24

Hemp: a ground water protecting crop?
Yields and nitrogen dynamics in plant and soil

Katja Hendrischke¹, Thomas Lickfett¹, and Hans-Bernhard von Buttlar^2,3

¹Georg-August-University, Institute for Agricultural Chemistry, Von-Siebold-Str. 6, D-37075 Goettingen, Germany
²Ingenieurgemeinschaft Landwirtschaft und Umwelt (IGLU), Buehlstr. 10, D-37073 Goettingen, Germany
³University of Kassel, Institute of Crop Science, Steinstr. 19, D-37213 Witzenhausen, Germany

        Hendrischke, Katja, Thomas Lickfett and Hans-Bernhard von Buttlar 1998. Hemp: a ground water protecting crop? Yields and nitrogen dynamics in plant and soil. Journal of the International Hemp Association 5(1): 24-28. The aim of this work was to examine the yields and nitrogen (N) dynamics of plant and soil in fiber hemp production, as affected by N fertilization and with special regard to the suitability of hemp cultivation in water catchments. Field experiments were laid out in 1996, cropping fiber hemp at three different levels of nitrogen fertilization and followed by field retting. Plant height, plant density, dry matter production, and nitrogen content in above-ground plant material and crop residues were determined. Additionally, soil N_min (0 - 90 cm) under the growing crop and after the harvest was recorded.
        Results indicate that both the soil parameters (soil fertility) and the seeding technique influence crop development and yield. Under conditions of high amounts of soil N min at the beginning of the experiment, applied nitrogen fertilizer showed no significant effect on dry matter yield. At harvest, almost no N_min was left in the soil. Nitrogen net mineralization of soil organic matter and plant residues increased after mowing and during the retting period, until December. This led to enhanced N_min status of soil after the vegetation period. This nitrate can be liable to leaching into ground water with the precipitation that occurs during winter and the following spring season. Hence, growing fiber hemp with field retting in water catchments might cause problems concerning nitrate content of water wells.

Introduction
        Hemp is regarded as a plant providing bioresources (oil and fiber) of high quality with environmentally compatible cultivation properties (e.g. renunciation of pesticides). In Germany, hemp cropping was excluded from the discussion on sustainable production of non-food-crops due to the prohibition of hemp cultivation which has existed since 1982. After repealing this prohibition in Spring 1996, a considerable deficiency of knowledge about hemp cultivation in Germany under contemporary conditions soon became evident. From the agricultural point of view, both the effect of nitrogen fertilizers on crop yield and the ecological consequences of its use in hemp cultivation is of particular interest. The use of nitrogen on plants and the dynamic and mobility of nitrogen in the soil are of special importance in regards to environmentally safe hemp production in order to avoid ground water pollution with nitrate. For this reason, in 1996, field experiments on hemp cultivation in Lower Saxony were carried out to answer the following questions:

• How does nitrogen fertilizer affect dry matter yield of fiber hemp?
• Which course does the nitrogen uptake of the crop follow?
• How does the soil N_min change during the vegetation period and after harvest?
• What quantity of crop residues and harvesting losses can be expected and how do they affect subsequent delivery of nitrogen to the soil by mineralization?

Materials and methods
         From April to December 1996, field experiments with fiber hemp (no seed harvesting) and a subsequent field retting were performed at four locations of southern Lower Saxony, Germany (Foehrste I + II, Elbingerode and Hoerden). The experimental fields in Foehrste were cultivated under a minimum tillage technique. The French monoecious variety Fedora 19 was sown in May with a seeding rate of 49-56 kg/ha to achieve a plant density of 300 plants m² . The experimental set-up was a large scale plot consisting of three nitrogen fertilization levels (0, 40, 80 kg N/ha, respectively, pre-emergence dressing) with one replication. During the vegetation period, the stage of development, plant height, plant density, fresh and dry weight of the plants and nitrogen content in above-ground dry matter were recorded six times. The area harvested was two m² each. The stage of development was described with EC numbers following a proposal of Von Buttlar et al. (1997). After mowing in September, the plants were left in the field for a six week dew retting period. Soil sampling for measuring N_min (CaCl₂ soluble nitrate and ammonia) took place every two weeks for the duration of the entire field experiment. N_t in oven-dried plant material was determined by thermal conductivity following dry combustion (DUMAS). In addition to the field experiment, the six week field retting period was simulated in an open greenhouse to quantify and qualify the alteration of the harvested crops.

Figure 1. Development and plant height of hemp crops at three levels of N fertilization.

Results and discussion
Development of hemp crops-
        At the experimental site in Foehrste I, hemp plants were already suppressed by weeds six weeks after seeding. Problems in seed bed preparation are considered to be the main reason for the stunted hemp; direct drilling in the mulch of the pre-crop set-aside seemed to restrict optimum growth of hemp plants shortly after emergence. For these reasons, the experiment at this site was canceled.
        At the experimental sites in Elbingerode, Hoerden and Foehrste II, crop development was almost uniform. Plants in Elbingerode were slightly ahead in early growth. The main elongation phase (EC 29) began after six weeks (Figure 1). Within 45 days after emergence, the hemp plants grew to 1.5 m, which corresponds to a daily growth rate of 3.3 cm. After 88 days of vegetative phase, the hemp crops reached the generative phase (EC 51). Grain ripening (EC 75) occurred 118 days after sowing.
        Nitrogen fertilization did not significantly affect plant height throughout the entire vegetation period (Figure 1). Results of the field experiment confirm that plant height is essentially influenced by location ("soil fertility") and duration of the vegetative phase (Höppner and Menge-Hartmann 1994). The plot mean at Elbingerode showed significantly taller plants at final harvest (226 cm) than the other sites, with a plot mean of 212 cm. The continuing increase in plant height after flower formation can be explained by a lengthening of the inflorescences (Bócsa and Karus 1997).

Dry matter yield-
        Averaging the three levels of nitrogen fertilization, above-ground dry matter yields were 13.5 t/ha in Elbingerode, 10.9 t/ha in Hoerden and 10.2 t/ha in Foehrste, respectively. The differentiation is assumed to be due to soil fertility and duration of the vegetative phase, which was one week shorter in Foehrste due to a later seeding date. There was no significant effect of nitrogen fertilization on dry matter yield at final harvest (plot mean). High amounts of soil Nmin at the beginning of the experiment (150 kg N/ha at 90 cm) may be an explanation for the lack of a distinct improvement of crop yield caused by an enhanced nitrogen dressing, as described by van der Werf (1994). At final harvest before the retting period, mean dry matter yield was 11.5 t/ha (three sites) which is equivalent to approximately 40 t/ha of fresh matter with a dry matter content of 29%. French breeders of the variety ‘Fedora 19’ indicated yields between 11.2 and 11.9 t/ha (FNPC 1994).

Figure 2. Nitrogen uptake of hemp (above ground plant material) during vegetation period at three levels of N fertilization.

Nitrogen uptake of hemp-
        During the main elongation phase, nitrogen absorption of the plants increased continuously. With the onset of flowering (EC 55) the increase of nitrogen absorption stagnated (Figure 2).
        Averaged over all nitrogen fertilizing levels at final harvest, plants at Hoerden und Foehrste had absorbed 123 kg N/ha and 159 kg N/ha at Elbingerode. The results show a clear correlation between nitrogen absorption and level of output. During the entire vegetation period higher N supply was followed by an enhanced nitrogen uptake.

Figure 3. Soil N_min in 0-90 cm depth during the growing period and after the harvest of hemp at three fertilization levels.

Soil N_min dynamics during vegetation period of hemp and after harvest-
        At the end of April 1996, soil N min in 0 - 90 cm depth was approx. 150 kg N at all three sites. Three weeks after seeding and application of nitrogen fertilizer, the different nitrogen fertilization levels became clearly visible as measured by soil N min (Figure 3).
        Shortly after emergence, fluctuations in the amounts of soil N min occurred. Nitrogen absorption of the growing crop and increased soil nitrogen mineralization can be held responsible for these changes (Scheller 1993). In the middle of July, a slight increase in soil nitrate appeared. At this time, subsequent delivery of soil nitrogen was greater than the crop's nitrogen uptake. Nitrogen was absorbed by the plants during the entire vegetation period, even after finishing the vegetative phase. By final harvest, hemp had been able to empty soil completely of N_min.
        Already during the field retting period, there was a slight increase of N_min in the top layer (0-30 cm) of all plots. This seemed to be due to mineralization of leached nitrogen compounds from the retting hemp plants and residues and might also be caused by enhanced activity of microorganisms due to the humid micro-climate in the deeply closed hemp canopy, and under the swath of retting hemp straw (Hesch 1995, Kainer 1996). The plant material lost during the retting period showed a relatively low C/N-ratio of 15:1 (plot mean) and is therefore rapidly mineralizable. These results support the assumption that an increased mineralization of plant residues occurs already in Autumn (Hanf 1996). Though there was no statistically significant effect of nitrogen fertilization on N_min after harvest, the plots with higher nitrogen dressing showed the tendency of greater soil N_min quantities. An obvious increase of soil N_min was caused by tillage for the following wheat crops. In November, soil N_min in the lower layers increased, indicating a beginning of vertical movement for nitrate nitrogen.

Alteration of plant material during the retting period-
        During retting, the plant material composition changed. At the beginning of the simulated retting experiment in the greenhouse, the dry weight of harvested plant material was 1.22 kg/m 2 averaged over the three nitrogen fertilization regimens (Table 1). Its nitrogen content was 1.35% and therefore this plant mass contained 16.6 g N/m². At the end of the retting period, the hemp straw had a dry weight of 1.03 kg/m² with 0.95% N, which leads to a nitrogen content of 9.9 g N/m². Dry matter content of retting material increased from 31% to 86%. C/N-ratio changed from 40:1 to 56:1 due to the reduction of nitrogen content in plant material. This can be partially explained by the loss of leaves and seeds from retting straw. On average the three nitrogen fertilization levels, 16% of dry matter and 40% of nitrogen were lost during retting, corresponding to an absolute nitrogen loss of 67 kg N/ha. These relatively high losses can be regarded as a consequence of the field retting procedure.

Nitrogen balance-
        Nitrogen balance data (Table 2) were calculated from nitrogen input by fertilizer, minus nitrogen export by nitrogen uptake of above-ground plant material at final harvest, which was reduced by 40% from nitrogen losses determined in the retting simulation experiment. Crop residues remaining in the field are not considered. The plot mean nitrogen balance of hemp production was negative. According to the nitrogen fertilization level, mean nitrogen balance ranged from -67 to -8 kg N/ha in N₀ and N₈₀, respectively.
        At a nitrogen fertilization level of 80 kg N/ha, nitrogen import and export were nearly equal (+/-10 kg N/ha). High soil N_min content at the beginning of the field experiment, as well as a great nitrogen mineralization potential after field retting, must be taken into account when interpreting N balance results.

Conclusions
         Hemp is considered to be a very productive crop when cultivated on medium or high quality soil. From the environmental point of view, reduction of pesticides is judged to be positive. With regard to environmental aspects the low nitrogen surplus of hemp N balance due to a high N uptake, has to be mentioned positively. Nevertheless, hemp production, including a field retting period, may cause problems of nitrate leaching in water catchments when high amounts of lost plant material is rapidly decomposed in Autumn. Hence, cropping fiber hemp as silage without field retting should be tested as an alternative method. In the future, soil N dynamics, even after cultivation, have to be integrated into an environmental safe production system.

• Buttlar, H.-B. von et al. 1997. Europaeische Hanfsorten im Standort-vergleich zweier deutscher Anbau-regionen. [Comparing European hemp varieties in two German farming regions] in nova-Institut (Eds): Bioresource Hemp ‘97, Proceedings of the Sympossium Frankfurt am Main, Feb. 27th-March 2nd, 1997 First edition, Hürth, September 1997.
• Bócsa, I. and M. Karus 1997. Der Hanfanbau: Botanik, Sorten, Anbau und Ernte. [The cultivation of hemp: botany, varieties, cultivation and harvest.] C.F. Mueller Verlag, Heidelberg, Germany.
• FNPC (Federation Nationale des Producteurs de Chanvres) 1994. in Katalyse e.V., Institute fuer angewandte Umwelt-forschung (Ed.) Nutzhanf vom konventionellen zum oekologischen Anbau. [Useful hemp - from conventional to organic cropping.] Stiftung Oekologie & Landbau, Sonderausgabe Nr. 67, Deukalion Verlag, Holm, Germany.
• Hanf, C.-H. 1996. Hanf wissenschaftliche Untersuchung zum Thema: Anbau, Ernte und Aufbereitung sowie Verwendung
• von Hanf, Teil 1, Zusammenfassung. [Scientific study on cultivation, harvesting, processing and possible use of hemp, part 1, summary.] Schriftenreihe Nachwachsende Rohstoffe Volume 7 (Eds. Fachagentur Nachwachsende Rohstoffe e.V.), Landwirtschaftsverlag GmbH, Muenster, Germany.
• Hesch, R. 1995. Hanf als Industrierohstoff [Hemp as raw material for industry]. Tagungsskript zum 1. NRW-Hanftag in Haus Duesse am 12.09.1995: Hanf ist (k)eine Wunderpflanze, aber.... Verein zur Foerderung des Anbaus und der Verwertung von Hanf e.V. und Arbeitsgemeinschaft Bauernblatt Westfalen e.V., Rheda-Wiedenbrueck, Germany.
• Höppner, F. and U. Menge-Hartmann 1994. Anbau-versuche zur Stickstoffduengung und Bestandesdichte von Faserhanf. [Cultivation tests on nitrogen fertilization and plant density in fibre hemp.] Landbauforschung Voelkenrode 44 (4): 314-324.
• Kainer, H. 1996. Agrarnutzen von Hanf. [Agricultural use of hemp.] Oesterreichisches HanfMagazin 1/96: 6-7.
• Scheller, E. 1993. Die Stickstoffversorgung der Pflanzen aus dem Stickstoff-Stoffwechsel des Bodens; ein Beitrag zur Pflanzenernaehrungslehre des oekologischen Land-baus. [Nitrogen supply of plants from soil nitrogen metabolism; a contribution to plant nutrition in organic farming.] Verlag Josef Margraf, Weikersheim, Germany.
• Werf, H. van der 1994. Crop physiology of fibre hemp (Cannabis sativa L.). Proefschrift Wageningen, Cip-Gegevens Koninklijke Bibliotheek, Den Haag, The Netherlands.

Acknowledgements
         This research project was financially supported by the Lower Saxony Regional Authority of Ecology (NLOE). We are gratefully indebted to the excellent technical assistance of Gabi Dormann.