IMPACTS OF LANDSCAPE MANAGEMENT OPTIONS AND PRACTICES ON ECOSYSTEM SERVICES IN THE CHANGING CLIMATE OF CENTRAL ETHIOPIA

Abstract

In Ethiopia, about 80 percent of the population is dependent on rainfed agriculture. However, land degradation coupled with climate variability and changes made the rainfed agriculture a risky venture for a century. Hence, moving from conventional practices to a combination of conservation and management practices and/or technologies is of paramount importance to make the agriculture resilient and sustainable. This study aims to comprehensively assess the impacts of different landscape management options and practices on multiple ecosystem services (ES) in the face of climate change across space and time. Meta-data analysis and synthesis were performed based on a critical review of previous studies to assess the impact of various land management (LM) practices on multiple ES provision in Ethiopia. Based on meta-data analysis and synthesis result, soil bund (SB), crop residue (CR), integrated conservation practice (ICP), berken plough (BP) and conventional practice (CP) were selected for empirical field experiment evaluation that was conducted for two seasons (2020-2021) at Kulumssa Agricultural Research Center (KARC). Thence, modeling of the impact of LM practices on soil water balance and water use efficiency (WUEET) was performed in the current and future 2050 climates using AquaCrop model. Retrospective and foresight analysis of landscape transformations were conducted for (1986, 2001, 2011, 2021) and (2031, 2051) periods. Thence, four antecedent landscape management scenarios (i.e., Business-As-Usual (BAU), Rapid Agricultural Expansion (RAE), Ecosystems Protection and Agricultural Development (EPAD) and Landscape Ecosystems Restoration and Conservation (LERC) were developed using the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) and multilayer perceptron–artificial neuron network (MLP–ANN) model. Ecosystem service valuation (ESV) and habitat quality modelling were performed to evaluate the landscape management scenarios by 2051 compared to the base year 2021. The meta-data synthesis result revealed that LM practices were efficacious in enhancing the delivery of ES bundles, but soil bunds and fanyaa juus reduced the grain yield on average by 24 and 22%, respectively. However, the landscape-level collective impact of different LM practices was exhibited without any ES trade offs. The empirical experiment result showed that the LM practices significantly reduced runoff and soil loss on average from 27.1 – 55.8% and 15 – 76%, respectively compared to conventional practice (CP). The lowest losses of TN, Av. P, OM, and OC were recorded sequentially from SB, ICP, BP, and CR treated plots. Although there were no significant differences among LM practices at 0-20 cm in bulk density, it was slightly higher under CR and ICP than under CP. Soil porosity under BP and ICP plots was significantly higher than that of CP. FC and WP were not significantly different among the LM practices. PAWC increased on average by 8, 18.7, 20.4 and 37.9% under SB, CR, ICP and BP, respectively. Soil organic matter (OM) and carbon (SC) stock in the lower 20–40 cm soil depths were significantly different in the order of ICP > CR > BP > CP>SB. The average grain yield increased by 8.1, 10, 10.1 and 12.1% under SB, CR, ICP and BP, respectively, compared to CP. Similarly, wheat biomass was also significantly different among LM practices and increased on average by 5.3, 15.8, 7.1 and 31.1% under SB, CR, ICP and BP, respectively, compared to CP. Under the current and future climates; WUEET, evapotranspiration (ET) and transpiration (T) were higher in order of ICP>CR>BP>SB>CP. The landscape analysis results xx revealed that cultivated land, bare land and built-up areas increased immensely and persistently in the years between 1986 and 2021. In contrast, forest, grazing land, wetland, and shrub-bush woodland significantly decreased by 37, 81, 38.4, and 25.7%, respectively. Forest, water, grazing land, wetland, and shrub-bush land had significant area losses between the study periods of 1986– 2031 and 1986–2051. Cultivated land, built-up areas and bare land are predicted to continue to gain more area in the years to come (2031 and 2050). Under the BAU and RAE scenarios, cultivated land increased by 146,548 ha (22%) and 193,965 ha (29%), respectively, whereas forest, water body, wetland and shrub-bush land decreased. However, forest cover increased by 31,725 ha and 100,080 ha, but bare land reduced by 8466 ha (21%) and 10,379 ha (25%) under the EPAD and LERC scenarios, respectively. The forest cover annual rate of change was 3.2% and 6% under the EPAD and LERC scenarios, respectively. As a result, the total loss of ESV was reduced by US$58.3 million and 85.4 million in the periods of 1986–2021 and 1986–2051, respectively. The total ESVs were reduced by USD 73 million and 27 million under the RAE and BAU scenarios, respectively, compared to the base year 2021. In contrast, the total ESV under the EPAD and LERC scenarios increased by USD 24.5 million and 79 million compared to the base year 2021. ES without/minimum trade-offs, well-designed bundling of different LM practices is crucial for agricultural landscape conservation. To enhance landscape ecosystem multifunctionality, sustainable intensification and ecosystem restoration strategies are needed