Soil and agricultural management practices (AMP) that are able to provide for an increasing population while meeting environmental existential challenges have gained considerable attention in recent times. Such AMP influence the soil profile and hydrological components for varying depths and patterns, depending on site-specific and environmental conditions. Though it is well known that management-induced changes of soil structure have consequences on soil hydraulic properties (SHP) and water fluxes, their dynamics through a season or on a long-term basis are hardly studied. Typically, an invariant soil pore system is assumed when modeling the transport of water and solutes in the soil system which leads to incorrect predictions of the dynamics of water balance components. Ultimately, this may lead to poor decision making and mismanagement of environmental resources. Hence, the present study quantifies the dynamics of SHP from existing studies and evaluates a model that is able to capture soil pore space dynamics following tillage. The objectives were to (1) investigate the quantitative effects of agricultural practices on soil structure and hydraulic properties and the subsequent response of the water balance components (2) evaluate a pore space evolution model for its capability in predicting the evolution of soil pore size distribution (PSD) for two cases: a) when there is a change in the tillage regime and/or land-use change b) in the months following tillage (3) derive corresponding soil water retention and hydraulic conductivity functions to incorporate them in hydrological models To achieve these objectives, first, a review of contemporary literature was undertaken to analyze the impacts of anthropogenic and environmental influences on SHP. The analysis indicated the relevance of studying temporal alterations of soil structure and SHP. Thereafter, a numerical model was evaluated for its ability to capture the dynamics of soil pore space with respect to time and pore radius using water retention parameter data sets from different parts of the world. The physically based coefficients of the model simulated the processes that were expected to occur after tillage. Furthermore, saturated hydraulic conductivity was obtained from the initial and final pore size distributions. Using the final pore size distribution curve and water retention function, the hydraulic conductivity function was also derived. The resulting water retention and hydraulic conductivity curves can directly be used as input in hydrological modeling studies. The results of the literature review indicate that, generally, soils show an abundance of large pores immediately after tillage. Those pores are not stable with time mainly due to precipitation and biological activity. Saturated hydraulic conductivity decreases in periods of rainfall along with the number of macropores and the overall porosity. Thus, the infiltration rates and capacities also decrease. However, the results of existing studies cannot be generalized owing to discrepancies in the dynamics of SHP, infiltration rates and soil moisture dynamics for soils under similar agricultural management practices. They are attributed mainly to a lack of standardization of research methodology as well as to site-specific conditions. Furthermore, it was also seen that incorporating the temporal dynamics of SHP in hydrological models produce more reliable and accurate modeling outcomes in comparison to studies with constant SHP as model input. The evaluation of the pore evolution model illustrated its suitability in capturing the temporal dynamics of soil pore space in response to tillage and environmental influences. High effective rainfalls and plant growth stages at which measurements were done affected the model performance. The use of sink/source terms and providing new initial conditions after high intensity rainfall events were provided as a means to improve the modeling outcomes. Though the model performed quite well in obtaining the water retention function as well as the saturated hydraulic conductivity and hydraulic conductivity functions, the high spatial variability in the sampling sites hampered with the model output. However, the main limitation lay in the lack of availability of sufficient data sets to calibrate and validate the model and its coefficients as well as for the derivation of SHP from the model. Overall, this study is a forerunner in predicting the temporal dynamics of soil structure and hydraulic properties. The established dynamics in the water retention and hydraulic conductivity functions can be used in hydrological simulations for planning land-use and management measures. The current study also reveals the need for more measurements and data sets that capture the alterations in soil hydraulic properties on a long-term basis.

Des données synchroniques de sols observées pendant 20 à 40 ans dans différentes zones agro-écologiques du Togo ont été utilisées pour analyser l’évolution de la fertilité du sol sous différentes pratiques culturales. Trois modèles de complexité croissante d’évolution du stock de C du sol ont été testés et le plus performant a été couplé au modèle QUEFTS pour apprécier l’effet des pratiques culturales sur l’évolution de cette fertilité. Bien que la fertilisation NPK ait été continuellement appliquée à la dose recommandée par la recherche (RR) ou à cette dose augmentée de 50% (1.5RR), les rendements initiaux de 1,5 à 2 t ha-1 pour le coton et 2 à 3 t ha-1 pour les céréales ont baissé pour ne représenter qu’environ 1 t ha-1 après 20 années de culture. L’alternance de 2 ou 3 années consécutives de jachère à la rotation culturale n’a pas amélioré le rendement des cultures, mais tend à stabiliser la production dans le long terme. Les stocks initiaux de C de 15 à 39 t ha-1 ont fortement baissé durant les premières années et ont atteint un équilibre après seulement 10 à 15 ans de culture. La baisse du stock de C a été peu dépendante de la quantité de C retournant au sol et a été mieux décrite avec le modèle à deux compartiments de C. Le taux de minéralisation de la fraction labile de C varie entre 0,15 et 0,24 et le stock de C stable représente 33-67% du stock initial ; pour une quantité de C ayant résisté à l’hydrolyse acide de 30% du stock initial. Les stocks de N du sol ont fortement baissé et cette baisse a été peu dépendante de la dose de fertilisation minérale. L’apport NPK à 1.5RR pendant 20 ans a contribué significativement à la baisse du pH du sol, puis a accentué de 20 et 40% les baisses respectives des concentrations de Ca et Mg. Nous avons conclu que l’apport de Ca/Mg couplé à celui de NPK à la dose RR serait (plus que le surdosage de P et K) une meilleure option pour gérer la fertilité minérale de ces sols. Le labour minimum, associé aux apports externes de C, devrait permettre de mieux gérer leur fertilité organique. / Using data observed during 20 to 40 years in different agro-ecosystems of Togo, we analyzed soil fertility dynamic under different agricultural management practices. Tree soil C models of varying complexity were tested, and the most accurate in terms of soil C dynamics description in these tropical soils was associated with the QUEFTS model and with nutrient partial balance to analyze the effect of each agricultural management practice. Results indicated that, although fertilizers N, P and K were continuously applied at a recommended rate by the research (RR) or at this rate increased in 50% (1.5RR), initial cotton and cereals yields of 1.5 to 2 t ha˗1 and 2 to 3 t ha˗1, respectively, decreased to around 1 t ha-1 after 20 years of soil cultivation. Alternation of 2 or 3 consecutive years of fallow plot with 3 years cropping did not improved crops yields as compared with continuous cropping system, but tends to stabilize crops production in the long-term. In the upper soil 20 cm, soil C stock of 15 to 39 t ha˗1 after vegetation clearance dropped faster during the first years to reached equilibrium after only 10 to 15 years of soil cultivation. Such a decrease was less depended on soil C input and was accurately descried with a two-pool C model, running on an annual time step basis. Calibrated decomposition rate of labile soil C ranged between 0.15 and 0.24, and stable soil C pool represented 33 to 67% of initial total C; while soil C that was resistant for hydrolyze acid was about 30% of initial total C. Soil N stock was also decreased faster and less associated with a rate of N input to soil. Application of NPK at 1.5RR during 20 years contributed significantly to soil pH decrease and pronounced soil Ca and Mg decreases, which respectively were 20 and 40% more than in plots that were not fertilized. We concluded that, association of Ca/Mg and NPK at a recommended rate would be (more than over-dosage of P and K) a promising option to achieving a sustainable management of mineral fertility of these soils. Minimum tillage, associated with external C input, would be an asset to a sustainable management of organic fertility of these soils.

Current mass consumption and production patterns have led to an unsustainable use of natural resources, including freshwater, which is one the most crucial natural resources for people and the planet. Agricultural production systems are alone responsible for 72 percent of all global water withdrawals. Therefore, companies with an agricultural supply chain, such as IKEA, play an important role in minimising the negative impacts on water due to agricultural production. In order to do so, companies need to conduct assessments to become aware of their contribution and thereafter develop water targets and strategies on how to mitigate and minimise the company’s water impacts. Several different guidances, methodologies, and tools aiding companies in how to do this have recently been developed, however there is not yet one common established methodology. Therefore, this research project aimed to test and evaluate a new assessment methodology and tool for IKEA, by conducting a water risk assessment on water availability of the sourcing locations of two of IKEA’s key agricultural raw materials: soy and palm oil, and identifying mitigation possibilities, in order to aid in the company’s further work with achieving a sustainable water management and material sourcing, and developing water targets and strategies. Soy is a prioritised raw material to work with in terms of environmental and social risk, and through the conducted water risk assessment, a high risk for water depletion was identified in many of the company’s most likely sourcing locations. The water risk assessment results for palm oil indicated no significant risk for water depletion in any of the sourcing locations. The high water depletion risk scores of the hotspots were found to most likely be due to several different causes, both climatic conditions and anthropogenic activities, including agriculture. Therefore, it was identified that there are possibilities for a company, such as IKEA, to aid in mitigating the water availability challenge of the hotspots through the company’s agricultural supply chain. For example, a few agricultural management practices for increasing the water use efficiency suitable for the different hotspots could be recommended and incentivised by the company to the local farmers of the hotspot sourcing locations. The WRA methodology used in this research project, following the guidance for setting enterprise water targets by Reig et al. (2021), was assumed suitable for IKEA to use and incorporate into a more comprehensive environmental assessment methodology for agricultural raw materials in order to develop water targets and strategies. The Aqueduct Food tool was assumed to be a robust tool for water risk assessment, and can be a suitable tool for IKEA to use to quickly gain a high-level picture of a material, location or water risk of specific interest. The water risk assessment results will mainly be useful in the future, when more comprehensive environmental assessments of agricultural raw materials have been conducted, and the water use has been assessed to a greater extent. Future research is needed to be conducted in order to confirm the results and recommendations of this research project, as well as to further complement the results for the needs of IKEA. This future research includes conducting field studies of the identified hotspots and the operations of the local farms, and conducting an equivalent water risk assessment of the two materials and sourcing locations on the water quality. / Nuvarande masskonsumtion och produktionsmönster har lett till en ohållbar användning av naturresurser, inklusive sötvatten, som är en av de mest avgörande naturresurserna för människor och planeten. Jordbruk står ensam för 72 procent av alla globala vattenuttag. Därför har företag med en jordbruksförsörjningskedja, som IKEA, en viktig roll i att hjälpa till att minska de negativa effekterna på vatten från jordbruk. För att kunna göra det behöver företag utföra bedömningar för att bli medvetna om sin vattenpåverkan och därefter utveckla vattenmål och strategier för hur de kan minimera den påverkan. Det har nyligen utvecklats flera olika vägledningar, metodiker och verktyg som hjälper företag att göra detta, men det finns ännu inte en gemensam etablerad metodik. Därför hade detta forskningsprojekt som syfte att testa och utvärdera en ny bedömningsmetodik och ett nytt verktyg för IKEA, genom att genomföra en riskbedömning av vattentillgången för produktionsplatserna för två av IKEAs viktigaste jordbruksråmaterial: soja och palmolja, och genom en litteraturstudie identifiera möjligheter att minska vattenpåverkan, för att bidra till företagets fortsatta arbete med att uppnå en hållbar vattenanvändning och materialförsörjning i jordbruksförsörjningskedjan och att utveckla vattenmål och strategier. Soja är ett prioriterat råmaterial att arbeta med gällande miljömässiga och sociala risker, och genom den genomförda bedömningen av vattenrisker identifierades en hög risk för vattenbrist på många av de produktionsplatser som företaget mest troligen köper in ifrån. Riskbedömningsresultaten för palmolja indikerade ingen signifikant risk för vattenbrist på någon av de produktionsplatser som företaget troligen köper in ifrån. Den höga risken för vattenbrist i de identifierade hotspotsen visade sig troligen bero på flera olika orsaker, både klimatförhållanden och antropogena aktiviteter, inklusive jordbruk. Därför identifierades det att det finns möjligheter för ett företag, som IKEA, att kunna hjälpa till att mildra vattentillgänglighetsutmaningen i hotspotsen genom företagets jordbruksförsörjningskedja. Till exempel kan ett par olika jordbruksförvaltningsmetoder för att öka vattenanvändningseffektiviteten som är lämpliga för de olika hotspotsen rekommenderas och uppmuntras av företaget till de lokala bönderna. Den WRA-metodik som används i detta forskningsprojekt, som följde vägledningen för att sätta upp vattenmål för företag av Reig et al. (2021), antogs lämplig för IKEA att använda och integrera i en mer omfattande miljöbedömningsmetodik för jordbruksråmaterial för att utveckla vattenmål och strategier. Aqueduct Food-verktyget antogs vara ett robust verktyg för riskbedömning för vatten och kan vara ett lämpligt verktyg för IKEA att använda för att snabbt få en översikt av ett material, en plats eller en vattenrisk av specifikt intresse. Riskbedömningsresultaten kommer främst att vara användbara i framtiden, när mer omfattande miljöbedömningar av jordbruksråvaror har gjorts och vattenanvändningen har bedömts i större utsträckning. Framtida forskning behövs för att bekräfta resultaten och rekommendationerna i detta forskningsprojekt, samt för att ytterligare komplettera resultaten för IKEAs behov. Denna framtida forskning inkluderar att genomföra fältstudier av de identifierade hotspotsen och de lokala sojafarmernas verksamhet, och att genomföra en likvärdig riskbedömning för vatten av de två materialen och produktionsplatserna som utförts i detta forskningsprojekt för vattenkvaliteten.