Fluxes of methane (CH₄) and carbon dioxide (CO₂) estimated by empirical models based on small-scale chamber measurements were compared to large-scale eddy covariance (EC) measurements for CH₄ and to ...a combination of EC measurements and EC-based models for CO₂. The experimental area was a flat peat meadow in the Netherlands with heterogeneous source strengths for both greenhouse gases. Two scenarios were used to assess the importance of stratifying the landscape into landscape elements before up-scaling the fluxes measured by chambers to landscape scale: one took the main landscape elements into account (field, ditch edge ditch), the other took only the field into account. Non-linear regression models were used to up-scale the chamber measurements to field emission estimates. EC CO₂ respiration consisted of measured night time EC fluxes and modeled day time fluxes using the Arrhenius model. EC CH₄ flux estimate was based on daily averages and the remaining data gaps were filled by linear interpolation. The EC and chamber-based estimates agreed well when the three landscape elements were taken into account with 16.5% and 13.0% difference for CO₂ respiration and CH₄, respectively. However, both methods differed 31.0% and 55.1% for CO₂ respiration and CH₄ when only field emissions were taken into account when up-scaling chamber measurements to landscape scale. This emphasizes the importance of stratifying the landscape into landscape elements. The conclusion is that small-scale chamber measurements can be used to estimate fluxes of CO₂ and CH₄ at landscape scale if fluxes are scaled by different landscape elements.
Strong gradients of decreasing soil fertility are found with increasing distance from the homestead within smallholder African farms, due to differential resource allocation. As nutrient use ...efficiency varies strongly along these gradients, such heterogeneity must be considered when designing soil management strategies, aimed at an improved overall resource use efficiency at farm scale. Here, we quantify the magnitude and study the origin of farmer-induced, within-farm soil fertility gradients as affected by biophysical and socio-economic conditions, and investigate farmers' perceptions of such heterogeneity. Farm transects, participatory resource flow mapping, farmers' classification of land qualities, and soil sampling for both chemical and spectral reflectance analyses were performed across 60 farms in three sub-locations (Emuhaia, Shinyalu, Aludeka) representing the variability found in the highlands of western Kenya. Differences between the various field types of a farm were observed for input use (e.g. 0.7-104 kg N ha(-1)), food production (e.g. 0.6-2.9 t DM ha(-1)), partial C (e.g. -570 to 1480 kg ha(-1)) and N (e.g. -92 to 57 kg ha(-1)) balances and general soil fertility status, despite strong differences across sub-locations. Concentration of nutrients in the home fields compared with the remote fields were verified for extractable P (e.g. 2.1-19.8 mg kg(-1)) and secondarily for exchangeable K (e.g. 0.14-0.54 cmol(+) kg(-1)), on average, whereas differences for soil C and N were only important when considering each individual farm separately. Farmers managed their fields according to their perceived land quality, varying the timing and intensity of management practices along soil fertility gradients. Fields classified by them as poor were planted later (up to 33.6 days of delay), with sparser crops (ca. 30% less plants m(-2)) and had higher weed infestation levels than those classified as fertile, leading to important differences in maize yield (e.g. 0.9 versus 2.4 t ha(-1)). The internal heterogeneity in resource allocation varied also between farms of different social classes, according to their objectives and factor constraints. Additionally, the interaction of sub-location-specific socio-economic (population, markets) and biophysical factors (soilscape variability) determined the patterns of resource allocation to different activities. Such interactions need to be considered for the characterisation of farming system to facilitate targeting research and development interventions to address the problem of poor soil fertility.
Shallow fresh water bodies in peat areas are important contributors to greenhouse gas fluxes to the atmosphere. In this study we determined the magnitude of CH₄ and CO₂ fluxes from 12 water bodies in ...Dutch wetlands during the summer season and studied the factors that might regulate emissions of CH₄ and CO₂ from these lakes and ditches. The lakes and ditches acted as CO₂ and CH₄ sources of emissions to the atmosphere; the fluxes from the ditches were significantly larger than the fluxes from the lakes. The mean greenhouse gas flux from ditches and lakes amounted to 129.1 ± 8.2 (mean ± SE) and 61.5 ± 7.1 mg m⁻² h⁻¹ for CO₂ and 33.7 ± 9.3 and 3.9 ± 1.6 mg mg m⁻² h⁻¹ for CH4, respectively. In most water bodies CH4 was the dominant greenhouse gas in terms of warming potential. Trophic status of the water and the sediment was an important factor regulating emissions. By using multiple linear regression 87% of the variation in CH₄ could be explained by PO₄³⁻ concentration in the sediment and Fe²⁺ concentration in the water, and 89% of the CO₂ flux could be explained by depth, EC and pH of the water. Decreasing the nutrient loads and input of organic substrates to ditches and lakes by for example reducing application of fertilizers and manure within the catchments and decreasing upward seepage of nutrient rich water from the surrounding area will likely reduce summer emissions of CO₂ and CH₄ from these water bodies.
The processes of nutrient depletion and soil degradation that limit productivity of smallholder African farms are spatially heterogeneous. Causes of variability in soil fertility management at ...different scales of analysis are both biophysical and socio-economic. Such heterogeneity is categorised in this study, which quantifies its impact on nutrient flows and soil fertility status at region and farm scales, as a first step in identifying spatial and temporal niches for targeting of soil fertility management strategies and technologies. Transects for soil profile observation, participatory rural appraisal techniques and classical soil sampling and chemical analysis were sampled across 60 farms in three sub-locations (Emuhaia, Shinyalu, Aludeka), which together represent much of the variability found in the highlands of western Kenya. Five representative farm types were identified using socio-economic information and considering production activities, household objectives and the main constraints faced by farmers. Soil fertility management and nutrient resource flows were studied for each farm type and related to differences in soil fertility status at farm scale. Farm types 1 and 2 were the wealthiest; the former relied on off-farm income and farmed small pieces of land (0.6-1.1 ha) while the latter farmed relatively large land areas (1.6-3.8 ha) mainly with cash crops. The poorest farm type 5 also farmed small pieces of land (0.4-1.0 ha) but relied on low wages derived from working for wealthier farmers. Both farm types 1 and 5 relied on off-farm earnings and sold the least amounts of farm produce to the market, though the magnitude of their cash, labour and nutrient flows was contrasting. Farms of types 3 and 4 were intermediate in size and wealth, and represented different crop production strategies for self-consumption and the market. Average grain yields fluctuated around 1 t ha(-1) year(-1) for all farm types and sub-locations. Grain production by farms of types 4 and 5 was much below annual family requirements, estimated at 170 kg person(-1) year(-1). Household wealth and production orientation affected the pattern of resource flow at farm scale. In the land-constrained farms of type 1, mineral fertilisers were often used more intensively (ca. 50 kg ha(-1)), though with varying application rates (14-92 kg ha(-1)). The use of animal manure in such small farms (e.g. 2.2 t year(-1)) represented intensities of use of up to 8 t ha(-1), and a net accumulation of C and macronutrients brought into the farm by livestock. In farms of type 5, intensities of use of mineral and organic fertilisers ranged between 0-12 kg ha(-1) and 0-0.5 t ha(-1), respectively. A consistent trend of decreasing input use from farm types 1-5 was generally observed, but nutrient resources and land management practices (e.g. fallow) differed enormously between sub-locations. Inputs of nutrients were almost nil in Aludeka farms. Both inherent soil properties and management explained the variability found in soil fertility status. Texture explained the variation observed in soil C and related total N between sub-locations, whereas P availability varied mainly between farm types as affected by input use.
Policy decisions are often taken at the regional scale, while crop models, supporting these decisions, have been developed for individual locations, expecting location-specific, spatially homogeneous ...input data. Crop models are able to account for the variation in climatic conditions and management activities and their effects on crop productivity. However, regional applications require consideration of spatial variability in these factors. Several studies have analyzed effects of using spatially aggregated climate data on model outcomes. The effects of spatially aggregated sowing dates on simulations of crop phenological development have not been studied, however. We investigated the impact of spatial aggregation of sowing dates and temperatures on the simulated occurrence of ear emergence and physiological maturity of winter wheat in Germany, using the phenological model of AFRCWHEAT2.
We observed time ranges for crop emergence exceeding 90
d, whereas for harvesting this was more than 75
d. Spatial aggregation to 100
km
×
100
km reduced this range to less than 30 and 20
d for emergence and harvest, respectively. Differences among selected regions were relatively small, suggesting that non-climatic factors largely determined the spatial variability in sowing dates and consecutive phenological stages. Application of the AFRCWHEAT2 phenology model using location-specific weather data and emergence dates, and an identical crop parameter set across Germany gave similar deviations in all studied regions, suggesting that varietal differences were less important among regions than within regions. Importantly, bias in model outcomes as a result of using aggregated input data was small. Increase in resolution from 100
km to 50
km resulted in slight improvements in the simulations. We conclude that using spatially aggregated weather data and emergence dates to simulate the length of the growing season for winter wheat in Germany is justified for grid cells with a maximum area of 100
km
×
100
km and for the model considered here. As spatial variability of sowing dates within a region or country can be large, it is important to obtain information about the representative sowing date for the region.
LINTUL3 is a crop model that calculates biomass production based on intercepted photosynthetically active radiation (PAR) and light use efficiency (LUE). It is an adapted version of LINTUL2 (that ...simulates potential and water-limited crop growth), including nitrogen limitation. Nitrogen stress in the model is defined through the nitrogen nutrition index (NNI): the ratio of actual nitrogen concentration and critical nitrogen concentration in the plant. The effect of nitrogen stress on crop growth is tested in the model either through a reduction in LUE or leaf area (LA) or a combination of these two and further evaluated with independent datasets. However, water limitation is not considered in the present study as the crop is paddy rice. This paper describes the model for the case of rice, test the hypotheses of N stress on crop growth and details of model calibration and testing using independent data sets of nitrogen treatments (with fertilizer rates of 0–400
kg
N
ha
−1) under varying environmental conditions in Asia. Results of calibration and testing are compared graphically, through Root Mean Square Deviation (RMSD), and by Average Absolute Deviation (AAD). Overall average absolute deviation values for calibration and testing of total aboveground biomass show less than 26% mean deviation from the observations though the values for individual experiments show a higher deviation up to 41%. In general, the model responded well to nitrogen stress in all the treatments without fertilizer application as observed, but between fertilized treatments the response was varying.
Crop simulation models are widely applied at large scale for climate change impact assessment or integrated assessment studies. However, often a mismatch exists between data availability and the ...level of detail in the model used. Good modelling practice dictates to keep models as simple as possible, but enough detail should be incorporated to capture the major processes that determine the system's behaviour. The objective of this study was to investigate the effect of the level of detail incorporated in process-based crop growth models on simulated potential yields under a wide range of climatic conditions. We conducted a multi-site analysis and identified that by using a constant radiation use efficiency (
RUE) value under a wide range of climatic conditions, the description of the process of biomass production may be over-simplified, as the effects of high temperatures and high radiation intensities on this parameter are ignored. Further, we found that particular attention should be given to the choice of the light interception approach in a crop model as determined by leaf area index (
LAI) dynamics. The two
LAI dynamics approaches considered in this study gave different simulated yields irrespective of the characteristics of the location and the light interception approaches better explained the differences in yield sensitivity to climatic variability than the biomass production approaches. Further analysis showed that differences between the two
LAI dynamics approaches for simulated yields were mainly due to different representations of leaf senescence in both approaches. We concluded that a better understanding and modelling of leaf senescence, particularly its onset, is needed to reduce model uncertainty in yield simulations.
High extraction of phosphate reserves and low phosphorus utilization efficiency in the food chain in China result in large P losses and serious environmental pollution. The P fertilizer industry, ...soil P surplus, livestock manure P and wastewater P recycling have been identified as the priority sectors based on summarizing several systemic and in-depth reviews of P flows analysis. Mineral P fertilizer production has reached 7.4 Mt P in 2012, which is more than seven times the value in 1980. The large P surpluses in arable land resulted in soil P accumulation of up to 64 Mt during the period 1951–2010. Livestock numbers have increased dramatically (more than ten times) during the period 1949–2012 in China, especially pigs and poultry, and so has the quantity of manure that they produce. The average loading of manure P on arable land in China has increased significantly from 9.5 kg P ha⁻¹ in 1980 to 20.4 kg P ha⁻¹ in 2010. Up to 0.49 Mt of wastewater P discharged without treatment also exerted great pressure on the environment in 2012. Based on an understanding of P interactions in these key sectors, an integrated set of policy options and technical measures is proposed. Taking P flows in China in 2010 as an example, if all of the strategies recommended in this study are adopted in P management, about 4.3, 2.5, 1.6 and 0.3 Mt of P resources, respectively, will be saved in the P fertilizer industry, arable land production, livestock manure and wastewater sectors.
Since the pioneering work of C.T. de Wit in the 1960s, the Wageningen group has built a tradition in developing and applying crop models. Rather than focusing on a few models, diversity is its ...trademark. Here we present an overview of the Wageningen crop and crop-soil modelling approaches along three criteria. The first criterion relates to the production situations the models are dealing with (i.e. potential, water and/or nutrient-limited, and actual production situations including pests, diseases and weeds). Second, models differ as a result of the objectives of model development, and hence required scale and degree of detail and comprehensiveness. Third, models have at least three potential application domains, i.e. research, education and support of learning and decision making processes.
We describe both summary and more comprehensive modelling approaches for the major production situations. An overview of most of the Wageningen models is presented together with a more detailed description of LINTUL, SUCROS, ORYZA, WOFOST and INTERCOM. Illustrations for each of the three application domains are presented, i.e. plant type design, guiding experimental research, education, yield gap analysis, evaluation of manure policies, crop growth monitoring system and analysis and design of farming and regional land use systems. We discuss common issues of model verification, model validation, model validity and data requirements, and present information on software implementation, model and software documentation and distribution policy. Finally, we reflect upon the Wageningen modelling approaches and identify a number of key issues for future research.
Major achievements of Wageningen modelling efforts include (1) a broad variety of approaches for modelling of systems at different scales and with different purposes; (2) their contribution to quantitative systems thinking in general, also for applications at higher hierarchical levels; (3) a strong linkage between crop modelling and higher education, both at undergraduate, graduate and post-doctoral level. To continually increase our understanding of crops and production systems a diversified approach must be cherished. At the same time we conclude that focus is required on a limited number of modules in a more integrated modelling framework for the benefit of analysing, evaluating and designing cropping systems. This review may be instrumental in the development of such an integrated framework.
► Linear interpolation can generate daily weather data from aggregated weather data.
► Simulations using interpolated data will deviate from those with actual data.
► Increasing day-to-day ...variability in weather conditions gives stronger deviations.
► Increasing detail in a model structure results in stronger deviations.
Weather data are essential inputs for crop growth models, which are primarily developed for field level applications using site-specific daily weather data. Daily weather data are often not available, especially when models are applied to large regions and/or for future projections. It is possible to generate daily weather data from aggregated weather data, such as average monthly weather data, e.g. through a linear interpolation method. But, due to the nonlinearity of many weather–crop relationships, results of simulations using linearly interpolated data will deviate from those with actual (daily) data. The objective of this study was to analyse the sensitivity of different modelling approaches to the temporal resolution of weather input data. We used spring wheat as an example and considered three combinations of summarized and detailed approaches to model leaf area index development and associated radiation interception and biomass productivity, reflecting the typical range of detail in the structure of most models. Models were run with actual weather data and with aggregated weather data from which day-to-day variation had been removed by linear interpolation between monthly averages.
Results from different climatic regions in Europe show that simulated biomass differs between model simulations using actual or aggregated temperature and/or radiation data. In addition, we find a relationship between the sensitivity of an approach to interpolation of input data and the degree of detail in that modelling approach: increasing detail results in higher sensitivity. Moreover, the magnitude of the day-to-day variability in weather conditions affects the results: increasing variability results in stronger differences between model results. Our results have implications for the choice of a specific approach to model a certain process depending on the available temporal resolution of input data.
PDF
