Reg Environ Change (2007) 7:63–77 DOI 10.1007/s10113-007-0032-6
ORIGINAL ARTICLE
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine for 2000 to 2070 I. A. Romanenko Æ V. A. Romanenkov Æ P. Smith Æ J. U. Smith Æ O. D. Sirotenko Æ N. V. Lisovoi Æ L. K. Shevtsova Æ D. I. Rukhovich Æ P. V. Koroleva
Received: 20 March 2007 / Accepted: 20 March 2007 / Published online: 25 May 2007 Springer-Verlag 2007
Abstract This study estimates the consequences of climate change on cropland with and without implementation of adaptation measures, paying special attention to the maintenance of soil organic carbon (C) stocks. We examine the possibility for regional sustainable agricultural management practice that combines both maintenance and gain in soil carbon level with profit maximization. Future scenarios of Regional Agricultural Production Systems (RAPS) were constructed for 2000–2070 based on linking the effects of global climate change, predicted change in
productivity parameters for the main agricultural crops, land-use and soil database parameters. The RAPS were used to examine profitability and feasibility of alternative agricultural scenarios, based on an economic model. A number of recommendations for decision making were proposed based on an assessment of the efficiency of adaptation in animal husbandry and in the crop production sector, after analysis of current percentage of perennial grass in rotation in comparison with future economic scenarios. Keywords Sustainable agriculture Economic model Soil organic carbon Adaptation Climate change
Figures in color are available at http://agro.geonet.ru/articles/carbon/ I. A. Romanenko All-Russian Institute of Agricultural Problems and Informatics, Kharitonievsky per., 21/6 Bld.1, 103064 Moscow, Russia V. A. Romanenkov (&) L. K. Shevtsova Pryanishnikov All-Russian Institute of Agrochemistry (VNIIA), Pryanishnikova st., 31a, 127550 Moscow, Russia e-mail:
[email protected] P. Smith J. U. Smith School of Biological Sciences, University of Aberdeen, Cruickshank Building, St. Machar Drive, Aberdeen AB24 3UU, UK O. D. Sirotenko All-Russian Institute of Agricultural Meteorology, Lenina st., 82, Obninsk, Kaluga Region 249020, Russia N. V. Lisovoi Institute for Soil Science and Agrochemistry Research named after O.N. Sokolovsky, Chaikovskogo st., 4, 61024 Kharkiv-24, Ukraine D. I. Rukhovich P. V. Koroleva Dokuchaev Soil Science Institute, Pyzhevsky Per.,7, 109017 Moscow, Russia
Abbreviations RAPS Regional agricultural production systems AEZ Agro-ecological zone GHG Greenhouse gas GIS Geographical information system SOC Soil organic carbon BAU Business as usual (BAU) scenario, optimal soil management scenario (OPT) and economically and environmentally sustainable agriculture (SUS) scenario SB Spring barley PG Perennial grass P Potatoes AG Annual grass APG Mixture of perennial and annual grass GO Green oats GVO Green vetch-oats SC Silage corn WR Winter rye WW Winter wheat O Oats SBPG Spring barley with perennial grass
123
64
OPG G PS
I. A. Romanenko et al.
Oats with perennial grass Grass Peas
Introduction The impacts of climate change on agriculture are highly dependent on complex interactions among a range of processes, including environmental, economic and technological ones. Based on a regional aggregation of climate scenarios, Fischer et al. (2005) found considerable potential for expansion of potential agricultural land in high latitudes, estimated for the Russian Federation as a 64% increase by 2080, with a decrease of the total area of prime agricultural land, particularly in the Ukraine. Crop growth model simulations for different Russian regions suggest that the net effects of future climate on crop productivity are positive, especially if adaptation to climate change is implemented, but with projected yield declines in the North Caucasus and Siberian regions. There is considerable potential for expansion of suitable land for various crops; for example, more than a fivefold increase for early ripening grain corn will be possible. A potential increase of 15–20% in total cereal-production potential and a 20–30% increase in fodder production potential are expected, even without adaptation (Izrael and Sirotenko 2003). Thus, a gain in potential agricultural land is possible for grain crops, early ripening varieties of vegetables, potatoes, sunflower and sugar beet. Introduction of labour-consuming industrial, and vegetable, crops will result in higher production costs, but this will be balanced by higher profitability, providing the opportunity for using increased incomes for agricultural improvements. Future climate is expected to change considerably the agricultural production systems in the Russian Federation and the Ukraine, but could also enhance ecological problems, cause structural economic reorganization and require the creation of new land-use and land management systems. New farming systems will require considerable economic input, not only to ensure improved production, but also to maintain or improve environmental standards. Agricultural intensification and expanding cropland area can potentially lead to additional greenhouse gas (GHG) emissions (Smith et al. 2007a). Reallocation of capital and human resources is necessary to mitigate the negative effects of global climate change on the economy and the environment by economic adjustment. The three most common approaches to estimate the impact of climate change on agriculture are cross-sectoral models, agronomic-economic models (Rounsevell et al. 2006) and
123
agro-ecological zones (AEZ) models (Mendelsohn et al. 2001; Fischer et al. 2005; Metzger et al. 2006; Schro¨ter et al. 2005). The first two approaches usually do not include adaptation, while the more comprehensive AEZ approach gives the best estimates based on existing soil– climate–crop relationships. For using AEZ as a forecasting tool, linking economic variables into the AEZ model is necessary. Mitigation options in agriculture include reducing the source of GHG emissions or using as a sink for carbon (C) storage (McCarl and Schneider 2000; Lewandrowski et al. 2004; West and Post 2002; Smith 2004a). Agriculture and forestry, operating on an extensive land base, represent relatively low-cost options to mitigate the buildup of GHGs and are the dominant activities for carbon/ GHG trading with a value of approximately USD10 tCO2-eq.–1 or lower (Murray 2004). For cropland, the activities that have the highest potential for storing C are afforestation, conversion of cropland to permanent grasses and conservation tillage, and better use of organic inputs (Smith et al. 2000). Lower C storage potentials include changing crop rotation, expanding the use of winter cover crops, eliminating fallow periods, changing fertilizer management, using more organic soil amendments, improving irrigation, shifting land to conservation buffers and restoring wetlands (Lewandrowski et al. 2004; Smith et al. 2000). The realistically achievable potentials for C sequestration are substantially lower and estimated at about 10–20% of the biological potential, suggesting that high variability of economic factors may limit the adoption of C sequestering practices (Smith 2004a). Not surprisingly, among numerous papers on the economics of controlling GHG emissions very few have focused on the analysis of policies to encourage of C sequestration in agricultural soils (Paustian et al. 2004). In this paper, we focus on the methodology for estimating possible climate-change adaptation in agriculture at the regional scale to provide an increase of incomes and potential C sequestration based on a set of economic indicators for European Russia and the Ukraine. The relationship between economic adaptation to climate change and C stock management in agricultural ecosystems for use in developing long-term adoption strategies at regional level is shown on Fig. 1. The objectives of this study were to: (1) integrate the impacts of different plausible future ecological, economic and climatic changes into consistent future scenarios for Regional Agricultural Production Systems (RAPS), (2) develop a methodology for modelling soil C and land management changes of RAPS as affected by climate, (3) develop a set of indicators for representing sustainable development of RAPS, and (4) propose links between the indicators and the basic parameters of the RAPS.
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine
Fig. 1 Structure of the ecologic-economic model for analysis of alternative management scenarios
Study region The region, which is only about 20% of the country’s territory, includes 74 Mha of arable land in European Russia, and represents 59% of total Russian arable land (Romanenko et al. 1998). It comprises 47 administrative regions, which are the basic units for agricultural statistics and economic analysis (Major indicators of RF 2000). Soil map of the Russian Federation, 1:2,500,000, Land use map of the USSR, 1:4,000, USSR Political and Administrative Map, 1:8,000,000, Natural and Agricultural Zoning Map were adjusted to administrative boundaries (Rukhovich et al. 2007). This gives 200 U that are assumed to be homogenous with respect to soil, climate, economic and land-use parameters. The procedure was slightly different for Ukraine, as this region is much more uniform with respect to the above-mentioned factors, and so the aggregation procedure results in fewer, larger units. For 41 million ha of arable land, 12 landscape-production areas were identified by overlaying meteorological and land-use information with current agricultural statistics for 25 administrative regions of the Ukraine. The details for this linkage within a Geographical Information System (GIS) are described in Rukhovich et al. (2007).
Methods Modelling methodology for RAPS The methodology for construction of the alternative agricultural production scenarios at regional level includes profitability and feasibility analyses based on assessment of the effect of global climate change on productivity parameters for the main agricultural crops, cost efficiency of crop growth and cattle breeding. For constructing future economic scenarios, a regional economic model is applied.
65
The methodology for construction of the scenarios was based on a description of RAPS, which were constructed by linking information on crop rotations, fertilization practices, vegetation period and crop growth characteristics (Romanenko 2005a), all of which are needed by models of C dynamics, with combinations of soil and climate data within the system. Regional specialization in the long-term is connected with changes in the highest possible yield for the main crop/region driven by climatic scenarios. Baseline figures of crop productivity were average yield data for 1990–2000 for the administrative regions, available from agricultural statistics (Agriculture in Russia 1998, 2002). The procedure of scenario construction includes the following successive steps: (a)
summing croplands for all crops within the classes: cereals, row crops, and grass; (b) defining the dominant crop within each class/region (potatoes or sugar beet for row crops, wheat or barley for grain crops, etc.); (c) identification of attributive data (the highest possible and real yields, sale price of the specific product, etc.) assuming only dominant crop growing within the class; and (d) calculation of costs for the specific crop across regions. The normative sources for determining the structure of livestock farming and crop production sectors within each unit are soil texture, initial soil C content (soil database), potential yield of the main crops (a dynamic crop growth model Climate–Soil–Yield outputs according four different climate scenarios; Smith and Powlson 2003; Smith et al. 2007b, c), milk, livestock and crop production input and output standards per 1 U (head) and 1 ha (Kuznetsov et al. 2002), metabolizable energy and dry matter per unit weight of the feedstuff component (Planning Agriculture Handbook 1974), constraints on regional crop rotation systems and share of the foodstuff components in animal rations (Planning Agriculture Handbook 1974), with correction of mineral fertilization rates so that they do not exceed ecological safe rates. The regional economic model we use includes an interrelated system of several submodels (Ognivtsev and Siptitz 2002), which tracks the processes of agricultural crop production, livestock farming production (separately for different animal systems), fodder production and soil fertility reproduction, defined for the available land resources of the RAPS. Relationships within the each submodel are based on balance calculations, linear and nonlinear functions and normative information. The model outputs for future scenarios assume regional RAPS change as a result of land-holder adaptation to climatic change only in the livestock farming and crop
123
66
I. A. Romanenko et al.
production sectors. A distinct feature is the assessment of soil fertility dynamics based on driving parameters from other submodels (crop yield, mineral and organic fertilizer rates, etc.). The list of necessary equations in the economic model includes: • • • • • • •
Soil fertility function; Manure production function (depends on livestock structure and number); Livestock structure and number functions (depends on feed production volume and structure); Feed production volume and structure (depends on crop production structure); Crop production structure (depends on arable land structure); Crop yield functions; and Cost functions for crop and livestock production (Fig. 1).
The model provides information on costs, incomes and profits for the specific production practices (including crop rotations and production inputs) using a whole-system approach. The RAPS parameters that define the profit gain or loss are the percentage of cows in the cattle herd, and the share of marketable production in the crop sector. Percentage of cows in the cattle herd determines whether a farm specializes in milk or meat production. Milk production in most regions of Russia is currently profitable, but meat production is not (Agriculture in Russia 2002). The ratio of market prices for different types of agricultural production is considered to be stationary. Linking of the model results with map units is based on the definition of a 10-year crop sequence within the calculated crop pattern. Changes in the livestock farming and crop production sectors are based on the solution of general linear programming (Ognivtsev and Siptitz 2002; Romanenko 2005b, c). The model solves a profit maximization routine based on costs and profits per unit of production using linear constraints, the main of which are: • • •
arable land cannot exceed the amount existing in the region; percentage of cows in the cattle herd cannot exceed 67% (reproduction constraint); and share of the foodstuff components in animal rations is not more than normative ones.
The main output parameters for evaluation include the structure of arable land, application rates of FYM and mineral fertilizers, percentage of cows in the cattle herd, the structure of livestock feed rations, and the share of marketable production in the crop growth sector. Most crop growth models take account of resource variability (such as water and nutrients) but do not consider
123
social or economic factors and, hence, cannot be applied directly to simulate both adaptation in agriculture and climate-induced changes (Reilly 2002). To eliminate this limitation, a dynamic crop growth model Climate–Soil– Yield (Sirotenko et al. 1995) has been run separately for each 10-year interval for all Russian and Ukrainian administrative regions (2000–2050), and the results of the crop model were then used as inputs for the economic model to select adaptation strategies. To consider differences between ‘‘adaptation’’ and ‘‘no adaptation’’ response by crop producers to climate change, a 20-year interval was applied for comparison since sequestration activities need to be practiced for periods of at least 15–20 years to provide sufficient time of terrestrial C sequestration (Lewandrowski et al. 2004) and to be able to detect soil organic carbon (SOC) differences with a reasonable sample size (Smith 2004b). We also consider a fast adoption of new feasible land management structures in 2010, 2030 and 2050 (i.e. autonomous adaptation) as a reasonable response to climate change (Romanenkov et al. 2005) to distinguish it from planned capital replacement. This approach allows us to determine whether, and what type of, action in land management might be taken depending on the climate impact. The crop model projected grain crop (with no detail for the specific cereal) and grass yield changes, given as percentages under limited or optimal N fertilization in dry-land conditions (Sirotenko et al. 1995). Simulated climate outputs were the same as those used in Smith et al. (2007c), i.e. data 2000–2070 from the HadCM3 climate model (IPCC 2001) using four IPCC emission scenarios—A1FI, A2, B1, B2 (Nakic´enovic´ et al. 2000)—were used for the crop modelling. HadCM3 provides a warming projection for the region by the year 2010 in the middle of the range given in IPCC. The climate database with 0.5 resolution and monthly mean data was used as inputs for those sites within the administrative divisions at which the crop model was run. The crop model considers CO2 effects on the efficiency of photosynthesis and water use by crops. Other model outputs include a shift in emergence and harvest dates due to climate change, and changes in crop growth rate in the specific administrative divisions (Sirotenko et al. 1995). Baseline figures for the calculation of crop productivity change were average yield data for 1990–2000 for the administrative regions, available from agricultural statistics (Agriculture in Russia 1998, 2002). Change in the yield of row crops was calculated separately for sugar beet, sunflower, fodder roots and potatoes as follows: modern agroclimatic analogues of future climate were found based on the sum of T > 10C, the vapour deficit and the temperature of the coldest month of the year (Sirotenko and Pavlova 2003). For the RAPS systems without adaptation, current yields of modern analogues were used for the economic modelling. For the system including adaptation,
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine
the highest possible yield under optimal mineral fertilization strategy of the current agro-climatic analogue for the specific row crop/region was used (Normatives for estimation mineral fertilizer requirements in agriculture 1985). At the same time, extrapolation from the modelled grain crops and grasses based on the statistical approach for row crop yield prediction was unsatisfactory. The possible role of soil in this is important, but its exact role remains uncertain and adds complexity to the assessment. Besides, analysis of potato yields under optimal fertilization revealed only small changes from region to region. The economic model describes the crop rotation structure as the percentage of cereals, row crops and grass (annual or perennial). Defined rotational patterns for each region were constructed based on a 10-year crop sequence, in accordance with agricultural statistics and expert knowledge. The output percentages were rounded to the nearest 10. The rotation for the ‘‘no adaptation’’ response within the specific region was constructed based on an appropriate rotation limited by regional agricultural statistical data about the area of different crops, and generally accepted crop sequences in the each region (Vorobyev et al. 1991; Kashtanov et al. 1994). The baseline year of the crop rotation for the RAPS system was the 2000 growing season. All future changes in rotations were imposed as changes relative to the baseline sequence of crops, or the introduction of a completely new rotation pattern. The latter was constructed based on the current agro-climatic analogues of future climate, or on the current rotations of nearby regions. In continental areas, which are predicted to experience a more arid climate in future, fallowing was added, since this is a typical practice for arid agriculture to minimize the effects of drought (Kashtanov et al. 1994). After construction of each crop sequence, a random-number generator was applied, constructed specifically for this task, to arbitrarily move the beginning of the rotation so that each sequence can begin at any phase of the rotation. The procedure of assigning specific crops to the specific year in the RAPS was followed by identifying the sowing and harvesting dates and the fertilization pattern during the 10-year rotation. Future sowing and harvesting shifts were available for spring cereals and grass as outputs of Climate–Soil–Yield model, or were set using expert judgement for winter cereals and row crops, based on the analysis of dynamics of the vegetation periods from current agro-climatic analogues.
67
important to note that at the end of 1980s, inputs in agriculture and yields were at their highest levels. As a result, the economic, crop and agricultural technology parameters of the future scenarios are based on reasonable grounds that indirectly include factors other than climate change, important for agricultural economy (Reilly 2002). Three scenarios were analysed: the business as usual (BAU) scenario, optimal soil management scenario (OPT) and economically and environmentally sustainable agriculture (SUS) scenario. The BAU scenario (without the implementation of any adaptation strategy), assumes crop yield change in 2000–2070 for fixed crop rotations and fertilization patterns. N mineral and FYM fertilizer rates were assumed to stay the same as in 2000 and applied to the most valuable cash crops in the rotation. Not more than 50 kg ha–1 N and 4.9 t ha–1 FYM were applied per year (average rates for arable land in 2000 were 8 kg ha–1 and 0.6 t ha–1, respectively). Farming practices can cause a negative nutrient balance and nutrient mining. This may additionally lead to depletion of the SOC pool (Lal 2004). Two alternative policy scenarios assume the implementation of adaptation in land management. The OPT scenario assumes an optimal RAPS structure for profit maximization. Yield forecasts of Soil–Climate–Yield for optimal N fertilization in dry-land conditions were used. N fertilization rates and timing were also optimized based on fertilizer recommendations for optimal yields (normatives for estimation mineral fertilizer requirements in agriculture 1985). FYM rates were equal to outputs of livestock farming production of the region, and were not allowed to exceed ecologically safe rates. In the SUS scenario, profit maximization was additionally restricted by imposing the condition that soil C must remain the same or increase. The combined effect of different management practices on steady state C values was estimated using the static Model of Humus Balance (Shevtsova et al. 2003). As this model was developed for soddy– podzolic (podzoluvisol) soils, the last scenario was calculated for only the 19 from 47 regions, representing those with podzoluvisols. Comparative characteristics of the three scenarios are shown in Table 1. Tables 2 and 3 show assumed differences in crop rotations and N fertilization for BAU and SUS scenarios in selected regions: Moscow and St. Petersburg (the most economically developed areas), Smolensk, Ryazan and Mariy-El (western, southern and eastern continental parts for the SUS scenario, respectively).
Business-as-usual and alternative policy scenarios For future scenarios, land-use patterns and agricultural prices were assumed to remain at current levels to give an economic assessment of agriculture’s potential to benefit from climate change and to avoid possible losses. It is
Linking modelled soil C and land management changes of RAPS Three soil carbon models—RothC (Jenkinson and Rayner 1977), CANDY (Franko et al. 1995) and the Model of
123
68
I. A. Romanenko et al.
Table 1 Comparative characteristics of the scenarios Scenario
Business-as-usual BAU
Optimal soil management OPT
Economically and environmentally sustainable SUS
Criteria definition
No criteria
Maximum profitability (P)
Maximum P with (DSOC) ‡ 0 constraint
Crop yield calculationa
Climate effect on current crop yield
Economic effective yield change projected at 10-year intervals
Crop rotation
Current 2000 rotation
New rotation pattern based on economic model outputs
Crop growth parameters
Shifts in vegetation period according to crop growth model outputs
Fertilization
Current 2000 rates years
a
Mineral N and FYM—optimal for plant nutrition, FYM rates based on outputs of economic model, correction for not exceeding ecological safe rates
Percentage change from average 1990–2000 data for the administrative regions
Table 2 Crop rotations and N fertilization for selected regions, BAU scenario Region
Units
Crop rotation
N rates per rotation
St. Petersburg
12
P-APG-PG-PG-G-WW-P-APG-PG-APG
240
Smolensk
34
SBPG-PG-PG-G-WW-APG-PG-PG-WR-P
Moscow
43, 45, 52, 54, 56, 57
APG-PG-PG-G-WW-P-SBPG-PG-PG-P
Ryazan
74, 75, 84
SBPG-PG-WW-P-OPG-PG-SB-WR-P-O
83
Mariy El
91, 103
SBPG-PG-PG-WW-P-OPG-PG-WR-P-O
80
Humus Balance (Sirotenko et al. 2002)—were used to estimate the impacts of climate change on agricultural soil carbon stocks in the European Russia and Ukraine. These models have been shown to perform well for agro-ecosystems of the former Soviet Union (Smith et al. 2001; Shevtsova et al. 2003). The dynamic models, RothC and CANDY, were run through the 70-year period of 2000– 2070 for each soil group within each unit (Smith et al. 2007c; Franko et al. 2007). The statistical Model of Humus Balance was tested for assessment conditions of zero C change or C accumulation in the selected years for the zone of podzoluvisols (Romanenkov et al. 2007). The RothC and CANDY model year-end SOC outputs (t ha–1) were multiplied by the proportion of that soil group in the unit and weighted mean soil C values within each unit were calculated for the each year from 2000 to 2070. The resulting values of SOC were analysed to determine the total SOC changes in cropland for 70 years as Tg per unit and assess aggregated impacts of changes in management practices or t ha–1 per unit to access local impacts of the practices. This criterion was used to constrain the management whereby the SOC may not decrease or increase, thereby defining the management used in the SUS scenario. The results were then used for comparing BAU-OPT for the whole territory and BAU-OPT-SUS scenarios for the selected 19 administrative regions as described in Smith et al. (2007c).
123
60 225
Results and discussion Comparison of SOC dynamics in BAU and SUS scenarios Comparing the average SOC stock over the final 10 year rotation (2050–2069) with the starting SOC value shows that SOC stocks can be increased (–0.9, 2.0, 1.2 and 3.1% increase for SUS under the A1FI, A2, B1 and B2 climate scenarios, respectively, compared to losses of 7.5, 5.0, 5.9 and 4.3% for BAU for the same climate scenarios). The economically sustainable management scenario (SUS), though applied for only a limited area within the total region, suggests that for this region at least, economically sustainable land management could not only reverse the negative impact of climate change, but could increase soil carbon stocks, most effectively in the B2 climate scenario (Smith et al. 2007c). Tables 4, 5, 6, 7 demonstrate differences in total SOC for the selected regions (Tables 1, 2) for climate scenarios A1FI (the greatest climate forcing) and B2 comparing BAU and SUS. Within one region, loss over 10 years for BAU within one unit can be as high as 2.8 t C ha–1 for A1FI scenario and 2.1 t C ha–1 for B2 scenario but only for units with high initial SOC stocks. Average losses in 2000–2060 for different units in Ryazan Region were 1.43
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine
69
Table 3 Crop rotations and N fertilization for selected regions, SUS scenario Region
Units
Crop rotation
N rates per rotation
Climate scenario A1FI, 2010–2030 St. Petersburg
12
P-GVO-PG-PG-WW-SC-WW-P-SB-SC*
680
Smolensk
34
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
390
Moscow
43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-P-GO-SBPG-PG-WR-SB
430
Ryazan
74, 75, 84
SBPG-PG-PG-WW-P-OPG-PG-PG-SC-P
480
Mariy El
91, 103
SBPG-PG-PG-WW-SC-GO-PG-PG-WR-P
460
12
P-SBPG-PG-PG-WW-SC-GVO-WR-P-SC
660
Smolensk
34
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
390
Moscow
43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-P-GO-SBPG-PG-WR-SB
430
Ryazan
74, 75, 84
SBPG-PG-PG-WW-P-OPG-PG-PG-SC-P
480
Mariy El
91, 103
SBPG-PG-PG-WW-SC-GO-PG-PG-P-O
420
St. Petersburg
12
P-SBPG-PG-PG-WW-SC-GVO-WR-P-SC
660
Smolensk Moscow
34 43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P GVO-PG-PG-WW-P-GO-SBPG-PG-WR-SB
390 430
Ryazan
74, 75, 84
SBPG-PG-PG-WW-P-OPG-PG-PG-SC-P
480
Mariy El
91, 103
SBPG-PG-PG-WW-SC-GO-PG-PG-P-O
420
A1FI, 2030–2050 St. Petersburg
A1FI, 2050–2070
Climate scenario B2, 2010–2030 St. Petersburg
12
SBPG-PG-PG-WW-SC-SC-PS-WR-P-P
590
Smolensk
34
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Moscow
43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Ryazan
74, 75, 84
SBPG-PG-WW-SC-SC-PS-WR-P-P
700
Mariy El
91, 103
SBPG-PG-PG-WW-SC-GO-PG-PG-WR-P
510
Climate scenario B2, 2030–2050 St. Petersburg
12
SBPG-PG-PG-WW-SC-SC-PS-WR-P-P
590
Smolensk
34
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Moscow
43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Ryazan
74, 75, 84
SBPG-PG-WW-SC-GO-PG-PG-WR-P
500
Mariy El 91, 103 Climate scenario B2, 2050–2070
SBPG-PG-PG-WW-SC-GO-PG-PG-WR-P
510
St. Petersburg
12
SBPG-PG-PG-WW-SC-SC-PS-WR-P-P
590
Smolensk
34
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Moscow
43, 45, 52, 54, 56, 57
GVO-PG-PG-WW-SB-GO-PG-PG-WR-P
420
Ryazan
74, 75, 84
SBPG-PG-WW-SC-GO-PG-PG-WR-P
500
Mariy El
91, 103
SBPG-PG-PG-WW-SB-GO-PG-PG-WR-P
510
and 1.23 t C ha–1 for A1FI and B2 scenarios, respectively, compared with 0.54 and 0.4 t C ha–1 losses in Moscow Region, where SOC stocks were 2–3 times smaller. These losses are comparable to the average losses of SOC between 2000 and 2060 in European Russia and Ukraine for BAU of 0.08–0.17 t C ha–1 year–1, with possible increased losses in the chernozem zone (Smith et al. 2007c). SUS management was able to provide 50–70% reduction in SOC loss compared with BAU for Mariy El Republic, 30– 50% for Ryazan Region, 26–100% for some units in Moscow
Region and 0.8–2.8 t C ha–1 C stock increase for the others, 5.8–7.0 t C ha–1 C stock increase for Smolensk Region and was ineffective for the St. Petersburg Region. A fast adoption of new land management structures in 2010 was able to reverse SOC loss or drastically decelerate it. This SOC gain was not unidirectional; usually it reached a steady state or had a flex point. This change in SOC, for all climate scenarios, is pronounced in the mid-term of the considered period, i.e. around 2030, and can be estimated from unequal percent changes of soil C (Tables 4–7). It is mainly
123
70
I. A. Romanenko et al.
Table 4 Soil carbon (t C ha–1) per polygon and percent change in soil carbon for selected regions, BAU scenario, climate A1FI Unit Arable land (ha)
Soil carbon (t C ha–1) per polygon (weighted mean)
% Change in soil C
2000 2010 2020 2030 2040 2050 2060 2010/2000 2020/2010 2030/2020 2040/2030 2050/2040 2060/2050 12
412,837.8 36.8
37.3
37.4
37.2
37
36.6
35.7
1.4
0.3
–0.5
–0.5
–1.1
–2.5
34
2,439,731 29.5
30.8
31.6
31.8
31.6
31.3
31.1
4.4
2.6
0.6
–0.6
–0.9
–0.6
43
457,074.4 37.6
37.5
37.3
36.9
36.2
35.5
34.8
–0.3
–0.5
–1.1
–1.9
–1.9
–2.0
45
226,827.2 36.7
36.5
36.3
35.8
35.1
34.3
33.6
–0.5
–0.5
–1.4
–2.0
–2.3
–2.0
52
301,187.2 38.6
38
37.4
36.8
36.1
35.5
34.7
–1.6
–1.6
–1.6
–1.9
–1.7
–2.3
54
377,020.6 38.5
38.1
37.5
37
36.3
35.5
34.7
–1.0
–1.6
–1.3
–1.9
–2.2
–2.3
56 57
208,583 43.4 801,812.8 22.6
43.1 22.4
42.7 22.2
42.1 21.9
41.2 21.6
40.2 21.1
39.4 20.5
–0.7 –0.9
–0.9 –0.9
–1.4 –1.4
–2.1 –1.4
–2.4 –2.3
–2.0 –2.8
74
1,428,118 72.9
71.9
70.9
70
69.1
68
66.2
–1.4
–1.4
–1.3
–1.3
–1.6
–2.6
75
381,151.1 98.8
97.5
96.1
94.4
92.4
90.5
87.7
–1.3
–1.4
–1.8
–2.1
–2.1
–3.1
84
339,751
96.1
95.1
94.1
93.1
92.1
90.4
87.7
–1.0
–1.1
–1.1
–1.1
–1.8
–3.0
91
59,739.7
37.8
37.4
36.9
36.4
35.8
35.2
34.3
–1.1
–1.3
–1.4
–1.6
–1.7
–2.6
103
276,507.9 34.1
33.9
33.5
33
32.4
31.7
31
–0.6
–1.2
–1.5
–1.8
–2.2
–2.2
Table 5 Soil carbon (t C ha–1) per polygon and percent change in soil carbon for selected regions, BAU scenario, climate B2 Unit Arable land (ha)
Soil carbon (t C ha–1) per polygon (weighted mean)
% Change in soil C
2000 2010 2020 2030 2040 2050 2060 2010/2000 2020/2010 2030/2020 2040/2030 2050/2040 2060/2050 12
412,837.8 36.7
36.8
36.7
36.5
36.5
36.4
36.1
0.3
–0.3
–0.5
0.0
–0.3
–0.8
34
2,439,731 29.5
30.8
31.6
31.9
32.1
32.2
32.1
4.4
2.6
0.9
0.6
0.3
–0.3
43
457,074.4 37.6
37.5
37.3
37.0
36.7
36.3
35.9
–0.3
–0.5
–0.8
–0.8
–1.1
–1.1
45
226,827.2 36.7
36.5
36.2
35.9
35.5
35.1
34.7
–0.5
–0.8
–0.8
–1.1
–1.1
–1.1
52
301,187.2 38.6
38.0
37.4
36.9
36.4
35.9
35.1
–1.6
–1.6
–1.3
–1.4
–1.4
–2.2
54 56
377,020.6 38.5 208,583 43.5
38.1 43.1
37.7 42.6
37.2 42.1
36.7 41.4
36.1 40.8
35.6 40.2
–1.0 –0.9
–1.0 –1.2
–1.3 –1.2
–1.3 –1.7
–1.6 –1.4
–1.4 –1.5
57
801,812.8 22.6
22.6
22.3
22.1
21.9
21.6
21.3
0.0
–1.3
–0.9
–0.9
–1.4
–1.4
74
1,428,118 72.9
72.0
70.8
69.9
69.1
68.2
67.3
–1.2
–1.7
–1.3
–1.1
–1.3
–1.3
75
381,151.1 98.8
97.6
96.1
94.5
92.7
90.8
88.7
–1.2
–1.5
–1.7
–1.9
–2.0
–2.3
84
339,751
96.1
95.0
94.0
93.0
92.1
90.9
89.6
–1.1
–1.1
–1.1
–1.0
–1.3
–1.4
91
59,739.7
37.9
37.5
37.1
36.7
36.2
35.6
35.0
–1.1
–1.1
–1.1
–1.4
–1.7
–1.7
103
276,507.9 34.1
33.9
33.5
33
32.4
31.7
31
–0.6
–1.2
–1.5
–1.8
–2.2
–2.2
connected with maximum increases in the potential yield in the SUS scenarios in this period for the territory where SUS scenario was applied (Sirotenko 2005). For Ryazan Region, SUS management was equally effective in SOC stock maintenance in 2010–2040, followed by a decline in 2040– 2060, and was more pronounced for the A1FI climate scenario than for B2. For Moscow Region, some acceleration of C loss was projected for BAU and SUS management in 2040–2060 for the A1FI climate scenario, while the greatest accumulation takes place around 2030. The main differences
123
in C accumulation between BAU and SUS management for Smolensk Region were also reached before 2040. Different regions, therefore, show similar patterns of SOC accumulation through time under SUS management, but even within one region introduction of sustainable practice can reverse climate driven loss of cropland soil carbon for some units, but only decelerate the loss in the neighbouring regions. For estimating the different behaviour of units in the same region, it is desirable to analyse the costs for maintaining or increasing SOC stocks.
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine
71
Table 6 Soil carbon (t C ha–1) per polygon and percent change in soil carbon for selected regions, SUS scenario, climate A1FI Unit Arable land (ha)
Soil carbon (t C ha–1) per polygon (weighted mean)
% Change in soil C
2000 2010 2020 2030 2040 2050 2060 2010/2000 2020/2010 2030/2020 2040/2030 2050/2040 2060/2050 12
412,837.8 36.7
36.5
36.6
36.8
36.2
35.2
34.3
–0.5
0.3
0.5
–1.6
–2.8
–2.6
34
2,439,731 29.5
30.8
33.3
34.5
34.9
35.1
35.3
4.4
8.1
3.6
1.2
0.6
0.6
43
457,074.4 37.6
37.5
37.7
37.5
37
36.4
35.8
–0.3
0.5
–0.5
–1.3
–1.6
–1.6
45
226,827.2 36.7
36.5
36.6
36.4
35.9
35.2
34.6
–0.5
0.3
–0.5
–1.4
–1.9
–1.7
52
301,187.2 38.6
38
37.7
37.4
37
36.5
35.8
–1.6
–0.8
–0.8
–1.1
–1.4
–1.9
54
377,020.6 38.5
38
37.8
37.5
37.1
36.5
35.7
–1.3
–0.5
–0.8
–1.1
–1.6
–2.2
56 57
208,583 43.4 801,812.8 22.6
43.1 22.4
43.5 23
43.5 23.4
42.8 23.5
42 23.2
41.4 22.8
–0.7 –0.9
0.9 2.7
0.0 1.7
–1.6 0.4
–1.9 –1.3
–1.4 –1.7
74
1,428,118 72.9
71.9
72.7
72.4
72
71.2
69.4
–1.4
1.1
–0.4
–0.6
–1.1
–2.5
75
381,151.1 98.8
97.5
97.5
96.7
95.2
93.4
90.6
–1.3
0.0
–0.8
–1.6
–1.9
–3.0
84
339,751
96.1
95.1
95.5
95.5
95.1
93.7
90.9
–1.0
0.4
0.0
–0.4
–1.5
–3.0
91
59,739.7
37.8
37.4
37.7
37.6
37.5
37.1
36.3
–1.1
0.8
–0.3
–0.3
–1.1
–2.2
103
276,507.9 34.1
33.9
34.2
34.2
34.1
33.7
33
–0.6
0.9
0.0
–0.3
–1.2
–2.1
Table 7 Soil carbon (t C ha–1) per polygon and percent change in soil carbon for selected regions, SUS scenario, climate B2 Unit Arable land (ha)
Soil carbon (t C ha–1) per polygon (weighted mean)
% Change in soil C
2000 2010 2020 2030 2040 2050 2060 2010/2000 2020/2010 2030/2020 2040/2030 2050/2040 2060/2050 12
412,837.8 36.7
35.1
35.2
35.1
35.0
35.0
34.7
–4.4
0.3
–0.3
–0.3
0.0
–0.9
34
2,439,731 29.5
30.8
33.4
34.8
35.7
36.3
36.6
4.4
8.4
4.2
2.6
1.7
0.8
43
457,074.4 37.6
37.5
38.2
38.4
38.5
38.3
38.1
–0.3
1.9
0.5
0.3
–0.5
–0.5
45
226,827.2 36.7
36.5
37.2
37.4
37.3
37.1
36.9
–0.5
1.9
0.5
–0.3
–0.5
–0.5
52
301,187.2 38.6
38.0
38.4
38.3
38.2
37.9
37.5
–1.6
1.1
–0.3
–0.3
–0.8
–1.1
54
377,020.6 38.5
38.1
38.5
38.7
38.5
38.1
37.7
–1.0
1.0
0.5
–0.5
–1.0
–1.0
56 57
208,583 43.5 801,812.8 22.6
43.1 22.6
43.5 23.7
43.5 24.4
43.3 24.8
43.1 24.9
42.6 24.8
–0.9 0.0
0.9 4.9
0.0 3.0
–0.5 1.6
–0.5 0.4
–1.2 –0.4
74
1,428,118 72.9
71.8
72.1
72.1
72.1
71.8
71.1
–1.5
0.4
0.0
0.0
–0.4
–1.0
75
381,151.1 98.8
97.6
97.5
96.5
95.5
94.0
92.2
–1.2
–0.1
–1.0
–1.0
–1.6
–1.9
84
339,751
96.1
95.0
95.3
95.2
95.1
94.5
93.6
–1.1
0.3
–0.1
–0.1
–0.6
–1.0
91
59,739.7
37.9
37.5
38.1
38.3
38.2
37.9
37.4
–1.1
1.6
0.5
–0.3
–0.8
–1.3
103
276,507.9 34.1
33.9
34.6
34.7
34.6
34.5
34.2
–0.6
2.1
0.3
–0.3
–0.3
–0.9
Indicators of sustainable development Control of SOC stocks through changes in agricultural practices is geographically dependent and can be substantially different for the adjacent regions. Figure 2 demonstrates changes in costs for maintaining or increasing SOC stocks by 2050 according to four climate scenarios for 19 regions representing the podzoluvisol zone. The most favourable is the scenario with the lowest climate forcing—B1, while the scenario with greatest climate forcing (A1FI) has the largest additional cost for
preventing SOC losses. Climate scenarios B2 and A2 produce similar cost patterns. Despite differences among scenarios, there is a core of regions in the central part of the zone where SOC can be maintained or increased either at zero, or minimal, additional cost. The most sensitive are regions in the south-eastern and north-western part of this territory. In the continental southeastern part, costs of maintaining of increasing SOC can be as high as 30–40%, or even 50–80% of total RAPS incomes. All of the main economic RAPS parameters for SUS management under A1FI and B2 scenarios show a uniform
123
72
I. A. Romanenko et al.
Fig. 2 Maps showing the change in costs for maintaining or increasing SOC stocks in 2050 across the four climate scenarios for 19 regions of European Russia representing the podzoluvisol zone, as a percentage of total RAPS income; a A1FI, b A2, c B1 and d B2
response except for profit. Variation of total livestock farming production is usually in the range 1–5%, and rarely 15–19% (Ryazan Region, Chuvashia; Table 8). Cereal productivity as estimated by the Soil–Climate–Yield model shows more than a 10% increase for the B2 scenario, compared to A1FI in the following regions: Mordovia Republic (19%), Ryazan (16%), Bryansk (13%), Nizhny Novgorod and Kaluga (12%). Increase in crop productivity determines additional C input to the soil through extra crop residues, but is not able to account for the substantial changes in costs for maintaining or increasing SOC stocks for neighbouring regions. There is 20–40% ha arable land–1 profit increase for the above-mentioned regions for the B2 climate scenario, relative to A1FI. In earlier studies of the comparative efficiency of different strategies for preventing SOC loss in arable podzoluvisols, the most effective strategy was found to be an increase in the percentage of grass in the crop rotation at the expense of row crops, followed by FYM applications and changes in the rates of mineral N application (Romanenkov et al. 2001). Comparison of changes in the percentage of perennial grass in the rotation 1990–2000, with those calculated by the economic model for the B1 emission scenario in 2050 for 19 regions under OPT and SUS management shows that the current percentage in the centre of the zone (50–70%) exceeds the threshold for SOC loss prevention under 2050 climatic conditions, except for
123
in the Bryansk, Ryazan and Moscow Regions (Table 9). For regions in the south-eastern part of the territory, the percentage of perennial grass in the rotation is always less than that necessary for the SUS scenario. This is sometimes substantial, with a deficit of more than 25–35%. The percentage of grass in the crop rotation is a useful practice for preventing SOC loss, and provides a cost-free or minimal cost basis for implementation of the SUS scenario. Another indirect indicator of change in the grassland and row crop area is the stocking density per ha of cropland (Table 10). Stocking density reduction under the SUS scenario in 2050 compared with the BAU scenario (without adaptation) means reduction of feedstuff ratio in the RAPS production sector and, hence, leads to additional costs for maintaining the SOC level. This is especially important when stocking density is far less than necessary for OPT management. The economic model indicates this to be the case for the SUS management in St. Petersburg, Vladimir, Moscow and Yaroslavl Regions in 2050. In these cases, when climate effects on crop production are more pronounced than on the animal husbandry RAPS sector, which holds for the study zone (Table 8), analysis of the future trends in these indicators is useful in defining the territories where costs for SUS management are predominantly climate-mediated, compared to OPT management. It is important to note that the most economically developed areas, such as Moscow
1.08
1.08 1.12
1.12 1.06
1.06 1.16
1.16 1.10
1.10 1.06
1.06 1.12
1.12 1.13
1.13 1.11
1.11 1.19
1.19 1.13
1.13 1.10
1.10 1.07
1.07
1.03 Poultry production
1.03
1.03 Eggs production
1.03
1.01
1.06 1.11
1.02 0.98
1.04 1.10
1.07 1.01
1.09 1.04
1.00 1.02
1.10 1.12
1.03 1.00
1.18 0.80
1.19 1.04
1.13 1.12
0.99 1.00
1.04 Pork production, life weight
1.03 0.98
1.02
Beef production
0.97
1.01
1.01 1.01 1.02 1.02 0.98 0.98 1.07 1.07 1.01 1.01 1.00 1.00 1.02 1.02 1.03 1.03 1.00 1.00 1.19 1.19 1.04 1.04 0.99 0.99 1.03 1.03 Salable milk product 0.98 Total meat production, 0.98 life weight
1.00 1.00
1.01 1.02
1.02 0.98
0.98 1.07
1.07 1.01
1.01 1.00
1.00 1.02
1.02 1.03
1.03 1.00
1.00 1.19
1.19 1.04
1.04 0.99
0.99 1.00 1.03 0.98 Total milk production
1.03 0.98 Stock density
1.00
1.03 1.04 1.00 1.09 1.03 1.01 1.05 1.05 1.02 1.19 1.06 1.02 1.03 0.99 Organic fertilizer rate
1.02
1.00
1.08 1.12
1.03 0.95
1.06 1.16
1.10 1.02
1.10 1.06
0.98 1.02
1.12 1.13
1.04 1.04
1.11 1.19
1.05 1.06
1.13 1.10
1.01 0.97
1.07 1.03
0.98
1.03
0.95 Perennial grass yield
1.31 1.08 1.83 1.17 1.10 1.22 1.25 1.33 4.78 1.37 1.16 1.09 1.05
Grain crops yield
1.02 Income per 1 ha of cropland
1.12
St. Petersburg Novgorod Pskov Vladimir Mariy El Mordovia Chuvashia Bryansk Kaluga Kostroma Moscow Ryazan Kirov Nizhny Smolensk Novgorod Parameter
Table 8 RAPS economic model parameters for SUS management in 2050, A1FI/B2 ratios in different regions
Constructing regional scenarios for sustainable agriculture in European Russia and Ukraine
73
and St. Petersburg regions, which have a more favourable capital-labour ratio and infrastructure, are in the above mentioned list. For these regions, the difference in costs of maintaining of increasing SOC between SUS and OPT management is 15–30%, depending of the emission scenario. In the continental part of the zone (Upper Volga Region), the soil and climate conditions are generally favourable for implementing intensive technologies to produce industrial crops and cereals. As a result, the percentage of perennial grass in the rotation is not sufficient for maintaining SOC stocks. The difference in the percentage of grass in the crop rotation between SUS and OPT management scenarios in 2050 is 40–50%, while that between BAU and OPT in 2050 is 20–30% (Table 9). In this case, the costs for maintaining or increasing soil C can be as high as 20–90% of total RAPS incomes (Fig. 2). The relationship between C losses and RAPS specialization can be compared to SOC losses in the 1980–1990s. According to Rodin and Krylatov (1998) in Mariy El, an annual SOC sink of 0.11 t C ha–1 in 1986–1990 turned to an C annual loss of 0.21 t C ha–1 in 1995, while in Mordovia during the same period, there was 2.5-fold increase in annual losses (–0.16 and –0.44 t C ha–1, respectively). This corresponds well with the highest percentage of row crops in Mordovia (more than a half of the area was in cropland in 2000), drastic decreases in organic and mineral fertilization during 1990–2000 and a subsequent 30–40% drop of cereal and grass yields in the two abovementioned regions. Besides, both regions have moderate pasture areas (3–4% from agricultural territories) and big areas of potatoes (7% in 1999 for Mariy El cropland; Shishov et al. 2001). For these unsustainable RAPS systems under different climate scenarios, change in costs of maintaining of increasing SOC can be as high as 15%, which is smaller than in the central regions. The highest costs are predicted for the scenario with greater climate forcing (A1FI), apart from in Kirov Region, where grain and grass yields are improved in future even without adaptation (BAU management). This is most notable in the northern part of the region, compared with negative changes without adaptation in the parts of the Upper Volga (Sirotenko 2005). When the difference in the percentage of perennial grasses in the rotation between SUS and OPT management scenarios is not too large (
