There is a growing interest in better understanding factors affecting agroecosystem functions and services. One key property of agroecosystems that has received much attention in recent years is the diversity of crop species. National crop species diversity has been associated with the stability of food production and used as a proxy for pollination services partly because local crop diversity enhances associated biodiversity . However, it is unclear how national crop diversity is related to local-scale diversity, and assuming that inferences made at one scale are maintained at other scales can be misleading. Crop species diversity also has a temporal dimension of great importance resulting from farmers’ crop rotations. These crop rotations can reduce pressure from pathogens, pests, and weeds and improve soil quality and yield stability . Yet, the temporal dimension of crop diversity is frequently ignored in most diversity assessments, mainly because it is hard to measure , and it is not clear how diversity in space and time relates. Therefore, a better understanding and treatment of the scale dependency of spatial crop species diversity and its relation with temporal diversity is needed to develop comprehensive theories of crop diversity effects on agroecosystem function. Most calls for more diverse farming systems do not consider regional differences that might limit farmers’ diversification opportunities or demand-side constraints . Moreover,wholesale grow bags most studies on crop species diversity used variation in diversity to explain other phenomena but there has been less progress in understanding factors that shape crop diversity .
Thus, to gain a quantitative understanding of diversity patterns and processes and improve comparisons between regions and countries, environmental and crop demand constraints to crop diversity must be identified. The first two chapters of this dissertation deal with some fundamental aspects of crop species diversity. The first chapter explores how crop diversity can be measured, depending on the dimension and scale considered and how these relate, using data for the conterminous United States. The second chapter examines which factors define and limit crop diversity. It outlines a framework for quantifying potential and attainable levels of crop species diversity, which is then applied at the global level. Crop species diversity in the USA increased during the first half of the 20th century, but it has gradually declined over the past 50 years in most of the country . This specialization, together with increasing use of inputs, allowed substantial yield gains, but it was also associated with important negative environmental impacts . Consequently, there is an interest in developing more diverse and sustainable cropping systems . Most studies on diversified cropping systems consider crop rotation diversification a key management practice to be developed and implemented . However, there have been no comprehensive efforts to analyze crop rotation diversity across the US, probably because it needs to be observed at a very high spatial resolution . The first chapter analyzes temporal and spatial crop species diversity patterns in the conterminous US and how they relate. It shows that crop rotation diversity is tightly associated with local diversity at a spatial resolution close to typical US farm size. It also shows that this diversity is lower for rotations that include major crops. The observable patterns in the spatial distribution of crop species and their subsequent diversity are realizations of underlying processes that need to be elucidated .
While drivers of current crop genetic diversity patterns have been studied , there is very little knowledge on processes shaping the diversity of crop species. However, some concepts from macroecology and bio geography might be applicable . For instance, one of the most established patterns in Ecology, latitudinal biodiversity gradients , can also be expected to regulate crop species diversity. But crop diversity depends on both natural and human-mediated processes. Thus, while some tropical regions might be suitable for many crop species , current diversity patterns are also affected by individual and structural factors shaping farming decisions and resulting in different specialization levels . The second chapter sets a theoretical framework of hierarchical levels of crop diversity that considers crop-specific environmental requirements and the demand for agricultural products. This framework is then used to analyze the environmental drivers of potential and attainable crop diversity and quantify diversity gaps. The results show that potential and attainable crop diversity are lower in temperate and continental areas than in tropical and coastal regions. Although current diversity follows these patterns to some extent, other processes also affect it, resulting in high spatial variability in diversity gaps. The third chapter of this dissertation is on a different topic. It contributes to a project to better understand the opportunities for improved management of acid soils in Africa. One of the initial steps of the project is developing a spatially-explicit analysis of the costs and benefits of liming in Africa. This analysis is founded on models for lime requirement estimation. However, the literature on lime requirement estimation methods is sparse and inconsistent, particularly for acid tropical soils. Thus, the third chapter focused on a comparison of lime requirement models. Acid tropical soils can have several problems affecting crop growth, such as aluminum and manganese toxicity and calcium and magnesium deficiencies .
These issues can be addressed by applying liming materials . The amount of agricultural lime required is often estimated with locally calibrated soil tests . Both soil testing and liming might be relatively cheap and easy to access for intensive commercial farmers, but that is not the case for most smallholder farmers in tropical developing countries . Lime application is relatively expensive in many tropical regions, and the experimental evidence on lime response is also limited. Furthermore, soil tests that work elsewhere cannot be assumed to work for these places and must be re-calibrated. Therefore, general models to estimate lime requirements from generally available soil data could be useful as a starting point in developing locally optimal liming recommendations and for strategic research on future lime use. The third chapter compares and evaluates different models for lime requirement estimation that can be used in acid tropical soils with readily available soil data and introduces an outperforming model developed based on past experiences and clear principles. It shows that there are important differences in model accuracy and prediction values and that liming estimates largely depend on the target soil chemical property of the model. Therefore, the most important soil acidity problems affecting crop yields must be identified to formulate liming recommendations in acid tropical soils. However,grow bags for gardening models for other acidity problems than aluminum toxicity still need to be developed.Low soil pH is associated with a high concentration of toxic elements in the soil solution, such as aluminum and manganese, and with low availability of phosphorus, calcium, and other plant nutrients . Soil acidity problems can be addressed with liming, the application of calcium or magnesium-rich materials that react as a base . Liming has been practiced for centuries , and its use is still expanding, particularly in tropical areas with acid soils. For example, it played a key role in the recent expansion of agriculture in the Brazilian Cerrado region . The amount of lime required to adjust soil acidity depends on the soil, the target crop, and the liming material used. In temperate regions, lime requirements are commonly estimated with locally calibrated quick tests using buffer solutions . These tests can be developed by comparing the buffer’s response to the soil with the soil response to lime in field or incubation studies or by slow titrations. Both the soil testing and the lime application may be a relatively small expense in intensively managed commercial farms, partly because lime is cheap and partly because the use of lime, when needed, increases the use efficiency of other inputs . Moreover, applying a bit more lime than needed means its benefits will last longer . Thus, blanket applications that err on the higher side are not very risky , so there is no need for a highly accurate determination of the amount of lime to apply. This situation is different for smallholder farmers in sub-Saharan Africa and other tropical regions . Lime may be relatively expensive, and its benefit may be relatively small if fertilizer use is low. Under these circumstances, it would be helpful to have accurate estimates of lime requirements. However, empirical evidence from these tropical regions is limited, and laboratory-based soil testing is often inaccessible. Furthermore, methods that depend on direct measurements of soil acidity in each field with buffer solutions cannot be assumed to work elsewhere and would have to be redeveloped. Models to estimate lime requirements from generally available soil data are needed for strategic research of potential lime use across tropical regions.
They can be particularly useful for subSaharan Africa, where the impact of soil acidity on crop productivity and nutrient-use efficiency is poorly understood . Lime requirement models could serve as a starting point to develop locally optimal liming recommendations for farmers and development practitioners and provide strategic information to national governments and the private sector on potential market sizes for lime for a region of interest. The latter is now possible thanks to the availability of high-resolution spatial products for most soil properties across the continent . Here we provide a comprehensive review of general lime requirement models for tropical acid soils that can be used with readily available soil data. We first introduce key concepts related to estimating lime requirements that have been a source of confusion and inconsistency. We then describe and discuss published lime requirement models for tropical soils and introduce a new model to estimate lime requirements. Finally, we show substantial differences in the estimated lime requirement for acid tropical soils when using these models and discuss their implications. Soils can be naturally acidic or become acidic because of agricultural practices such as the use of acidifying fertilizer and the removal of elements with harvested products. In the tropics, many soils in humid areas are inherently acidic because intense weathering processes have resulted in the displacement and leaching of basic exchangeable cations and the accumulation of exchangeable acidity . The main problem with soil acidity in the tropics is not the low pH as such, but rather the associated aluminum toxicity that constrains crop growth . The purpose of liming should be, therefore, to remove Al toxicity, considering the sensitivity of the target crops, together with other possible constraints such as Ca and Mg deficiencies , but not to increase pH for its own sake . Acidity saturation is the fraction of the effective cation exchange capacity of the soil occupied by exchangeable acid cations . In tropical soils, nearly all exchangeable acidity comprises exchangeable Al3+ and, thus, Al saturation approximates acidity saturation . For that reason, acidity saturation is often used as a proxy for Al toxicity . Many lime requirement models estimate the lime rate required to lower the acidity saturation to a target level that does not affect crop yield . The terms exchangeable acidity and exchangeable Al3+ have been used interchangeably in tropical soil literature, with the term exchangeable Al3+ more commonly used in older literature . Indeed, several authors of the lime requirement models reviewed here measured acidity saturation but referred to it as Al saturation . Consequently, some models were originally formulated for exchangeable Al3+ but derived from exchangeable acidity measurements. Some highly weathered acid soils can have very low ECEC and, thus, low exchangeable Ca2+ and Mg2+ but low acidity saturation, resulting in Ca and Mg deficiencies without Al toxicity problems . Therefore, some lime requirement models based on acidity saturation also estimate the lime rate needed to cover such deficiencies . Such mineral deficiencies can also be addressed with compost or inorganic fertilizers such as calcium nitrate, which might be more convenient in soils with no Al toxicity problems. An alternative approach to alleviating soil acidity problems aims to raise the “base saturation” to a certain level rather than focusing on acidity saturation . Base saturation is the sum of all exchangeable bases divided by the Cation Exchange Capacity at pH 7 . CEC7 is different from ECEC, especially in acid soils, where CEC7 ≫ ECEC. For ECEC, exchangeable acid cations are extracted with a neutral unbuffered salt. In contrast, a pH 7 buffer solution is used for CEC7, which extracts both exchangeable and non-exchangeable acidity , comprising the potential acidity.