Home About Browse Search

Klimatförändringars effekter på jordbrukets växtproduktion i Sverige

scenarier och beräkningssystem

Eckersten, Henrik and Kornher, Alois (2012). Klimatförändringars effekter på jordbrukets växtproduktion i Sverige. Uppsala: (NL, NJ) > Dept. of Crop Production Ecology, Sveriges lantbruksuniversitet. ; 14


Authors/Creators:Eckersten, Henrik and Kornher, Alois
Title:Klimatförändringars effekter på jordbrukets växtproduktion i Sverige
Subtitle:scenarier och beräkningssystem
Alternative abstract:

The English summary of this report is based on a conference abstract of the NJF seminar “Risk assessments/risk management, forecasting pests and diseases of field crops in a changing climate” in Kristianstad, Sweden in November 2011 (Eckersten et al., 2011b).

It has been evaluated that climate change probably will cause increased need of plant protection (Eckersten et al., 2007; 2008; Fogelfors et al., 2009) which in turn would increase the risk of negative impact on the environment. To predict possible changes in the future needs of plant protection in Swedish agriculture it is of interest to know how dates for sowing, phenological stages and harvest might change in future. It is also of interest to know which crops that might be grown. This ultimately depends on which crops the farmer can grow economically, which to a large extent is determined by socio-economic factors, but also on how the yield and quality of the crops change in a changing climate. To provide a base for evaluating changes of future needs of pest control we assessed the sowing dates of spring barley and maize, anthesis date (DC61) in spring barley, floral initiation (DC31) and anthesis dates in winter wheat, flowering date (DC65) in winter wheat and winter rape, harvest date and yield of spring barley and quality, harvest date and yield of fodder maize (Eckersten et al., 2011). We also assessed the change in risk of frost after sowing under climate change, and an index for successful growth of wine.

To assess possible future changes in these variables we further developed a computational framework (Eckersten et al., 2001) for handling inputs, executing the computations and handling the outputs, needed to assess effects on crop of possible future changes of the climate. Inputs are, except for the climate variables, also the crop and cultivar specific parameter of the models. The parameters were either achieved from other experimental studies (Hermann et al., 2005) or derived by comparing modelled values with observed data. We used the Delta-method for downscaling climate change scenarios on a monthly and 50x50km scale to daily and local values. The monthly changes projected by the climate scenarios were applied to daily observed weather (SMHI) during the reference period 1961-87 (~1975; also the reference period 1981-2010 was used). The climate change scenarios of the IPCC A1b-emission scenario were used (SMHI, 2011), giving an average annual temperature increase of 2.5, 3.6 and 3.7oC for Lund (55.6oN), Uppsala (59.8oN) and Luleå (65.5oN), respectively, from ~1975 to ~2085. The corresponding changes in precipitation were +84, +55 and +85 mm, respectively. Solar radiation (global radiation) generally decreased under climate change, although only a little. Direct effects of increased atmospheric carbon dioxide concentrations on crop were not considered. The models used to assess the development stages were the accumulated sum of daily mean air temperature (TDMv) above 5oC from 1 January. At a certain accumulated value (TSum) the phenological stage occurred. TSum was estimated to fit regular observations in Malmöhus County during 1988-2010 made by the Swedish Board of Agriculture as part of their regular monitoring programme of crop status. Sowing and harvest dates of spring barley was estimated with the corresponding temperature sum to fit Swedish crop cultivar trials (2003-2009) and sowing date of maize to fit field experiments in Germany 1998-2002. The yield of spring barley was simulated with the growth model FOPROQ32 (Torssell and Kornher, 1983), and fodder maize yield and quality with the MAISPROQ model (Hermann et al., 2005). Both models depend on a daily weather index for air temperature, global radiation and soil water availability. The spring barley model was calibrated against regional yields in Sweden for selected years between 1965 and 2009, and the fodder maize model against the Janna cultivar in German field trials. Harvest date of maize was set to the date when the simulated dry matter achieved 34% level, or by latest 31 October. For wine the Huglin index based on TDMv and daily maximum temperature was used (cf. Trnka et al., 2011). We also assessed the uncertainties in the projections for differences in climate change scenarios, development stage models, sowing and harvest strategies, calibration data, cultivars, and climate reference period used in the Delta-method for downscaling.
Concerning an English summary of the results we refer to the NJF seminar abstract by Eckersten et al. (2011b).

In conclusion, the projected effect of climate change on development stages of winter wheat, winter rape and spring barley were similar among the crops. The observed changes that have already occurred between the previous “30” year-period (1961-87; ~1975) and the “today” period (1981-2010; ~1995) were quite large in Götaland and Svealand. The future assessments were uncertain, depending on which reference period that was used in the Delta-method for downscaling. Using the reference period ~1975 we achieved a small change in temperature between the nearest future period (2011-2040) and the “today” period (~1995), whereas when using the ~1995 as the reference period a considerable temperature increase was achieved. The reason of this uncertainty remains to be investigated but is speculated to be related to how the modelled climate fit the corresponded observed climate at the locations used in this study.

By the end of the century the development stages might occur about one and a half to two and a half weeks earlier than “today”, and the current dates for Lund might be found in Uppsala, and those of Uppsala today might be found in Umeå, when using the ~1975 reference period in the Delta-method. When the ~1995 reference period, or the A2 climate scenario, was used the corresponding development stages occurred two to more than three weeks earlier than “today”. For sowing dates the projected changes were larger than for development stages, when estimated according to the accumulated temperature sum. However, then there was an increased frost risk which might slightly delay sowing for spring barley, but less for maize. The expansion of fodder maize cultivation area depends on how frequent a successful fodder quality is achieved. It was predicted that by the end of the century Uppsala might achieve about similar frequency of successful years as Lund has “today”. This will cause a considerable advanced harvest date, ranging from end of August in Lund to the beginning of October in Uppsala by ~2085.

The uncertainties in the assessments, in addition to the choice of the reference period used in the Delta-method, were mainly dependent on development stage models and climate scenario inputs used. The daylength based model gave less climate change response and better prediction of between year variations than the model that only was based on temperature, which compared with the daylength model, predicted the observed regional differences better. The county used to calibrate the models had little influence on the assessed response to a climate change, but a considerable influence on the absolute date values of different regions. The uncertainty due to variations in emission scenarios and GCM model version were also high. The A2 based temperature scenario was strongly ahead of the A1b based scenarios.

Series/Journal:Report from the Department of Crop Production Ecology, SLU
Year of publishing :2012
Number of Pages:62
Place of Publication:Uppsala
Publisher:Institutionen för Växtproduktionsekologi, Sveriges lantbruksuniversitet
ISBN for printed version:978-91-576-9067-8
Publication Type:Report
Full Text Status:Public
Agris subject categories.:F Plant production > F01 Crop husbandry
P Natural resources > P40 Meteorology and climatology
Subjects:Obsolete subject words > FORESTRY, AGRICULTURAL SCIENCES and LANDSCAPE PLANNING > Plant production > Agronomy
Keywords:Klimatförändringsscenarier, Grödutveckling, Beräkningar
Permanent URL:
ID Code:8590
Department:(NL, NJ) > Dept. of Crop Production Ecology
External funders:Swedish Board of Agriculture
Deposited By: Prof Henrik Eckersten
Deposited On:14 Feb 2012 11:48
Metadata Last Modified:02 Dec 2014 10:48

Repository Staff Only: item control page