 438 of the psychromelers was centralized, and was some 3ms 1 . Soil temperature was mea- sured with Pt-500 sensors. According to the effective temperature sum in degree days (ETS), with a threshold tem- perature of 5.O°C, the beginning of the grow- ing seasons were warmer than average in 1983, 1984and 1986, but in 1985 and 1987 they were cooler, and temperatures also remained cool throughout most of these two seasons. Fairly high ETS values were observed in 1984, in June and July 1986 and in July 1987 (Fig. 3.3). Soil temperatures in May 1985 and May 1987 were 2—3°C below average (Table 3.3). 3.4. Air humidity Air humidity was measured psychrometri- cally at the mast by using dry-air tempera- tures. Water vapour pressure and relative hu- midity were calculated as follows. Saturation water vapour pressure (e 5, hPa) was calcu- lated (Morton 1975): (3.1) e s = 6.11 x exp (17.27 x ta/(ta + 237.3)) where ta = dry-air temperature (°C). Water vapour pressure (e) was calculated: (3.2) e = esw —0.67 (ta —t w ) where esw =saturation water vapour pressure at wet-bulb temperature (t w ). Values for relative humidity (r) were calcu- lated using the following formula: (3.3) r =e/esa where esa = saturation water vapour pressure at dry-air temperature ta . The air humidity in crops was measured us- ing Humicap HM2I sensors constructed by Vaisala Oy. 3.5. Wind speed and wind direction Wind speed was measured at four levels of the mast, using WAAIS sensors made by Vaisala. The top of the mast had a crossarm assembly to support an anemometer WAAIS and a wind vane WAVIS. 3.6. Potential evapotranspiration By using a modified version of Ivanov’s equation (Ansalehto et al. 1985), a long ser- ies of potential evapotranspiration (PET) values at Jokioinen (1929—87) was calculated for comparison. Modification was made in order to obtain the best fit for comparisons with the PET values determined with the Penman equation (Penman 1956). For the whole growing season, the cumula- tive sum of the daily PET values was lower than average in 1984—1987. The values for 1983 were on the average level. In 1984 and 1985 there were, however, periods when the PET sum was higher than average (Fig. 3.4). Makkink (1957) proposed the following equation for estimating potential evapotran- spiration (E*) from grass: (3.4) E*=o.6l ! 0.12 mm/day D+ Yp L where D = the slope of the curve of the satu- ration vapour pressure vs the temperature, Yp = the psychrometric constant (0.67 hPa/K), Sj = global radiation and L = the la- tent heat of vaporization of water. Aslyng and Hansen (1982) used a simpli- fied version for the calculation of E*: n s (3.5) E* = 0.7 —- D+ Yp L We have calculated D using the formula of Morton (1975) (equation 3.1) and L using that of Hankimo (1964): (3.6) L = 2494-2.29 xta where ta = the dry-air temperature at a height of 2 m. In the EVAPO submodel of WATCROS the following formula, which is the simple average of equations 3.1, 3.5 and 3.6, has been found to be satisfactory: (3.7) E* =0.606 (0.399+ 0.0139 ta ) S/2.47. For the WATCROS model E* was calcu- lated with the modified version of Penman (1956), too (Aslyng 1976) (3 8) E* P(R n — °r) , Y f(v)(e.-ea ) L(D + YP) D + Yp where R n = the net radiation above grass, Gg = ground heat flux and f (v) = the function of wind. (3.9) f (v) =0.263 (0.5 +0.54 v) where v = the mean wind speed, ms Daily net radiation values were given by the nearby Meteorological Observatory. The ground heat flux values were estimated in one- week intervals during the growth period in 639 1983—85, using measurements by Kulmala (1970): (3.10) Gg = -5.5-7.959 dt s where dts =ts 2—ts , and ts | = the mean soil temperatures in soil layers 10...- 100 cm. In the last two study years, 1986—87, Gg was calculated as earlier, but the daily values were computed according to the distribution of net radiation. 3.7. Precipitation Precipitation was measured both manually and automatically. Manually observations were made using a Finnish standard gauge, Tretyakov, at one point on the field. Precipitation was recorded automatically on both the non-irrigated and the irrigated plots, using a tipping bucket rain gauge with a reso- lution of 0.1 mm. 3.8. Precipitation deficit Precipitation deficit (D p) was calculated as follows: (3.11) Dp = E*-P where E* = the potential evapotranspiration (PET) and P = precipitation. In 1983—1987 the precipitation deficit dur- ing the growing season was less than average; 1984 in particular was very wet. Only in 1986 was there a period, in June—August, when the precipitation deficit was greater than aver- age (Fig. 3.5). 4. The biological programme The Nordic research programme (1982—85) wanted to include plant species common to all participating countries. In Finland the Porno cultivar was used for the barley tests, and Tar- mo was the timothy variety used in 1982—87. The turnip rape cultivar used as the test plant was Span in 1982—86 and Kova in 1987. Of these plants only rape (Span) was cultivated in Denmark and Norway, too. Plots of barley and turnip rape were estab- lished in the standard way each year. Timo- thy stands were clear seeded in 1982 and es- tablished with a cover crop barley in 1984. Be- cause of the clear seeding and winter damages (Table 4.1), the growing seasons of 1982 and 1984 were discarded in timothy modelling. The barley plant stands were been quite dense in all years exept 1982. Turnip rape stands sprouted poorly throughout the study (Ta- ble 4.2). Barley and turnip rape were fertilized with NPK (16-7-13) and timothy with NPK (20-4-8) fertilizers. The amounts of nutrients as kg per hectare for barley and turnip rape were 80—100 kg nitrogen (N), 35—40 kg phos- phorous (P) and 40—65 kg potassium (K). For grass, the amounts after the year of establish- Table 4.1. Wintering and Total available Carbohydrates (TAC) in root DM of timothy. Year Plot TAC % Stand Density % Spring Autumn Spring Autumn 1983 Cl 10.3 26.8 89 79 C 2 9.8 26.8 98 75 1984 Cl/Bl 9.1 18.6 65 >9O C2/B2 7.2 18.9 25 >9O 1985 Bl 7.3 17.1 85 85 B 2 7.6 18.2 81 80 1986 Bl 6.8 6.8 72 68 B 2 3.5 7.8 73 76 1987 Bl 6.1 8.2 63 75 B 2 4.1 10,0 70 75 640 Table 4.2. Shooting of barley and turnip rape in 1982—87 per m- 2 . Tear Barley Turnip rape 1982 333 108 334 179 1983 520 288 427 256 1984 587 245 501 269 1985 518 282 500 320 1986 437 270 507 287 1987 539 410 512 389 Mean 1983—87 505 300 ment were: first cut 100—110 kg N, 20 kg P and 40—60 kg K; second cut 80 kg N, 20—35 kg P and 40—65 kg K; third cut 40—60 kg N, 20—30 kg P and 30—50 kg K. In 1982 and 1984 timothy was established by using 500 kg per hectare of NPK (16-7-13) fertilizer. Plant protection was considered important for avoiding the influence of weeds, plant dis- eases and pests on yield and crop green area. The chemicals used and the timing of their sprayings are shown in Table 4.3. In 1984, tur- nip rape was sowed twice, but insect pests also caused some damage to the second plant stand despite protection. During the growing season, the plants were monitored according to the programme of bi- ological measurements. The central measure- ments were made weekly, except for some parameters which were monitored infrequent- ly during the growth period or only at the time of harvest (Table 4.4). 5. Plant growth and development 5.1. Crop surface Instead of the leaf area index (LAI), Aslync and Hansen (1982) adopted the total crop area index (CAI), the green area index (GAI) and the yellow area index (YAI). These indices are the accumulated areas of leaves, stalks, stems and ears divided by the cor- responding land surface. The total crop area influences the interception of radiation and precipitation, and the total green area cor- responds to photosynthesis. The growth period here is defined as the period from emerging to ripening for barley and turnip rape. In the case of timothy, growth is assumed to start at the beginning of the thermal growth period (>5°C) and to end at the time of the last harvest. The green and yellow crop area of studied plant species was measured with an automat- ic leaf area meter (HAYASHI DENKOD AAM-7). The yellow crop area was measured during the years of 1985—87). In the best years, timothy grass reached a GAI value over 15, barley near 10 and turnip rape only 8 (see chapter 8). 5.2. Root growth In 1982—84 only the amount of main roots of the tillage layer (o—2o cm) was measured at the time of cutting in the autumn. From the year 1985 on, root growth was monitored more carefully in order to learn how fast the Table 4.3. Plant protection schedule in the experimen- tal field. TimothyYear Turnip rape Barley 1982 23/6 Decis 6/7 Dipro 2/7 Decis 23/8 PCNB 9/7 Sumicidin 28/12 Avicol 1983 13/6 Decis 9/6 Dipro 25/5 Actril S 20/6 Decis 1984 18/6 Decis 18/6 Actril S 4/7 Decis 1985 19/6 Butisan 19/6 Dipro 16/5 Actril S 19/6 Decis 19/6 Roxion 24/6 Decis 27/6 Decis 4/7 Decis 1986 13/6 Roxion 18/6 Roxion 1987 26/6 Decis 23/6 Herbalon 3/7 Decis 641 642 Table 4.4. Biological programme for the experimental field. Management Barley Turnip rape Timothy Seeding rate 180 8 25 (kg/ha): (500/m2 ) 1987, 12 Seeding depth (cm): 3—5 3 1.5 Emergence date: when 50 Vo sprouting CAI (starting at 10 cm height, whole crop): 6x30 cm at raw weekly Fresh weight (starting one week after CAI): 3x 1.5 m 2 weekly Cutting height (cm): 5 DM determinations: 2 x 200 g IOO°C Height measurements: 5 points/1.5 m 2 Heading date; at the time of Ist ear/m 2 >5O Vo Maturity date: determined Root sampling (at end harvest): 2—3x50 cm at raw (15 cm depth) Number of plants (at end harvest): 3—6 x 1 m at raw Number of ears/panicles (at end harvest): 3 —6 x 1 m at raw Straw yield DM (cutting 5 cm): 4 —6x20 m 2 (fall) Grain yield (barley 15 Vo and turnip rape 9 Vo moisture content): 4—6x20 m 2 (fall) 1000 seed weight 3 x 100 seeds roots penetrate to the clay soil and the quan- tity with which they remain in the field. In 1985 root density in the soil (cm/cm 3) was measured by Newman’s (1966) method for es- timating the total length of root sampling (Ta- ble 5.1). According to Madsen (1978), the ef- fective root depth comprises at least 0.1 cm root per cm 3 soil. In 1986 and 1987, root depth growth was measured from emergence to the time when a root depth of 60 cm was attained. Accord- ing to these results the average root penetra- tion speed was 1.3 cm per day for barley, 1.2 cm per day for turnip rape and 0.7 cm per day for timothy. Root depth growth was not the same for the whole growth period (Table 5.2). According to Jakoiisi n’s (1976) formula of root growth, with a threshold soil temperature of 4°C, the soil temperature of the root penetration zone did not restrict the root growth of barley or turnip rape during 1983 87. At the time of sowing the soil tempera- ture of the tillage layer was uniformly + 10°C or more. According to this formula, soil tem- perature restricted the root growth of timo- thy about 1 to 2 weeks after the onset of the growth period. The studies showed that the maximum ef- fective root depth (dr ) remained <75 cm for all threeplant species. Salonen (1949), study- Table 5.1. Root length, cm in cm* soil in 1985. Depth Date cm 28/5 11/6 8/7 Timothy 00—10 9.0 10—20 0.7 20—30 0.3 0.4 1.8 30—40 0.2 0.9 40—50 0.0 0.5 50—60 0.2 Burley 00—10 17 10—20 0.7 2.9 20—30 I -2 30—40 12 40—50 0.8 Turnip rape 10—20 1.6 20—30 0.9 30—40 0-5 Table 5.2. Root depth growth. Year Date (Days) Soil Depth depth growth cm cm/day Barley 1986 3/6—23/6 (21) 0—25 1.2 24/6—14/7 (21) 25—60 1.7 Avg. 3/6—14/7 (42) o—6o 1.4 1987 5/6—29/6 (25) 0—25 1.0 30/6—28/7 (29) 25—64 1.3 Avg. 5/6—28/7 (54) 0—64 1.2 Turnip rape 1986 3/6—23/6 (21) o—lo 0.5 24/6—21 /I (28) 10—60 1.8 Avg. 3/6—21/7 (49) o—6o 1.2 1987 8/6—29/6 (22) o—l 9 0.9 30/6—28/7 (29) 19—60 1.4 Avg. 8/6—28/7 (51) o—6o 1.2 Timothy 1986 25/4—19/5 (24) o—2o 0.8 20/5—16/6 (28) 20—35 0.5 17/6—14/7 (29) 35—60 0.9 Avg. 25/4—14/7 (81) o—6o 0.7 1987 23/4—lB/5 (25) 0— 6 0.2 19/5—16/6 (29) 6—51 1.6 17/6—20/7 (35) 51—61 0.3 Avg. 23/4—20/7 (89) o—6l 0.7 ing barley and timothy root growth in differ- ent soil types, showed a maximum root depth of 35—85 cm for barley and 40—70 cm for timothy in clay soils. In Denmark the aver- age effective root depth in clay soil has been 100 cm containing 170 mm water as a root zone capacity. In 1983—1985 the amount of main roots was only 400—500 kg for barley and 250—400 kg of dry matter (DM) per hectare for turnip rape. Careful washing of soil samples to a depth of 60 cm in 1986 introduced root DM yields of barley 1000—1500 kg and 500—550 kg per hectare for turnip rape. Timothy had a 2—4 ton root DM mass per hectare, but that sum also contained old, dead roots (Table 5.3). 5.3. Dry matter production Crop growth was measured weekly throughout the study period. Cuttings were measured weekly from a plant height of about 20 cm, the measurements continuing until the end harvest. Daily above ground (>5 cm) dry matter production per hectare for barley af- ter emergence was 50—90 kg of DM per day Table 5.3. Total amount of roots (kg DM/ha) in 1986. Plot Soil Depth (cm) 00—10 10—20 20—30 30—40 40—50 50—60 00—60 Barley Cl 991 265 158 44 20 43 1521 C 2 632 139 110 71 63 18 1033 Avg. 812 202 134 58 42 30 1277 Turnip rape A 1 280 102 95 56 22 555 A 2 305 54 82 56 28 525 Avg. 292 78 88 56 25 540 Timothy Bl 2478 170 110 37 22 2817 B 2 3377 195 120 32 24 9 3757 Avg. 2928 182 115 34 23 4 3287 643 and increased during the next four to six weeks to a maximum value of 200—300 kg of DM per day. At the very early phase of the de- velopment, turnip rape growth was very slow, but the pace of growth increased quickly af- ter the sprouting period to a level of 100—150 kg of DM per day. The maximum values were 150—250 kg of DM per day. The DM growth of timothy varied enourmously from one year to another and also between cuts. During spring, daily grass growth was 50—100 kg of DM, and later reached the maximum valueof 200—250 kg of DM per day. Summer grass growth varied between 30 and 180 kg of DM per day. The autumn growth of timothy ranged from 10to 100 kg ofDM per day. (see chapter 8.) 6. Results of the end harvestings 6.1. Barley and turnip rape Table 6.1 shows the results of dry matter production at harvests, divided into grain DM yield and total production above ground lev- el (> 5 cm). End harvests were done with an experimental harvester on four to six subplots, each 20 m 2 in size. Straw and grain were sepa- rated, but it was impossible to keep the cut- ting height constant at a stubble height level of 5 cm. When results of end harvests were compared to those of the last periodic cut- tings, the existing error was estimated to be about 10—15 % of the actual total dry mat- ter yield. Plant height and harvest analyses are presented in Tables 6.2 and 6.3. Harvest in- dex values of barley and turnip rape were cal- culated as the dry matter yield of grain (seed) at the end harvest per the maximum measured (> 5 cm) crop dry matter yield of the growth period. The shoot height growth of barley ended at the time when the maximum GAI values were reached. In the case of turnip rape, height development was related to the time of blooming. The best growing conditions for DM production of the studied crops occurred in 1983, which was warm and dry. In 1985 and 1986, irrigation had a positive effect on plant height development, but the yields were great- er only in 1986. 1987 was a cloudy, cold and rainy year and was the least favourable year for plant production since 1962. 6.2. Timothy Timothy plots were established in the spring of 1982 and of 1984. These years were not in- cluded in crop growth simulation(Table 6.4). Table 6.1. Growth period, dry matter production above ground level (>5 cm) and grain yield of barley and turnip rape at end harvest (kg DM/ha). Year Plot Dates of Non-irrigated Irrigated Sowing Emerging Ripening Harvest Grain Total Grain Total Barley 1982 B 27/5 5/6 23/8 7/9 4023 6525 3224 5896 1983 A 10/5 20/5 8/8 11/8 4408 8580 4167 8276 1984 B 15/5 21/5 10/8 15/8 3483 7031 3591 7571 1985 A 24/5 2/6 23/8 26/8 3891 6617 3283 6824 1986 C 27/5 3/6 23/8 28/8 3630 6122 4604 7920 1987 A 26/5 5/6 8/9 17/9 2994 6294 3012 6208 Turnip rape 1982 A 27/5 7/6 13/9 16/9 1702 3832 1815 4326 1983 B 10/5 20/5 18/8 25/8 1482 5388 1647 5636 1984 A 31/5 4/6 3/9 10/9 650 1735 877 2748 1985 C 24/5 3/6 9/9 12/9 1726 4833 1560 4603 1986 A 27/5 3/6 13/9 19/9 1323 3835 1566 5188 1987 C 26/5 6/6 23/9 2/10 1450 5330 1259 5403 644 Table 6.2, Harvest analyses of Porno barley in 1983—87. Year/Plot Weight Plant Harvest height index1000 Heclo- s u ... cm H seed g litre kg 1983 A 1 39.2 66.5 98 0.52 A 2 38.8 67.2 99 0.53 1984 B 1 33.4 62.1 101 0.39 B 2 33.3 60.4 102 0.39 1985 A 1 29.7 59.1 108 0.41 A 2 37.3 62.9 95 0.53 1986 Cl 38.2 60.3 89 0.55 C 2 41.3 62.7 71 0.50 1987 A 1 38.3 61.9 79 0.42 A 2 39.0 63.3 76 0.38 Mean 1983—87 36.8 62.7 92 0.46 Table 6.3. Harvest analyses of Span turnip rape (Kova 1987) in 1983—87. Year/Plot Weight Plant Harvest 1000 height index seed g cm H 1983 B 1 2.44 111 0.25 B 2 2.31 110 0.20 1984 A 1 2.61 85 0.26 A 2 2.46 90 0.18 1985 Cl 2.83 127 0.25 C 2 2.40 105 0.28 1986 A 1 2.73 100 0.25 A 2 2.60 78 0.28 1987 Cl 2.88 88 0.22 C 2 2.82 88 0.24 Mean 1983—87 2.61 98 0.24 The best growing season for grass production occurred in 1983, when the total DM yield above ground was about 11 tons of DM per hectare (Table 6.5). For timothy, the growth of plant height was equal to the development of GAI. The spring of 1985 was cold, and the yields of the first end cut were the lowest of any of the studied years. In 1985 and 1986, irrigation after the first end harvest hastened the de- velopment of crop growth. In 1987 grass de- velopment was slow and only two end cuttings were taken, but the DM yield was still about 9 tons per hectare. 1. The model 7.1. Simulation of the crop area Aslyng and Hansen (1982) used the long- term average CAI and GAI values of the stud- ied crops when developing their simulation model. For the crops surface development of used model, it is necessary to know the date of emergence (JDAYI), the date when the GAI values reach 5 (JDAY2), the date of the developmental point when the GAI decreases below 5 (JDAY3) and the dateof full maturi- ty, when the GAI =0 (JDAY4). In cereals the CAI is 0 before emergence; it then rises ex- ponentially to 5: (7.1) CAI =G m (exp (2.4 (t —JDAYI)/ (JDAY2 —JDAYI)) 1)/10 where t = time, JDAYs = Julian days from the beginning of the year and Gm = the maximum value of GAI. In the model of maximum value of GAI (G m) is stated as sto the onset of ripening. GAI values over 5 had a minor influence on the amount of energy absorbed by the crop (Aslyng and Hansen 1982). After the onset Table 6.4. Dry matter production (kg DM/ha) and plant height of timothy in the years of establishment. Year/Plot Dates of DM yield Plant height Sowing Emerging Harvest 1 2 I 2 1982 C 1984 B 27/5 14/5 7/6 24/5 6/9 1417 1898 59 45 5/9 1731 2591 45 40 KEY: Plot 1 irrigated, 2 =non-irrigated 645 Table 6.5. Dry matter production (kg DM/ha) and plant height (cm) of timothy in grass years. Year/Plot Dates of Plant height DM yield Onset Harvest I 2 1 2 1983 C 20/4 16/6 103 105 6654 6893 2/8 59 62 2817 2777 26 15489/9 36 778 Sum 11019 10448 1985 B 8/5 24/6 66 67 2231 2958 12/8 83 95 4058 2750 26/9 33 25 983 1240 Sum 7272 6948 1986 B 25/4 13/6 76 74 5118 4995 4/8 79 43 3211 1172 8/9 26 35 558 911 Sum 8887 7078 1987 B 30/4 25/6 76 75 4269 4340 10/9 96 86 4593 5067 Sum 8862 9407 KEY: Plot: l=irrigated, 2= non-irrigated of ripening, the GAI and CAI values decrease linearly to full ripening (GAI=O), and in cereals the CAI at harvest equals 2 (YAI = 2): (7.2) GAI =G m -G,„ (t - JDAY3)/ (JDAY4 —JDAY3) (7.3) CAI =G m —(G m —Y)(t JDAY3)/ (JDAY4 —JDAY3) where Y = the yellow area index (YAI). The same type of crop area model was ap- plied for turnip rape as for barley, fit to the development of rape. The turnip rape stand did not always reach a GAI value of 5. In 1984 the maximum value of 2.5 was used for rape, because leaf area development was poor. The simulation model of GAI for grass was developed for Italian ryegrass cut five times during the growing season (Aslyng and Han- sen 1982). We used the same type of GAI model for timothy in a three cut system. De- velopment of GAI to the value of 5 was the same as for barley; a GAI unit of 0.5 was ad- ded to describe the plant stand after cuttings: (7.4) CAI = Gn, (exp 2.4 (t JDAYI)/ (J DAY2 - J DAY 1))- 1)/10+ (0.5) For the timothy simulation model of GAI, the Gm-value of 5 remains stable until har- vest. Aslyng and Hansen (1982) used the com- mon development of GAI, which depends on the temperature sum. Our data of GAI in all plant species showed that GAI development varied yearly and was not entirely dependent on the effective temperature sum (ETS,> 5°C). In our study it was more reliable to simulate the real yearly development of GAI for all three species studied (Fig. 7.1—7.3). 7.2. Actual evapotranspiration Daily evapotranspiration was calculated ac- cording to the EVAPO simulation model (Aslyng and Hansen 1982). The input pa- rameters of this submodel are precipitation (P), irrigation (I) and potential evapotranspi- ration (E*). The daily values of maximum ef- fective root depth (d r ) and crop area index (CAI) are also needed. The soil water capaci- ty is divided into two parts. The topsoil reser- voir (S*) is considered to occupy the upper 10 cm layer of the soil, and is stated contain- 646 2 ing 10 mm water. Heinonen (1985) proposed that this reservoir is a “microrelief of the sur- face”, which delays the beginning of the flow of the surface water. S,* is not independent of the root zone reservoir, which is considered to be a function of soil type and the effective root depth. St* can be readily evaporated from the soil surface, and root zone reservoir is available to the plants in the root zone. The priori assumption is that the actual evapotranspiration (E) can reach, but cannot exceed, the potential evapotranspiration. In the EVAPO model a break point (E/E* = 1.0) is adopted, when 50 % of the water is used by the plants. The pressure head at which soil water begins to limit plant growth seems to range between a pF-value 2.6 and 3 (Feddes et al. 1978). Denmead and Shaw (1962) Fig. 7.1. Simulated and measured green area index (GAI) of barley in 1982—87 647 reported the influence of particular meteoro- logical conditions on the relationship between actual and potential transpiration. They de- termined the point of soil moisture at which the wilting of the plants increased at the same rate as the increase in potential transpiration. Long and French (1967) showed that loss of soil moisture by evaporation occurred main- ly from the upper 30 cm of soil, in conditions when the soil contained less water than it does at field capacity. Drying below this depth is caused by the extraction of water by the roots. In the WATCROS model, the break point of the soil moisture function value of 0.5 (50 %) was used from May to August, and a value of 0.6 (60 %) was used for april and September for all plants. In the potential case of the simulation, irrigation was applied, Fig. 7.2. Simulated and measured green area index (GAI) of turnip rape in 1982—87 648 when the 40 % fraction of the root zone ca- pasity was utilized. The upper limit of water applied in an irrigation was 50 mm. According to Saucier (1970), evaporation behaves in the same way as does the net radi- ation. When the evaporative demand is dis- tributed between the soil (Ef) and the crop (E*) and using BEER’s law (see equation 7.18.), the following equations can be drawn: (7.5) Ef = E*-KG (7.6) E* =E* (1 -e~KC ) (7.7) E* g =E* (1 -e~KG ) (7.8) E* y=E* E*. where K = the extinction coefficient (equal to a net radiation of 0.6), G = the green area in- dex (GAI), C= the crop area index (CAI), E* g = the evaporative demand of the green active crop area and E* y= the evaporative demand of the yellow, inactive, crop area. The model operates on a daily basis; at the beginning of each step the amounts of precipi- tation and irrigation, called the potential in- terception storage (S*), are supplied to the reservoir. Jensen (1979) proposed that the plant can, at most, hold water equal to 0.5 mm H2O on the crop surface (C) and on the crop green area (G). The interception storage can further be distributed as follows: (7.9) sr = 0.5C (7.10) 5,% = 0.5 G (7.11) s* y =sr —S* g where S* g = the potential interception storage of the green crop area and S,* y = the intercep- tion storage of the yellow crop area. The rest of P and I are supplied to the top- Fig. 7.3. Simulated and measured green area index (GAI) of timothy in 1983, 1985—87 649 soil and root zone reservoir. Extra water is transferred to a through-flow reservoir, where it remains for three days if not evapotranspi- rated during that period. After this, water is drained out of the root zone as a deep perco- lation. At the beginning, water is extracted from the topsoil at a potential rate as long as there is water in the reservoir; next it is extracted from the root zone, at a rate equal to 0.15 E*. E s = Ef ;S,>Ef (7.12) E, = 0.15 E*; S, < Es * where Es = the actual evaporation of the soil and S, = the actual topsoil water storage. E* g extract water from the green crop area and E* y from the yellow crop area in the potential rate equal to the actual values: F = S • S F**-'c, y cc, y » y Ec, y F =F* • S >F*Cc, g Cc,g » °l.g- CC,g (7.14) Ec , g~S|, g + E T ; S, go.s S* (7.16) ET =E|—; 0<5,<0.5 S* 0.5 S* Et = 0 ; Sr Rm (7.27) Rg = O P < R1 g lx m where rg = the growth respiration coefficient. When the given formulas are combined, the amount of recently assimilated carbohydrates (P n |) not yet converted into structural plant dry matter can be calculated: (7.30) il ~)Q\ D _ i I'm X W )[/.Zö) P . = !—— •P > p nin1 + (0.5 X0.7 Xr ) ’ 8 m Umsi) P _ r y W (7.29) Pn ;= —25 —; P B R m1 + (0.5 x rm) 8 m (7.32) the total amount of accumulated plant dry matter cannot be harvested; some of it re- mains in the field. As thebasis of their calcu- lations, Aslyng and Hansen (1982) used 3 ton DM losses in the field; the same value was used in Norway (Kvifte 1987). This study also used 3 ton losses for our plant species, because of difficulties in determinations of the actual root yields in the heavy clay soil. (7.33) Simulation of the daily non-harvested DM yield from the total DM production should take into consideration the formula of plant dry matter partitioning between roots and shoots (Heemst 1986, Keulen and Seligman 1987). The daily stubble, root etc. mass loss (WL ; ) is calculated by using the simulated daily potential DM production (P„ ,) until the maximum root depth (d r) or the full 3 ton DM loss is reached: WL ,i = 0.5 P„, WL i = 0.659-0.01 P n j (barley) WL j= 0.644 —O.Ol P n | (t. rape) (t>t3 )WL> j= 3.0 where P n ( = the daily potential net produc- tion, including the loss, t 0 = the beginning of simulated potential DM production, t, = 2 weeks from tO, t 2 = the time, when 3 ton DM loss is reached and t 3 = the time of maximum root depth (Fig. 7.4). After calculations of these formulas (7.30— 7.33), the total loss of 3 tons per hectare was reached, on average, 47 days from sowing for barley and 61 days for turnip rape. In Great Britain the maximum root yield of cereals were obtained at about 50 days from sowing (Welbank et al. (1973). In the Danish and in the our version, DM Fig. 7.4. Estimation of daily non-harvested dry matter (stubble, root, etc. mass loss) for barley and turnip rape (explanation see text). 652 loss of timothy was calculated using a linear obtained through optimal irrigation treat- function from the onset of growth to the max- imum root depth: 3 0(7.34) W, j= (timothy). dr Where dr = the maximum effective root depth (75 cm). The model continues to operate with W h until the total lost dry matter yield (3 tons) is reduced: (7.35) W h = P n -WL where W h= the harvested dry matter yield. Grain (barley) and seed (turnip rape) yields (W g ) were calculated as follows: (7.36) W g = H x Wh where H = the harvest index. The harvest index values applied were H=0.45 for barley and H = 0.25 for turnip rape. Timothy yield is considered to have a harvest index of H= 1.0. 7.5. Water limited crop production Aslyng and Hansen (1982) used a linear relation between transpiration and potential crop production in calculating the water limit- ed (actual) plant gross production (P g): (7.37) Pg P* g where E = transpiration plus the evapora- tion of water intercepted on the green active crop surface, E* g = the same in the potential case and P* = the potential gross production. The WATCROS model does not consider the effect of water stress on the crop green area. Aslyng and Hansen (1982) prefer gross production to net production because of difficulties and errors in estimating respira- tion. Determination of actual production differs from that of potential production in only one essential aspect. Potential crop production is ment, but the actual crop production is entire- ly dependent on water, in the soil, available to the plant and on root depth growth. Aslyng and Hansen (1982) and Kvifte (1987) used the same maximum efficient root depth of 100 cm. According to root measurements made in 1986 and 1987, the maximum root depth in the Finnish studies was assumed to be 75 cm. 8. Results and discussion 8.1. Barley and turnip rape In 1983—87 the time of the GAI develop- ment of barley was 29(27 —31) days for the increasing phase. It was 23 (17 —34) days at GAI>5, and 33 (19 —45) days in the decreas- ing phase. The simulation for the increasing phase of GAI, using an exponential function, succeeded well. The simulated DM production for barley was 30—45 % in the increasing phase of GAI, 35—55 °7o at GAI>S and 15—30 % in the decreasing phase of GAI. In the prevailing weather conditions the simulated actual (water limited) DM produc- tion of barley in the experimental field was the highest in 1983 and 1984, whereas the actual DM production of turnip rape reached the maximum level in 1982 and 1983. Simulated total potential DM yields containing root, stubble etc. mass loss were 13.2—15.5 tons per hectare for barley and 12.1—15.3 tons per hectare for turnip rape exept in 1984, when it was only 7.0 tons DM per hectare. The simulated development of actual dry matter production for barley was somewhat greater than the values measured (Fig. 8.1). The difference was even greater for turnip rape than for barley (Fig. 8.2). Such behaviour is explained by the fact that the WATCROS model takes into consideration only the limits in water use, but not the excess of water in the soil. In Finnish climatic and soil condi- tions, root development seems to be more re- stricted than in the Danish conditions. Soil 653 conditions and root development seem to lead the plant to a lower PAR use of stand in Fin- land than in Denmark. In some years the grain yields at Jokioinen district were higher than the simulated poten- tial grain yields in the experimental field (Ta- bles 8.1 and 8.2). In such cases harvested yields that were greater than simulated poten- tial ones may be the result of differences in sowing time, the development of GAI in re- lation to incoming radiation and the values of the harvest index. For example, in the first study year (1982), the crop surface develop- ment of barley was poor, which decreased the simulated yield. In 1987, though generally a rainy and cool year, the development of GAI Fig. 8.1. Above ground dry matter (DM) production of barley in 1982—87. E =emergence day 654 was great due to the high incoming radiation in July, and the simulation of potential DM production of barley was also high, 15.0 tons per hectare. The measured DM production re- mained low due to the excess of water and low temperature during the filling period of grain. According to the results of the simulated potential yield, irrigation was meaningful dur- ing the last three tests years (1985—87) for barley but only in 1983 and 1986 for turnip rape. The simulated model introduced require- ments for irrigation; these varied from 30 to 165 mm in the growing season for barley and 15—170 mm for turnip rape. In the rainy sea- sons of 1984and 1987, occasionally there was too much water in the field for crop produc- Fig. 8.2. Above ground dry matter (DM) production of turnip rape in 1982—87. E =emergence day 655 Table 8.1. Production results of Porno barley including 15 % moisture at Jokioinen (grain tons/ha). Year Measured: Simulated: KVO KJO Climatic field Potential Actual I 2 MAK PEN MAK PEN 1982 7.5 7.3 3.84.7 5.55.5 5.55.4 1983 4.6 7.3 4.95.2 6.26.2 6.06.1 1984 3.3 4.2 4.24.1 6.36.3 6.36.3 1985 5.6 6.4 3.94.6 5.45.4 4.65.3 1986 4.0 5.8 5.44.3 6.66.6 5.65.8 1987 4.2 4.6 3.23.2 6.46.4 5.66.2 KEY; Department of Agricultural Centre: KVO =Crop Science, KJO =Plant Breeding Formula of Potential evapotranspiration (PET): MAK =Makkink, PEN = Penman Plot; 1 =Irrigated, 2 = Non-irrigated Table 8.2. Production results of Span turnip rape (Kova 1987) including 9 % moisture at Jokioinen (seed tons/ha) Year Measured: Simulated: KVO KJO Climatic field Potential Actual 1 2 MAK PEN MAK PEN 1982 2.7 1.9 2.1 1.9 3.2 3.2 3.1 3.0 1983 2.8 2.6 1.8 1.6 3.4 3.4 3.1 3.3 1984 1.9 0.7 1.0 0.7 1.6 1.5 1.6 1.6 1985 2.0 1,7 1.9 2.5 2.5 2.4 2.5 1986 1.8 1.7 1.5 3.2 3.2 2.7 2.9 1987 2.0 2.4 1.4 1.6 2.6 2.6 2.6 2.6 KEY: (see Table 8.1) lion, a factor which was not taken into con- sideration in this growing model. 8.2. Timothy For timothy, the real annual development of crop surface was simulated separately for each of the cuts. Aslyng and Hansen (1982) used the long term mean development of GAI for Italian ryegrass. The maximum values measured for GAI were often much greater than the value, Gm =5, used in the simulation model. In Norway Kvifte (1987) used the value of Gm = 7 for the first cut and Gra = 5 for the second cut in the simulation of GAI of timothy. The total DM yields measured for timothy were only about 60 % of the simulated ones (Table 8.3). The simulation of timothy DM production succeeded best in 1987, but was still unsatisfactory (Fig. 8.3). Timothy had the best production conditions in 1983. The simu- lation model introduced the need for irriga- tion for the whole growth period; it varied from 55 to 150 mm per year. In the field, ir- rigated plots had better yields than the non- irrigated plots only in 1985 and 1986, which were relatively dry years. The total annual simulated poatential DM yields of timothy were 15.4—20.2 tons per ha. One reason for the differences between the simulated grass dry matter yields and the ac- tual measured DM yields may have been the effect of soil temperature, which was a 656 Table 8.3. Production results of Timothy grass at Jokioinen (tons DM/ha). Year End Measured: Simulated: KVO Climatic field Potential Actual 1 2 MAK PEN MAK PEN 1983 I 5.8 6.6 6.9 7.5 7.5 7.5 7.5 II 5.2 2.8 2.8 6.0 6.0 5.9 5.9 111 1.4 1.5 0.8 3.7 3.7 3.5 3.6 Sum 12.4 11.0 10.4 17.2 17.2 16.9 16.9 1985 I 3.9 2.2 3.0 6.1 6.1 5.4 5.4 II 3.7 4.1 2.8 4.6 4.6 3.9 4.0 111 l.O 1.2 2.1 2.1 2.1 2.1 Sum 7.6 7.3 6.9 12.8 12.8 11.4 11.5 1986 1 4.7 5.1 5.0 7.0 7.0 6.3 6.4 II 2.3 3.2 1.2 6.1 6.1 4.5 4.7 111 0.6 0.9 2.5 2.5 2.5 2.5 Sum 7.0 8.9 7.1 15.6 15.7 13.3 13.6 1987 I 3.5 4.3 4.3 6.4 6.4 5.9 6.3 II 4.4 4.6 5.1 6.0 6.0 6.0 6.0 Sum 7.9 8.9 9.4 12.4 12.4 11.9 12.3 KEY: (see Table 8.1) 657 Fig. 8.3. Above ground dry matter (DM) production of timothy in 1983, 1985—87 growth-reducing factor in one to two weeks after the onset of the growth period. Other in- fluencing factors are the same as those noted for barley and turnip rape. However, the simulated second and third yields were much too high compared to the first yield of timo- thy. The reason for this may be the natural growth rhythm of timothy, which includes a very slow start of regrowth after cuts. 9. Conclusions The Water Balance and Crop Production Simulation (WATCROS) model of Danish origin was tested in Finnish climatic and soil conditions as a part of the Nordic Project (NKJ-47). Different from the WATCROS model, the simulated crop surface was determined as the real development of GAI, owing to the impor- tance of GAI in the absorbance of PAR. In the simulation model of potential evapotranspiration, modified versions of Makkink (1957) and Penman (1956) were tested. As a result, the calculated values of potential evapotranspiration by Makkink or Penman led to the same result of the simu- lated actual DM yields of the three studied plant species. The Makkink was used as the basis of calculations. The constants used in the WATCROS mod- el are as follows: gross C02 single leaf as- similation=0.83 mg m^ 2 s-1; albedo = 6 % of PAR; the factor converting stored energy to plant structural DM = 70 g DM M.l the max. GAI = 5; the harvest index, =0.45 for barley, =0.25 for turnip rape and = 1.0 for timothy; the extinction coefficient = 0.8 for timothy and barley and = 0.65 for turnip rape; the extinction coefficient for net radiation = 0.6; the growth respiration coefficient = 30 %; the maintenance respiration coefficient = 1.5 % for barley and turnip rape and = 4.0 % for timothy; the maintenance respi- ration Q lO = 2; PAR = 48 % of global radia- tion; gross photosynthetic efficiency = 8 %; the stubble, root etc. mass loss in harvest = 3 ton DM per hectare; the maximum effective root depth =75 cm; the speed of root depth growth = 1.3 cm per day for barley, = 1.2 cm per day for turnip rape and = 0.7 cm per day for timothy; the point of soil moisture func- tion=0.5 for May—August, =0.6 for April and September; the capasity of topsoil evapo- ration reservoir = 10 mm; the fraction of the root zone capasity utilized =0.40, when irri- gation is applied; the largest amount of water applied in an irrigation = 50 mm. In the WATCROS model the simulated water limited crop production fit well to the actual measured crop production in Denmark (Aslyng and Flansen 1982) and, in modified form, also in Norway (Kvifte 1987). In Fin- land the Danish version of the model in- troduced higher simulated actual production than measured in actual yields. Reasons why the Danish model, unless modified, does not seem to fit to Finnish cli- matic and soil conditions are as follows: The Danish model consentrates on the water limited production conditions, excluding the excess of water in the soil and in the plant. The maximum efficient root depth growth in Finland (75 cm) is more limited than in Den- mark (100 cm). Also the daily depth growth of the roots in Finland was lower (from 0.7 to 1.3 cm) than in Denmark (1.5—2.0 cm per day). The gross photosynthetic efficiency 8 % from PAR, used in Danish model seems to be an overestimate in the Finnish conditions, owing to soil type and excess of water in the soil and in theplant, but also to the tempera- ture conditions in shoot growth, and especial- ly, in root growth. The Finnish studies on simulation models occurred during growing seasons character- ized by heavy amounts of precipitation. The WATCROS model did not include possible losses of nitrogen in the calculations. Water, PAR efficiency and possibly nitrogen should be taken into consideration when construct- ing a production model for Finnish climatic and soil conditions. Acknowledgements. This paper is connected with a joint Nordic project, NKJ-47, ‘Effect ofclimatological 658 factors on crop growth and production in Nordic Coun- tries’. We wish to thank computer operator R. Merkkinie- mi and the staff of the Department of Crop Science of the Agricultural Research Centre and the staff of the Meteorological Observatory at Jokioinen and computer operator, T. Koskela, M. Sci. of the Finnish Meteoro- logical Institute in Helsinki for their important assistance. We are also grateful to Dr. T. Karvonen of the Finnish Field Drainage Centre in Helsinki for inspiring discus- sions. We thank the Academy of Finland and the Minis- try of Agriculture and Forestry for financing the study. References Ansalehto, A., Elomaa, E., Nordlund, A. & Pilli- Sihvola, Y. 1985. Maatalouden sääpalvelukokeilu kesällä 1984. MTTK, Tiedote 2/85. 127 s. Jokioinen, Aslyng, H.C. 1976. Kiima, Jord og Planter. Kultur- teknik I, 5. Den. kgl Veter.- og Landbohosk. 368 p. Kopenhavn. Aslyng, H.C. & Hansen, S. 1982. Water balance and Crop production simulation. Hydrotechnical Labora- tory. The Royal Veterinary and Agricultural Univer- sity. 200 p. Copenhagen. Biscoe, P.V. & Gallagher, J.N. 1977. Weather, dry matter production and yield. (Eds) Landsberg and Cut- ting. Environmental Effects on Crop Physiology, p. 75—100. Academic Press. New York. Denmead, O.T. & Shaw, R.H. 1962. Availability of soil water to plants as affects by soil moisture content and meteorological conditions. Agron. J. 54: 358—390. Elomaa, E. 1987. Experiences in automation of agrometeorological observations in Finland. Sixth Sym- posium on Meteorological Observations and Instrumen- tation of the Amer. Meteorol. Soc., January 12—16, 1987. 4 p. New Orleans, Louisiana. & Pui 11, S. 1985. Variationer i globalsträlning, effec- tiva lemperatursumma, nederbörd, potentiella evapotranspiration och nederbördsunderskott i relation till växtproduktion i Södra Finland. NJF-seminarium 77: 19—27. Uppsala 24—25 September 1985. Jord- bruksmeteorologi. Aktuell och potentiell växtproduk- tion. Nordiska Jordbruksforskarens Forening. Uppsa- la. —, Ilola, A. & Pulli, S. 1986. Final report of project NKJ-47 in Finland. 11 (+3) p. Agricultural Research Centre. Jokioinen. Feddes, R.A., Kowalik, P.J. & Zaradny, FI. 1978. Simulation of field water use and crop yield. 189 p. Centre for Agricultural Publishing and Documentation. Wageningen. Gallagher, J.N. 1976. The Growth of Cereals in Rela- tion to Weather. Ph. D. Thesis. University ofNottin- gham. 158 p. Nottingham. Goudriaan, J. 1982. Potential production process. (Eds) Penning de Vries and van Laar. Simulation of plant growth and crop production, p. 98—113. Centre for Agricultural Publishing and Documentation. Wagenin- gen. Hankimo, J. 1964. Some computations of the energy ex- change between the sea and the atmospheric in the baltic area. Finn. Meteorol. Office Contr. 57: 1—26. Hansen, S., Jensen, S.E. & Aslyng, H.C. 1981. Jord- brugsmeteorologiske observationer, statistisk analyse og vundering 1955—1979. Hydroteknisk Laborato- rium. Den kgl. Veter. og Landbohosk. 414 p. Kobenhavn. Heemst, H.D.J. 1986. Crop phenology and dry matter distribution. (Eds) Keulen and Wolf. Modelling of agricultural production: weather, soils and crops, p. 27—40. Centre for Agricultural Publishing and Documentation. Wageningen. Heinonen, R. 1985. Soil Management and Crop Water Supply. 4th ed. Dept. Soil Sci. Sweden University Agric. Sci. 103 p. Uppsala. Jakobsen, B.F. 1976, Jord, rodvaekst og stofoptagelse. (Eds) Hansen, Jakobsen og Jensen. Simuleret plan- teproduktion. Den kgl, Veter.- og Landbohosk. 34 s. Kobenhavn. Jensen, S.E. 1979. Model ETFOREST for calculating ac- tual evapotranspiration. (Ed.) Halldin. Comparison of forest water use and energy exchange models. Int. So- ciety for Ecological Modelling (ISEM) p. 165—172. Copenhagen. Keulen, H. Van & Seligman, N.G. 1987. Simulation of water use, nitrogen nutrition and growth of a spring wheat crop. 310 p. Centre for Agricultural Publishing and Documentation. Wageningen. Kulmala, A. 1970. Heat balance of the earth’s surface at Jokioinen (60.8 N, 23.5 E), Summer 1968. 69 p. Finn. Meteorol. Office Contr. 74. Helsinki, Kvifte, G. 1987. Crop Production and Growth Model for Cereals, Rape and Grass at Aas, Norway. Acta Agric. Scand. 37; 137—158. Long, I.F. & French, B.K. 1967. Measurement of soil moisture in the field by neuthron moderation. J. Soil. Sci. 18; 149—166. Madsen, H.B. 1978. Jordbundskartering og bonitering. Belyst ved hjaslp af jordens vandretention, bygs rodud- vikling og simuleret planteproduktion. Folio Ge- ographica Danica X, 5. Licentiatafhandling, Koben- havns Universitet. 183 s.+ 6 bilag. Kobenhavn. Makkink, G.F. 1957. Ekzameno de la formulo de Pen- man. Repr. Netti. J. agric. Sei. 5: 290—305. 659 Morton, F.l. 1975. Estimating evaporation and transpi- ration from climatological observations. J. Appi. Meteorol. 14,4: 488—497. Newman, E.l. 1966. A method of estimating the total length of root in a sample. J. appi. Ecol. 3; 139—145. Penman, H.L. 1956. Evapotranspiration: An introduc- tory survey. Neth. J. agric. Sei. 4: 8—29. Penning De Vries, F.W.T. 1980. System analysis and models of crop production. (Eds) Penning de Vries and van Laar. Simulation of plant growth and crop produc- tion. p. 9—19. Centre for Agricultural Publishing and Documentation. Wageningen. Robson, M.J. 1981. Respiratory efflux in relation to tem- perature of simulated swards of perennial ryegrass with contrasting soluble carbohydrate contents. Ann. Bot. 48: 269—273. Saarinen, J., Pulli, S. & Elomaa, E. 1986. Sääkentän käyttö kasvin potentiaalisen sadon määrittämisessä. Suomen Maatal.tiet. Seuran Tied. No 7: 90—97. Saavalainen, J. & Rintanen, S. 1986. Uusi kenttämit- tausmenetelraä kyllästyneen maan hydraulisen johtavuuden mittaukseen. Vesitalous 3; 31—34. Salonen, M. 1949. Tutkimuksia viljelykasvien juurten sijainnista Suomen maalajeissa. Suomen Maatal.tiet. Seuran Julk. 70, 1: I—9l.1 —91. Saucier, B. 1970. Micrometeorology on Crops and Grasslands. (Eds) Landsberg and Cutting. Environmen- tal Effect on Crop Physiology, p. 39—55. Academic Press. London. Welbank, P.J., Gibb, M.J., Taylor, P.J. & Williams, E.D. 1973. Root growth of cereal crops. Roth. Exp. Stat. Report for 1973 part 2: 26—66. Ms received January 29;, 1988, SELOSTUS Tanskalaisen kasvumallin testaaminen ohralla, rypsillä ja timoteilla Suomen olosuhteissa A. Ilola, 1 E. Elomaa2 and S. Pulli 3 ' Kasvinviljelyosasto, Maatalouden tutkimuskeskus 31600 Jokioinen 2 Havainlolekninen toimisto, Ilmatieteen laitos PL 503 00101 Helsinki 1 Kasvinjaloslusosasto, Maatalouden tutkimuskeskus 31600 Jokioinen Biologis-meteorologinen aineisto kerättiin Jokioisissa vuosina 1982—87. Ohran, rypsin ja timotein potentiaa- linen ja vesirajoitteinen kuiva-ainesato simuloitiin tans- kalaisen WATCROS-mallin mukaan. Tärkeimpiä biologisia mittauksia olivat kasvustoalan (GAI), kuiva-ainesadon, juurtenkasvun ja maan kosteu- den viikoittainen seuranta sekä puintiajankohdan salo- analyysit sekä sadetetuilta että sadettamattomilta lohkoil- ta. Simuloinnissa tarvitut ja myös kentältä mitatut me- teorologiset parametrit olivat puolestaan päivittäinen auringon kokonaissäteily, ilman lämpötila ja sadanta. WATCROS-mallilla simuloidut kuiva-ainesadot olivat yleensä suurempia kuin koekentältä saadut sadot. Jatko- tutkimuksissa tulisikin selvittää fotosynteettisesti aktiivisen auringon säteilyn tehokkuus Suomen kasvuoloissa, sekä maan liiallisen märkyyden jakasveille käyttökelpoisen ty- pen huuhtoutumisen vaikutus kasvien kasvuun ja tuotan- toon. 660