The project recommends the following:

  1. In Fiji, Tuvalu and Kiribati the project was greatly hindered from pandemic and in Fiji cyclones. Further follow-up four-year trials undertaking farming system nutrient management should be formulated and implemented. Additional training and awareness should be made to key stakeholders on nutrient budgeting work especially on commercial crops.
  2. Continued capability improvement of in-country staff not only in sustainability farming system but also project management, communication, information technology and research extension
  3. Improved connections with University of the South Pacific and Fiji National University to develop environmental science graduate students and staff research and training either through higher education opportunities or collaborative research projects.
  4. Soil carbon data from Tonga and Fiji shows that there has been a significant decline in stocks since the 1990s. It is critical for on-going agriculture production to develop farming systems, a consumer base and policies that build soil carbon. Current Australian research has shown it is possible to build soil carbon in dry land cropping systems through appropriate nutrient management. Voluntary carbon markets may be a mechanism to improve farm income through international carbon sequestration purchases. SMCN-2020-139 will need to undertake briefings with Government agencies and others about the soil carbon findings and to enable the develop of a policy platform and research strategy to improve soil carbon sequestration. This co-developed strategy and platform should be tested within the SMCN-2020-139 field trials.
  5. Organic farming systems are being promoted in all Pacific Island Countries and Territories. There is a need to develop a sustainable compost industry for agricultural production systems and ensure that composts are tailored to the specific soil contrasts.
  6. GxExM research. There is an opportunity to investigate GxExM interactions for important Pacific Island Countries and Territories crops. Current research has utilised typically market available varieties and this limits our ability to determine genetic traits that may be suitable for different soil types and climate and associated farming system management. This research should also include nutritional quality of the harvested materials.
  7. Mid-infrared spectroscopy has been successfully used to augment soil laboratory analysis at the Fiji Agricultural Chemistry Laboratory, Ministry of Agriculture. During the project pilot soil spectral reference libraries for Fiji and Tonga have been developed but need to be expanded and built upon facilitated through conducting soil surveys of representative Pacific Island Countries and Territories soils. Recommended next steps at Fiji Agricultural Chemistry Laboratory, Ministry of Agriculture are to refine and establish protocols/standards for applying existing calibration models to new spectral data including the selection of a set of sub-samples for wet chemistry analysis to improve existing calibration models, together with the development of a soil database capturing the soil information collected following standard methods and protocols. Rapid infrared spectroscopy-based analysis should be further expanded to other Pacific Island Countries and Territories, and the use of infrared spectroscopy offerings capable to be taken to the field should be explored, such as near-infrared and portable X-ray fluorescence spectroscopy. Mid-infrared, near-infrared and possibly portable X-ray fluorescence spectroscopy should also be explored within a broader Pacific Island Countries and Territories laboratory impact and business plan.
  8. Pacific Soil Portal needs to be transferred to the Pacific Community to improve linkages to users and decision makers. The Portal should also be used to develop new soil mapping and land capability mapping in sub-regions for land use planning.
  9. Development of nutrient recommendations for taro:
    1. The proposed conceptual framework needs validation, and soil nutrient indexes for major elements established for taro as well as other crops used in rotation. Yield-to-nitrogen response relationships need to be determined at key locations and information combined with climatic and nutrient input cost data to be able to provide annual fertiliser (Nitrogen) recommendations. The communication effort may be coordinated by extension officers, and information delivered to farmers on an annual basis to assist with their nutrient decisions.
    2. The relative effects of nutrient source (organic amendment, mineral fertiliser) and placement (surface-applied, soil incorporation) on the crop agronomic performance require further investigation. There is a need to quantify pathways of nutrient losses and better understand the mechanisms involved. These are important considerations to ensure nutrient recovery in harvested plant material (use efficiency) is maximised and the risk of environmental losses of nutrients is minimised.
  10. Nutrient balance and intensification of taro production systems:
    1. There is a need to refine the field-scale nutrient balance calculations reported in this work to assist the development of long-term nutrient management policies aimed at maintaining the productivity of taro soils. Establishment of permanent experimental sites, carefully designed for long-term monitoring of soil nutrient dynamics and agronomic performance of taro, could serve to that purpose. Information derived from these sites can be then used to develop nutrient advice.
    2. Intensification of taro production may be possible through the establishment of double taro cropping systems whereby a second taro crop is planted (intercropped) before the main crop has been harvested. The main objective of this planting system would be to be able to harvest two crops in for example 12 months instead of 16 months, which is more common. This approach would require adjusting plant nutrition to ensure any adverse impacts on soil fertility and crop productivity are avoided. It will also require careful management of soil water, particularly regarding weeds control, and optimisation of the planting window for the second crop. Virtual trials in Agricultural Production Systems Simulator (APSIM) could be used to inform the design and management of double taro cropping systems and assist the establishment of future experimental work.
    3. Interest in reducing herbicide and manufactured fertiliser use, particularly to produce food crops, could increase adoption of tillage for weed control and application of organic amendments to meet crops demand for nutrients. This will require the development of best management practices for tillage and understanding of the fertiliser replacement value and nutrient release characteristics of organic amendments. Tillage management protocols will ensure adverse effects on soil are minimised; namely: (a) increased rate of oxidation of soil organic matter due to manipulation of soil for weed control and seedbed preparation, (2) increased risk of soil compaction due to tillage (most soil) and traffic (low bearing capacity), (3) soil structural damage, and (4) reduced rate of infiltration (reduced surface cover) leading to increased runoff and erosion, and therefore nutrient and sediment transport to surface waters.
  11. Further development of the taro module in the APSIM modelling framework:
    1. The taro module in APSIM was developed by Crimp et al. (2017) using data collected in Fiji, Vanuatu, and Tonga. The existing module has limitations as it is site and variety specific. Further development of this module will improve its capability to simulate the response of taro to projected changes in climate in the Pacific and to identify strategies for farming systems adaptation.
    2. If this model could be developed for release, then they would be useful given the importance of taro (and cassava) as staple for large numbers of people globally.
    3. Significant savings in field-based experimental work could potentially be realised if this was, in part, replaced by virtual trials, leaving field trials for verification of modelling outcomes.

Click here to see the Recommendations presentation.



Crimp, S., Lisson, S., Gleadow, R., Hargreaves, J., Gabriel, E., Meier, E., Nishi, M., Nauluvula, P., Melteras, M., Leo, P. (2017). Understanding the response of taro and cassava to climate change. Final Report No.: FR2019-72 (ACIAR Project No.: PC-2012-011), pp. 140. Available at: (Accessed 05 Dec. 2021).