Yield response of potato to spatially patterned nitrogen application

Although crop response to nitrogen fertilization has long been studied, classical experimental designs have led to inadequate accounting of spatial variability in field properties and yield response. Analytical methods to explicitly account for spatial variability now exist but the complementary modification of experimental design is still developing. There is a need to combine these analytical methods with non-traditional experimental design. A 2-year study was implemented to assess the response of potato (Solanum tuberosum cv. Kennebec) yield to nitrogen fertilizer rate. We used a transect-type plot design where four nitrogen treatments (0, 56, 112, and 280 kg N ha^?1) were applied systematically in a continuous sinusoidal pattern along longitudinal transects. Measured field properties included topography, soil texture, pre-application soil nitrate levels, and plant available soil water content. A random field linear model was used to simultaneously account for treatment effects and soil properties. The results showed that treatment effects were significantly different from each other; however, if spatially correlated errors were accounted for, these differences were smaller and significance levels lower. Nitrogen response functions varied widely throughout the field. Of the covariates, only clay content proved important in explaining spatial differences in response to N. The sinusoidal response pattern of N was similar over the 2 years but the amplitudes varied due to differences in weather. Interactions between uncharacteristically high rainfall and a sandy field soil may have minimized discernable effects of the other covariates. The results demonstrated how the statistical analysis of potato response to a patterned application of nitrogen fertilizer can take advantage of spatial correlations to understand the response of potato to nitrogen application over larger areas.     

Future risk of flooding: an analysis of changes in potential loss of life in South Holland (The Netherlands)

Potential loss of life is considered an important indicator of flood risk. We examine the future development of potential loss of life due to flooding for a major flood prone area in The Netherlands. The analysis is based on projections and spatial distribution of population under a high economic growth scenario and a loss of life model. Results show that the projected population growth in flood prone areas is higher than average in the Netherlands between 2000 and 2040. Due to this effect the potential number of fatalities is projected to increase by 68% on average for 10 different flood scenarios, not including impacts from climate change and sea level rise. Just sea level rise of 0.30 m leads to an average 20% increase in the number of fatalities. The combined impact of sea level rise and population growth leads to an estimated doubling in the potential number of fatalities. Taking into account increasing probability of flooding due to sea level rise and extreme river discharges, the expected number of fatalities could quadruple by 2040. The presented results give a conservative and upper bound estimate of the increase of the risk level when no preventive measures are undertaken. It is found that the consideration of the exact spatial distribution of population growth is essential for arriving at reliable estimates of future risk of flooding.     

Carbon accumulation through ecosystem recovery

A market has emerged for carbon sequestered through reforestation. The opportunity to restore ecosystems through this market rather than establish plantations is demonstrated by an Australian case study. In the state of Queensland there are vast areas that have been cleared relatively recently and could be restored to ecosystems with high resilience and important biodiversity values with appropriate management. In order to foster opportunities for carbon accumulation through ecosystem recovery spatially explicit information on sequestration rates, management recommendations, and clear definitions of ancillary biodiversity benefits need to be defined.     

Estimating carbon in savanna ecosystems: rational distribution of effort

Minimising the cost of repeatedly estimating C (C) stocks is crucial to the financial viability of projects that seek to sell C credits. Depending on the price of C, this may imply less or more sampling effort than would be applied for science objectives. In systems with heterogeneous C pools, such as savannas, this translates into a variable-effort sampling strategy that maximises the marginal additional C that can be claimed per incremental unit of effort expended. Analysis of a savanna in north-eastern South Africa indicates relatively modest returns per hectare due to the small C quantities and low sequestration rates. Under these conditions, areas in excess of 1,000 ha and infrequent sampling frequencies of 5–10 years are required to make such projects financially viable. For such projects the sample variance, number of samples, cost per sample and establishment costs have negligible impacts on financial viability. It was also found that the soil-C pool contributes up to three times the net returns of the aboveground C pool and provides a strong argument to monitor soil C for certification and market trading. The financial viability estimates, however, do not include the management or opportunity costs incurred in changing the land use. The economies of scale identified in this study combined with the massive area covered by savannas indicate that these additional costs can be covered. Further research is recommended to quantify these costs and interrogate the feasibility of large scale (in excess of 10,000 ha) C-sink projects in savanna systems.     

A database for integrated assessment of European agricultural systems

A major bottleneck for data-based policy making is that data sources are collected, managed, and distributed by different institutions, residing in different locations, resulting in conceptual and practical problems. The use of dispersed data for agricultural systems research requires the integration of data sources, which means to ensure consistency in data interpretations, units, spatial and temporal scales, to respect legal regulations of privacy, ownership and copyright, and to enable easy dissemination of data. This paper describes the SEAMLESS integrated database on European agricultural systems. It contains data on cropping patterns, production, farm structural data, soil and climate conditions, current agricultural management and policy information. To arrive at one integrated database, a shared ontology was developed according to a collaborative process, which facilitates interdisciplinary research. The paper details this process, which can be re-used in other research projects for integrating data sources.