The multiple ecosystemservices provided by healthy soil are well known and include soilcarbon sequestration to mitigate climate change, a medium for plant and agricultural production and regulating the hydrologic cycle. Despite the wide recognition of the importance of these services, drivers of soilorganiccarbon (SOC) dynamics across various land uses in East Africa are poorly understood. The objectives of this study were threefold: to quan- tify SOC stocks across Tanzania; assess the effect of landcover and erosion on SOC; and investigate the relation- ship between inherent and dynamic soil properties under diverse land uses. The Land Degradation Surveillance Framework (LDSF) was used to assess the variability of ecological metrics at different spatial scales. SOC was quantiﬁed within and between different landcover types (forest, woodland, shrubland, grassland and cropland) in Tanzania. A total of 2052 soil samples from 1082 –1000 m 2 plots were collected from seven 100-km 2 sentinel sites in 2010. Composite soil samples were collected at each plot from two depths (0–20 and 20–50 cm) and cu- mulative soil mass samples were collected to 100 cm. Soil samples were analyzed using a combination of tradi- tional analytical laboratory methods and mid-infrared spectroscopy (MIR). Model performance of MIR spectral predictions for carbon was good, with an R 2 of N0.95 and RMSEP of 4.3 g kg −1 , when using an independent validation datasets. Woodland and cropland were the most frequently occurring vegetation structure types in the sampled sites, with 388 and 246 plots, respectively. Average topsoil OC (and range) was 12.4 (1.5–81.4) g C kg −1 (n = 1082) and average subsoil OC (and range) was 7.3 (0.64–53.8) g C kg −1 (n = 970) for the seven sites. Forested plots had the highest mean topsoil organiccarbon concentrations (17.3 g C kg −1 ) followed by cropland (13.3 g C kg −1 ), for all sites included in the study, but with high levels of variability between sites. Soil mass at 30 cm was measured and these data were used to calculate carbon stocks for the different landcover types. An approach based on remote sensing was explored for the mapping of SOC stocks at 30 cm for Tanzania using Moderate Resolution Imaging Spectroradiometer (MODIS) imagery from 2012. Results indicate that the use of image reﬂectance for the mapping of SOC stocks has promising potential, with R 2 values ranging from 0.77 to 0.81 and RMSEP values from 0.90 to 1.03 kg m −2 for the three validation datasets. There is high utility of these maps for strategic land management interventions that prioritize ecosystemservices.
The processes of urban expansion and densi ﬁ cation both involve land-use changes that impact upon ecosystem function and service pro- vision (Gaston et al., 2010). Until recently, a major gap in current under- standing of urban ecosystems concerns the nature and property of their soils particularly in Western Europe (Lehmann and Stahr, 2007; Byrne, 2007; Ef ﬂ and and Pouyat, 1997), despite the critical role of soil in supporting most urban ecosystemservices. However, recent research investigating citywide SOC stocks, beneath both urban greenspaces and impervious surfaces, has demonstrated that urban areas are capa- ble storing much larger quantities of SOC than previously realised (Edmondson et al., 2012). Indeed, average storage across an entire U.K. city (assuming 0 kg SOC m −2 storage beneath buildings which cover 15% of the city) was 14.5 kg m −2 , to 1 m depth (Edmondson et al., 2012). These measurements contrast with the earlier estimates of SOC in suburban areas in England, which gave a mean value of 6 kg m −2 , based on the untested assumption that these areas store half the carbon of agricultural grasslands in the same region (Bradley et al., 2005). Furthermore, these new data reveal that urban SOC stocks are 12% higher than the mean storage value for agricultural grasslands and only 15% lower than the mean storage value for woodlands in the English national SOC inventory (Bradley et al., 2005), necessitating a radical revision of the widely held misconception of low ecological values of urban soils.
Wetland management and conservation have been subjects of research especially due to the pressure exerted on the wetlands with increasing population (Whigham et al. 1993, Rebelo et al. 2010, Mombo et al. 2011)(Rebelo et al. 2010). Emphasis on the need for development and implementation of management policies geared towards protection and regulation of wetlands providing ecosystemservices is on the rise (Millennium Development Goals and Nations Unies 2011, Sachs 2012, Leemhuis et al. 2017). However, for the sustainable development decisions, there is a need for data highlighting the overall interrelated functioning of the ecosystem. An important aspect of understanding an ecosystem is knowledge on the spatial location, extent, and dynamics of geophysical parameters. Complex nature of wetlands and high temporal variability favor the utilization of remote sensing in wetland assessments (Klemas 2011, Tiner et al. 2015). Vast extent and repeated coverage facilitate the continued monitoring of land parameters that influence wetland environments and their use (Guo et al. 2017). Water conservation and runoff regulation are among the wetland functions making wetlands a preferred ecosystem for increased food production (Chen and Wong 2016). Exploitation of the ecosystem potential for agriculture while neglecting the negative impacts of the natural resources degrades the functional capability of the wetland (Rockström and Karlberg 2010, Rockström et al. 2017, Rosa et al. 2017). Information on water availability to crops guides farmers on irrigation times whereas seasonal forecasts aid in the determination of planting and harvest times for maximum yield (Rosa et al. 2017). Historical study on the trends of geophysical parameters such as soil moisture could act as a plant distribution indicator geared towards plant species selection based on their tolerance to environmental stress (Li et al. 2013). Such decisions could include a shift in the cropping system and adaptation of climate resilient crops based on the soil water availability (Verhoeven and Setter 2010, Nicol et al. 2015).
the microhabitat heterogeneity, and adequately conserve its associated biodiversity (Tylianakis et al. 2005).
3.4.2 Herb diversity as affected by landscape position
The higher gamma diversity in coffee on upland soils compared to lowland soils may be related to farmers’ preferences to select areas with less steep slopes for pastures and rice compared to coffee agroforestry. On upland soils, coffee is grown on steeper slopes (mean 16 o , min 8 o and max slope 30 o ) than pastures (mean 13 o , min 5 o and max slope 25 o ) and rice (mean 10 o , min 5 o and max slope 15 o ). The contrast in slope and related drainage conditions between upland and lowland soils is therefore weaker for pasture and rice than for coffee. Moreover, severe compaction of pastures (as shown by bulk density) and erosion of rice fields (Schoorl et al. 2006) may have had such a strong influence on plant diversity that any relationship of plant diversity with topographic position in these land uses was masked. Meanwhile, in coffee agroforestry systems the lack of compaction and erosion may have preserved some of the original soil conditions and heterogeneity which still shows up in the relation with plant diversity. As was the case in our coffee agroforestry systems, the distribution of palms was also affected by landscape position (valley versus hill crests) in a Central Amazonian forest, which was attributed to differences in soil moisture and drainage conditions (Kahn and Castro 1995). Similar results were reported for understory herbs in an Ecuadorian rain forest (Poulsen and Balslev 1991).
We used spatial analysis to assess the Land Use LandCover (LULC) changes, and studied the impacts of LC changes on conservation of buffer zone of the Selous Game Reserve (SGR) and their implication on community’s livelihood in Vikumbulu Ward of Kisarawe District, Tanzania. Socio-economic data from Kisarawe District and TNBS were linked to spatial data to offer an inte- grated perspetive of LULC change in the Ward. Three cloud free image dates of 1998, 2011 and 2015 were analysed using System for Automated Geoscien- tific Analyses (SAGA) GIS for three categories of landcover, i.e. forest, wooded grassland and bare land/settlements/cultivation. Vikumbulu demo- graphic and socio-economic data were linked to spatial data applying distance as a function of LULC change. Spatial analysis has shown a decreasing trend of forest and woodland cover in Vikumbulu Ward between 1998 and 2015. The sharp decline indicates increasing social economic activities such as shifting agriculture and charcoal burning as an outcome of population growth and poverty. Rapid conversion of forest cover to wooded grassland occurred between 1998 and 2015 in Vikumbulu Ward. However, loss of forest cover was associated with a decreasing trend in wooded land in the ward between 2011 and 2015. As there was only 0.15% area under crop cultivation in Vi- kumbulu until 2015, it is highly likely that LC change is caused by charcoal burning and shifting cultivation. This study suggests developing integrated strategies that target LULC change, conservation and people’s livelihoods to effectively improve the current situation in rural areas of Tanzania.
From Figs. 3 and 4 and Table 9, we could get that the average ecosystem service value in the Nenjiang River Basin decreased gradually from the northwestern up- stream mountainous area to the southeastern down- stream plain. The higher valuable area are located in the Da and Xiao Xing ’ anling Ranges, such as Mohe, Hulun- beier, and Heihe city, while the lower valuable area are Baicheng, Songyuan, and Suihua city in the downstream plain. However, the Xing ’ an League, which also belongs to the Da Xing ’ anling mountainous area, is of a lower average ecosystem service value due to four great grass- land reclamation activities from the 1950s (Pan et al. 2002; Su et al. 2005). According to statistics, the area of farmland in the Xing ’ an League increased about 44.21 % from 1996 to 2000 and 6.63 % from 2001 to 2005 (Ying, 2009).
Types of land use, intensity of cultivation and fertilizer sources are major factors responsible for soil properties transformation. Some inherent soil properties such as soil pH and texture as well as climatic conditions might also have substantial influences on soil properties transforma- tions. Site 1 soil was lower in SOC and SON stocks than Sites 2 and 3; while Site 2 soil was lower in macro AF and AS than Sites 1 and 3. The later was also deficient in Cu, Fe and Zn due to its inherent properties and pH. Con- trastingly, Site 3 revealed higher SOC and SON stocks, but had limited P content. Interestingly, the variation in SOC and SON stocks, and nutrient concentrations be- tween the crop lands (Site 1 and Site 2) suggested that external fertilizer supply should take into account the indigenous nutrient supply potential of soil and the macro and micro nutrients balance for sustainable crop production. Generally the studied soils showed medium to high CEC and high percent base saturation (PBS), reflecting fairly good fertility. In nutshell, these findings are useful as a baseline for future intervention and sus- tainable utilization of the Andosols of the rift valley of Ethiopia. Furthermore, detailed studies of soil mineral- ogy and soil fertility based on use of test crops with addi- tion and omission of nutrients are suggested.
expected low organic C content of glacial and ﬂ uvioglacial strati- ﬁ ed sediments, based on the reported DOC values for selected UK lithologies — e.g. sands and gravels; Carboniferous limestone (UK Environmental Agency data reported in Buss et al., 2005). Lower soilorganiccarbon content is also known to be more generally typ- ical for long term arable cropping systems than for grassland or forest soils (e.g. Chantigny, 2003; Gregorich et al., 1995). The lon- ger the duration of arable cropping, the greater is the decrease in the soil water extractable and labile organic C content (Haynes, 2000; Saviozzi et al., 1994). Soil C is known to be present in subsoil horizons at low concentration; however according to a review by Schmidt et al. (2011), this C can be still important because it repre- sents > 1/2 of the global soil C stocks (Jobbagy and Jackson, 2000). Overall mean DOC concentrations were 1.78 mg/l under M, 1.38 mg/l under NC and 1.35 mg/l under NR. Mean groundwater DOC concentra- tion was signi ﬁ cantly higher under M treatment than under NC and NR (p b 0.05; Table 3). Fig. 3b shows higher DOC concentrations observed under M during the over-winter growth period compared to the other two treatments. The effects of the treatments become more pronounced and obvious after the winter cover growth period when DOC concentra- tions under M exceeded those under NC and NR (between January and March/April of 2008 and 2009; Fig. 3b).
The summary statistics on the areal extent of each class in the NLCD data are reproduced in Table 4. As previously these estimates were derived directly from the map and make no account for the effect of misclassification error. Although this is a relatively accurate map, with an estimated accuracy of 84% (Wickham et al., 2013; for the map at Anderson level I), the effect of misclassification error can still be large and was explored. For this, error-adjusted estimates of class extent are required. The latter are provided by Wickham et al. (2013) and also reproduced in Table 4. Estimates of the value of ecosystemservices for the conterminous USA were obtained using the original mapped areal extents and the misclassification error-adjusted estimates of class area. It was apparent that by taking the map at face value, the estimate of ecosystemservices value for the conterminous USA was ~US$1118 billion yr -1 . This is substantially larger than the US$773 billion yr -1 reported by Konarska et al. (2002) based on the earlier NLCD map for 1992. Given that the monetary value per-unit area associated with each class was constant, this outcome could indicate a substantial increase in the value of ecosystemservices and might perhaps be interpreted as reflecting the results of successful policy outcomes or some other change that acted to increase the extent of wetlands. Alternatively, the difference in the valuations may actually be of uncertain meaning because the estimates themselves have not been corrected for misclassification error. The latter is possible for the 2006 map because of the rigorous accuracy assessment undertaken
This model was successfully used in reconstructing the 1970–2010 Slovak crop and grass- lands topsoil SOC stock development trajectory and also for estimating the current SOC lev- els on a national scale (Barančíková et al., 2010, 2012, 2013). However, these authors record that the results provide only approximate SOC stock estimates because of limitations in the spatial resolution of gridded data (10x10 km) on organiccarbon inputs from management, the monthly weather records and the initial 1970’s SOC stock estimated from the soil map and profile data from the Slovak National agricultural soils inventory used to run the RothC model. This research also neglects marginal cropland and grassland areas and does not con- sider landcover change during the simulation period (Barančíková et al., 2012, 2013).
LIST OF TABLES
Table 1 Summary of management practices used for the GEMS model parameterization. The crop rotation probabilities should be read horizontally from time 1 to time 2; each row sums to 1 .................................................................................................................... 57 Table 2 Description of the 9 landcoverland use (LCLU) classes and the number of training pixels used for the classification.......................................................................... 60 Table 3 Soft-to-hard confusion matrix results for the 9 landcoverland use classes. The cell values report percentages of the total area; a total of 305 428 pixels were considered. The percent correct is 97.79% and Kappa-coefficient is 0.98. Grey fields, along the diagonal, represent for each class, the percentage correctly classified. The classes are: 1. Plantation; 2. Water; 3. Bare soil; 4. Rainfed agriculture; 5. Wetlands; 6. Mangrove; 7. Mud flats; 8. Irrigated agriculture; 9. Savanna grassland (Table 2). ................................ 67 Table 4 Comparison of the minimum, mean and maximum SOC (Figure 3) and NPP (Figure 4) simulated for the 9 LCLU classes using the year 2000 hard classification (Figure 1). Only pixels where SOC and NPP was modeled are considered (i.e., not water bodies, clouds, cloud shadows, settlement areas, or where there was no Landsat data). . 72 Table 5 Comparison by agro-ecological zone of the minimum, mean and maximum SOC (gC/m 2 ) (Fig. 3) for the 9 LCLU classes using the year 2000 hard classification (Fig. 1). The LCUC percentage area in each zone is shown in parentheses. Only pixels where SOC was modeled are considered (i.e., not water bodies, clouds, cloud shadows, settlement areas, or where there was no Landsat data). ..................................................................... 74 Table 6 Summary statistics of the mean of the 30 soft decision tree SOC estimates for year 2000. The statistics are summarized with respect to the 9 LCLU classes defined by the hard decision tree classification (Figure 1). The mean study area mean SOC is 1217.4 gC/m 2 . Only pixels where SOC was modeled are considered (i.e., not water bodies, clouds, cloud shadows, settlement areas, or where there was no Landsat data). .............. 75 Table 7 Summary statistics of the mean study area hard and soft decision tree (DT) soilorganiccarbon (SOC) (gC/m 2 ) model estimates illustrated in Figures 7 and 8, for the no, low and high climate change scenarios, for selected years ............................................... 82 Chapter 3:
Stevens, François ; Bogaert, Patrick ; van Wesemael, Bas
The spatial distribution of soil properties often displays complex and multiscale patterns of variation. It results from multiple soil processes acting simultaneously but at different scales. Hence, characterizing the influence of a given controlling factor on the soil property is made more difficult by the variation due to other controlling factors. In this context, separating the variation of the soil properties by spatial scales could allow disentangling the combined effect of controlling factors and would provide a qualitative and quantitative characterization of controlling factors separately. In this paper, geostatistical tools have been used to separate the scales of variation of two soil properties (i.e. SOC and texture) coming from a legacy dataset in the Belgian Loess Belt. Scale components were predicted separately and the relationships between soil properties were analyzed at different scales. Results illustrated that the contents of a given soil property in different dep...
One of the consequences of burial and redistribution is the increase in spatial variability in SOC concentration in the surface layer (Chaplot et al., 2009) and the attendant impacts on biogeochemical and microbial properties (Park et al., 2014). Erosion-induced transport of SOC over the eroding landscape can accentuate gaseous flux (Page et al., 2004; Smith et al., 2007) with an attendant loss of SOC pool (Boix-Fayos et al., 2009); Glendaell and Brazier, 2014; Hancock et al., 2010; Liang et al., 2009; Korup and Rixen, 2014; Martinez-Mena et al., 2008; Quinton et al., 2006; Yan et al., 2005). Thus, it is argued by pedologist and agronomist, that landscapes subjected to accelerated soil erosion are a major source of GHGs (Fig. 2; Table 2; Chappell et al., 2013; Lal, 2013; Lal and Pimentel, 2008) with a strong adverse impact on the C cycle (Ito, 2007; Lal, 2003; Kuhn et al., 2009) and ecosystem C budget (Gao, 2007; Lal, 2003).
et al. 2007; Ritter 2007). Moreover, the recovery rate of SOC
is greater at the early stage after a land use change than during the later stages (Coleman et al. 1997).
City suburb is an urban-rural transition zone. The rapid progression of the urbanization process in recent years has resulted in constant increases in the degree of agricultural intensification and land use intensity. In particular, the progression of urbanization has resulted in relatively large changes in land use types and management practices for agricultural lands, which will have important impacts on SOC content (Fu et al. 1999). The Pinggu District, located in a northeastern suburb of Beijing, China, has evolved from an agricultural area that mainly focused on grain production to an important fruit and vegetable production base for Beijing. Thus, the Pinggu District is highly representative of an urban-rural ecotone. Hence, an evaluation of the effect of land use change on the spatiotemporal variation in organiccarbon in this area is of utmost significance for implementing urban low-carbon and sustainable agriculture.
(Schulp & Verburg 2009; van Wesemael et al. 2010). Negative trends in the SOC stock balance can be treated politically via various measures taken in crop and soil management practices (e.g. Common Agricultural Policy, EC 2006b). Information on soil quality is essential for any decision-making on crop and soil management (EC 2006b). In Slovakia, the soil quality for decision-making needs is expressed by so called “soil production potential index” which is a complex functional characteristic of landscape attributed to soil-landscape units (land-evaluation units) delineated on large-scale maps based on climate, topography and soil characteristics in- cluding soil type, soil texture, stoniness and soil depth but not SOC content itself (Džatko 1981; Linkeš et al. 1996). A practical method how to acquire data on the SOC stock changes both in space and time for political decision-making is to employ process-based simulation models such as e.g. RothC (Smith et al. 2005, 2007; Milne et al. 2007; van Wesemael et al. 2010, etc.). RothC model was successfully applied also for the condi- tions of Slovakia. Data infrastructure was built up to run the model over the agricultural land of Slovakia and the validation of modelling results was done on several levels (Barančíková 2007; Barančíková et al. 2010a, b).
with volumetric cores for determining bulk density (BD). Core-collected soil samples were oven dried (105 °C) to constant weight prior to calculating their BDs (Blake and Hartge 1986). Particle size was determined by the pipette method (Gee and Bauder 1986). Soil pH was measured in deionized water (water–soil ratio 1:1) by glass electrode (McLean 1982). Soil samples were finely ground using a ball grinder (Oscillating Mill MM400; Retsch, Newtown, PA, USA) before SOC and N analysis. A CHN analyzer Fig. 1 Location of the study area and sites, Xitou tract of Experimental Forest, National Taiwan University, Nantou County, Taiwan (TP tea plantation, BM Bamboo forest, JC Japanese Cedar, TW Taiwania)
conversion. Unless these metrics are applied at an appropriate frequency, they will not provide a dynamic 421
response. In contrast, we have shown that by replacing landcover with u S* specific for wind erosion, and then
The use of gypsum in combination with other treatments has been found to improve overall soil properties to a greater extent than the use of gypsum on its own. For example, where lime and gypsum were combined in the amelioration of a sodic red- brown earth (pH <6.5), Valzano et al. (2001b) found higher levels of plant growth coupled with significant increases in total C in the soil over a period of three years. It was found that whilst gypsum was more effective than lime in displacing exchangeable and soluble Na, a combination of the two was more efficient at maintaining soil electrolyte levels and improving soil physical and hydraulic properties. This was due to the different solubilities of the two amendments, as gypsum could provide Ca 2+ during the early stages of remediation due to its higher solubility, enhancing soil physical properties to allow greater throughflow of water into the soil, which would, in turn, allow for greater dissolution of lime in the later stages. Similarly, when gypsum was used as an ameliorant in conjunction with stubble retention and appropriate crop rotations, Valzano et al. (2001a) found interactions between all treatments which aided the improvement of soil properties. Gypsum addition decreased soluble and exchangeable Na + concentrations, improving structural stability and hence, improved soil water relations. This results in higher crop yields, which build up SOC levels, thereby further improving soil structure. When stubble is burnt, macroporosity is reduced due to lower levels of biological activity and a reduction of throughflow of chemical amendment. The retention of stubble provides surface protection and prevents crust formation due to protection from raindrop impact. This improves infiltration while the use of leguminous crops may facilitate the leaching of gypsum through the soil profile, remediating soil properties at depth.