Philippa Kaur

Saponari, M. et al. Infectivity and transmission of Xylella fastidiosa by Philaenus spumarius (Hemiptera: Aphrophoridae) in Apulia, Italy. J. Econ. Entomol. 107, 1316–1319. https://doi.org/10.1603/EC14142 (2014).Article 

Google Scholar 
Cornara, D., Bosco, D. & Fereres, A. Philaenus spumarius: When an old acquaintance becomes a new threat to European agriculture. J. Pest Sci. 91, 957–972. https://doi.org/10.1007/s10340-018-0966-0 (2018).Article 

Google Scholar 
Weaver, C. R. & King, D. Meadow spittlebug, Philaenus leucophthalmus (L.). Ohio Agric. Exp. Stn. Res. Bull. 741, 258 (1954).
Google Scholar 
Halkka, A., Halkka, L., Halkka, O., Roukka, K. & Pokki, J. Lagged effects of North Atlantic Oscillation on spittlebug Philaenus spumarius (Homoptera) abundance and survival. Glob. Change Biol. 12, 2250–2262. https://doi.org/10.1111/j.1365-2486.2006.01266.x (2006).Article 

Google Scholar 
Cruaud, A. et al. Using insects to detect, monitor and predict the distribution of Xylella fastidiosa: A case study in Corsica. Sci. Rep. 8, 15628. https://doi.org/10.1038/s41598-018-33957-z (2018).Article 
CAS 

Google Scholar 
Godefroid, M. et al. Climate tolerances of Philaenus spumarius should be considered in risk assessment of disease outbreaks related to Xylella fastidiosa. J. Pest Sci. 2021, 1–14. https://doi.org/10.1007/s10340-021-01413-z (2021).Article 

Google Scholar 
Farigoule, P. et al. Vectors as sentinels: Rising temperatures increase the risk of Xylella fastidiosa outbreaks. Biology 11, 1299. https://doi.org/10.3390/biology11091299 (2022).Article 

Google Scholar 
Drosopoulos, S. & Asche, M. Biosystematic studies on the spittlebug genus Philaenus with the description of a new species. Zool. J. Linn. Soc. 101, 169–177. https://doi.org/10.1111/j.1096-3642.1991.tb00891.x (1991).Article 

Google Scholar 
Godefroid, M. & Durán, J. M. Composition of landscape impacts the distribution of the main vectors of Xylella fastidiosa in southern Spain. J. Appl. Entomol. 146, 666–675. https://doi.org/10.1111/jen.13003 (2022).Article 

Google Scholar 
Karban, R. & Strauss, S. Y. Physiological tolerance, climate change, and a northward range shift in the spittlebug, Philaenus spumarius. Ecol. Entomol. 29, 251–254. https://doi.org/10.1111/j.1365-2311.2004.00576.x (2004).Article 

Google Scholar 
Chmiel, S. M. & Wilson, M. C. Estimation of the lower and upper developmental threshold temperatures and duration of the nymphal stages of the meadow Spittlebug, Philaenus spumarius. Environ. Entomol. 8, 682–685. https://doi.org/10.1093/ee/8.4.682 (1979).Article 

Google Scholar 
Yurtsever, S. On the polymorphic meadow spittlebug, Philaenus spumarius (L.) (Homoptera: Cercopidae). Turk. J. Zool. 24, 447–460 (2000).
Google Scholar 
Ahmed, D. D. & Davidson, R. H. Life history of the meadow spittlebug in Ohio. J. Econ. Entomol. 43, 905–908. https://doi.org/10.1093/jee/43.6.905 (1950).Article 

Google Scholar 
Whittaker, J. B. Cercopid spittle as a microhabitat. Oikos 21, 59–64. https://doi.org/10.2307/3543839 (1970).Article 

Google Scholar 
Drosopoulos, S. New data on the nature and origin of colour polymorphism in the spittlebug genus Philaenus (Hemiptera: Aphorophoridae). Ann. Soc. Entomol. Fr. NS 39, 31–42. https://doi.org/10.1080/00379271.2003.10697360 (2003).Article 

Google Scholar 
Bodino, N. et al. Phenology, seasonal abundance, and host-plant association of spittlebugs (Hemiptera: Aphrophoridae) in vineyards of Northwestern Italy. Insects 12, 1012. https://doi.org/10.3390/insects12111012 (2021).Article 

Google Scholar 
Cornara, D. et al. Natural areas as reservoir of candidate vectors of Xylella fastidiosa. Bull. Insectol. 74, 173–180 (2021).
Google Scholar 
Gargani, E. et al. A five-year survey in Tuscany (Italy) and detection of Xylella fastidiosa subspecies multiplex in potential insect vectors, collected in Monte Argentario. Redia 104, 75–88. https://doi.org/10.19263/REDIA-104.21.09 (2021).Article 

Google Scholar 
Morente, M. et al. Distribution and relative abundance of insect vectors of Xylella fastidiosa in olive groves of the iberian peninsula. Insects 9, 175. https://doi.org/10.3390/insects9040175 (2018).Article 

Google Scholar 
Delong, D. et al. Spittle-insect vectors of Pierce’s disease virus. I. Characters, distribution, and food plants. Hilgardia 19, 339–356 (1950).Article 

Google Scholar 
Bodino, N. et al. Phenology, seasonal abundance and stage-structure of spittlebug (Hemiptera: Aphrophoridae) populations in olive groves in Italy. Sci. Rep. 9, 1–17. https://doi.org/10.1038/s41598-019-54279-8 (2019).Article 
CAS 

Google Scholar 
Wiegert, R. G. Population energetics of meadow spittlebugs (Philaenus spumarius L.) as affected by migration and habitat. Ecol. Monogr. 34, 217–241. https://doi.org/10.2307/1948501 (1964).Article 

Google Scholar 
Dongiovanni, C. et al. Plant selection and population trend of spittlebug immatures (Hemiptera: Aphrophoridae) in olive groves of the Apulia region of Italy. J. Econ. Entomol. 112, 67–74. https://doi.org/10.1093/jee/toy289 (2019).Article 

Google Scholar 
Bodino, N. et al. Spittlebugs of mediterranean olive groves: Host-plant exploitation throughout the year. Insects 11, 130. https://doi.org/10.3390/insects11020130 (2020).Article 

Google Scholar 
Villa, M., Rodrigues, I., Baptista, P., Fereres, A. & Pereira, J. A. Populations and host/non-host plants of spittlebugs nymphs in olive orchards from northeastern Portugal. Insects 11, 720. https://doi.org/10.3390/insects11100720 (2020).Article 

Google Scholar 
Antonatos, S. et al. Seasonal appearance, abundance, and host preference of Philaenus spumarius and Neophilaenus campestris (Hemiptera: Aphrophoridae) in olive groves in Greece. Environ. Entomol. 50, 1474–1482. https://doi.org/10.1093/ee/nvab093 (2021).Article 

Google Scholar 
Hasbroucq, S., Casarin, N., Ewelina, C., Bragard, C. & Grégoire, J.-C. Distribution, adult phenology and life history traits of potential insect vectors of Xylella fastidiosa in Belgium. Belg. J. Entomol. 92, 2569 (2020).
Google Scholar 
Mesmin, X. et al. Interaction networks between spittlebugs and vegetation types in and around olive and clementine groves of Corsica; implications for the spread of Xylella fastidiosa. Agric. Ecosyst. Environ. 334, 107979. https://doi.org/10.1016/j.agee.2022.107979 (2022).Article 

Google Scholar 
Albre, J., García-Carrasco, J. M. & Gibernau, M. Ecology of the meadow spittlebug Philaenus spumarius in the Ajaccio region (Corsica)—I: Spring. Bull. Entomol. Res. 111, 246–256. https://doi.org/10.1017/S0007485320000711 (2021).Article 

Google Scholar 
Andersson, P., Löfstedt, C. & Hambäck, P. A. Insect density–plant density relationships: A modified view of insect responses to resource concentrations. Oecologia 173, 1333–1344. https://doi.org/10.1007/s00442-013-2737-1 (2013).Article 

Google Scholar 
Hambäck, P. A., Inouye, B. D., Andersson, P. & Underwood, N. Effects of plant neighborhoods on plant–herbivore interactions: Resource dilution and associational effects. Ecology 95, 1370–1383. https://doi.org/10.1890/13-0793.1 (2014).Article 

Google Scholar 
Otway, S. J., Hector, A. & Lawton, J. H. Resource dilution effects on specialist insect herbivores in a grassland biodiversity experiment. J. Anim. Ecol. 74, 234–240 (2005).Article 

Google Scholar 
Lago, C. et al. Flight performance and the factors affecting the flight behaviour of Philaenus spumarius the main vector of Xylella fastidiosa in Europe. Sci. Rep. 11, 17608. https://doi.org/10.1038/s41598-021-96904-5 (2021).Article 
CAS 

Google Scholar 
Casarin, N. et al. Investigating dispersal abilities of Aphrophoridae in European temperate regions to assess the threat of potential Xylella fastidiosa-based pathosystems. J. Pest Sci. https://doi.org/10.1007/s10340-022-01562-9 (2022).Article 

Google Scholar 
Bodino, N. et al. Dispersal of Philaenus spumarius (Hemiptera: Aphrophoridae), a vector of Xylella fastidiosa, in olive grove and meadow agroecosystems. Environ. Entomol. 50, 267–279. https://doi.org/10.1093/ee/nvaa140 (2020).Article 
CAS 

Google Scholar 
Santoiemma, G., Tamburini, G., Sanna, F., Mori, N. & Marini, L. Landscape composition predicts the distribution of Philaenus spumarius, vector of Xylella fastidiosa, in olive groves. J. Pest. Sci. 92, 1101–1109. https://doi.org/10.1007/s10340-019-01095-8 (2019).Article 

Google Scholar 
Cappellari, A. et al. Spatio-temporal dynamics of vectors of Xylella fastidiosa subsp. pauca across heterogeneous landscapes. Entomol. Gen. 42, 515–521. https://doi.org/10.1127/entomologia/2022/1427 (2022).Article 

Google Scholar 
Avosani, S., Tattoni, C., Mazzoni, V. & Ciolli, M. Occupancy and detection of agricultural threats: The case of Philaenus spumarius, European vector of Xylella fastidiosa. Agric. Ecosyst. Environ. 324, 107707. https://doi.org/10.1016/j.agee.2021.107707 (2022).Article 

Google Scholar 
Allier, C. & Lacoste, A. Processus dynamiques de reconstitution dans la série du Quercus ilex en Corse. In Vegetation Dynamics in Grasslans, Healthlands and Mediterranean Ligneous Formations 83–91 (Springer, 1981).Delbosc, P., Bioret, F. & Panaïotis, C. Plant landscape of Corsica: Typology and mapping plant landscape of Cap Corse region and Biguglia Pond (Springer Nature, 2020).Book 

Google Scholar 
Chessel, D., Dufour, A.-B. & Thioulouse, J. The ade4 package—I: One-table methods. R. News 4, 5–10 (2004).
Google Scholar 
Biedermann, R. & Niedringhaus, R. The Plant-and Leafhoppers of Germany: Identification Key to All Species (Wabv Fründ, 2009).
Google Scholar 
Stöckmann, M., Biedermann, R., Nickel, H. & Niedringhaus, R. The Nymphs of the Planthoppers and Leafhoppers of Germany (WABV, 2013).
Google Scholar 
INRAE-CBGP. Arthemis DB@se – ARTHropod Ecology, Molecular Identification and Systematics. https://arthemisdb.supagro.inrae.frhttps://doi.org/10.15454/TBGRIB. Accessed 2021.Xu, T. & Hutchinson, M. ANUCLIM version 6.1 user guide. Aust. Natl. Univ. Fenner Sch. Environ. Soc. Canberra 2011, 256 (2011).
Google Scholar 
Quintana-Seguí, P. et al. Analysis of near-surface atmospheric variables: Validation of the SAFRAN analysis over France. J. Appl. Meteorol. Climatol. 47, 92–107. https://doi.org/10.1175/2007JAMC1636.1 (2008).Article 

Google Scholar 
Hijmans, R. J., Phillips, S., Leathwick, J. & Elith, J. Dismo: Species Distribution Modeling https://CRAN.R-project.org/package=dismo (2017).Faraway, J. J. Extending the Linear Model with R: Generalized Linear, Mixed Effects and Nonparametric Regression Models (Chapman and Hall/CRC, 2006).MATH 

Google Scholar 
Brooks, M. E. et al. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R. J. 9, 378–400. https://doi.org/10.3929/ethz-b-000240890 (2017).Article 

Google Scholar 
R Core Team. R: A Language and Environment for Statistical Computing https://www.R-project.org/ (2019).Hardin, J. W. & Hilbe, J. M. Generalized Linear Models and Extensions 4th edn. (Stata Press, 2018).MATH 

Google Scholar 
Harrison, X. A. Using observation-level random effects to model overdispersion in count data in ecology and evolution. PeerJ 2, e616. https://doi.org/10.7717/peerj.616 (2014).Article 

Google Scholar 
Bolker, B. M. et al. Generalized linear mixed models: A practical guide for ecology and evolution. Trends Ecol. Evol. 24, 127–135. https://doi.org/10.1016/j.tree.2008.10.008 (2009).Article 

Google Scholar 
Hartig, F. DHARMa: Residual Diagnostics for Hierarchical (Multi-Level/Mixed) Regression Models https://CRAN.R-project.org/package=DHARMa (2020).Lüdecke, D., Ben-Shachar, M. S., Patil, I., Waggoner, P. & Makowski, D. performance: An R package for assessment, comparison and testing of statistical models. J. Open Sourc. Softw. 6, 3139. https://doi.org/10.21105/joss.03139 (2021).Article 

Google Scholar 
Fox, J. & Weisberg, S. An {R} Companion to Applied Regression Third edn, https://socialsciences.mcmaster.ca/jfox/Books/Companion/ (Sage, Thousand Oaks CA, 2019).Lenth, R. V. Emmeans: Estimated Marginal Means, Aka Least-Squares Means https://CRAN.R-project.org/package=emmeans (2021).Hothorn, T., Bretz, F. & Westfall, P. Simultaneous inference in general parametric models. Biom. J. 50, 346–363. https://doi.org/10.1002/bimj.200810425 (2008).Article 
MATH 

Google Scholar 
Fernández-Mazuecos, M. & Vargas, P. Ecological rather than geographical isolation dominates Quaternary formation of Mediterranean Cistus species. Mol. Ecol. 19, 1381–1395. https://doi.org/10.1111/j.1365-294X.2010.04549.x (2010).Article 
CAS 

Google Scholar 
Berenbaum, M. R. & Feeny, P. P. 1. Chemical mediation of host-plant specialization: The papilionid paradigm. In Specialization, Speciation, and Radiation (ed. Tilmon, K.) 3–19 (University of California Press, 2008). https://doi.org/10.1525/california/9780520251328.003.0001.Kapantaidaki, D. E., Antonatos, S., Evangelou, V., Papachristos, D. P. & Milonas, P. Genetic and endosymbiotic diversity of Greek populations of Philaenus spumarius, Philaenus signatus and Neophilaenus campestris, vectors of Xylella fastidiosa. Sci. Rep. 11, 3752. https://doi.org/10.1038/s41598-021-83109-z (2021).Article 
CAS 

Google Scholar 
Mesmin, X. et al. Ooctonus vulgatus (Hymenoptera, Mymaridae), a potential biocontrol agent to reduce populations of Philaenus spumarius (Hemiptera, Aphrophoridae) the main vector of Xylella fastidiosa in Europe. PeerJ 8, e8591. https://doi.org/10.7717/peerj.8591 (2020).Article 

Google Scholar 
Denancé, N. et al. Several subspecies and sequence types are associated with the emergence of Xylella fastidiosa in natural settings in France. Plant Pathol. 66, 1054–1064. https://doi.org/10.1111/ppa.12695 (2017).Article 
CAS 

Google Scholar 
EFSA, Delbianco, A., Gibin, D., Pasinato, L. & Morelli, M. Update of the Xylella spp host plant database—systematic literature search up to 31 December 2020. EFSA J. 19, 6. https://doi.org/10.2903/j.efsa.2021.6674 (2021).Article 

Google Scholar 
Soubeyrand, S. et al. Inferring pathogen dynamics from temporal count data: The emergence of Xylella fastidiosa in France is probably not recent. New Phytol. 219, 824–836. https://doi.org/10.1111/nph.15177 (2018).Article 

Google Scholar 
Roy, J. & Sonié, L. Germination and population dynamics of Cistus species in relation to fire. J. Appl. Ecol. 29, 647–655. https://doi.org/10.2307/2404472 (1992).Article 

Google Scholar 
Whittaker, J. B. Density regulation in a population of Philaenus spumarius (L.) (Homoptera: Cercopidae). J. Anim. Ecol. 42, 163–172. https://doi.org/10.2307/3410 (1973).Article 

Google Scholar 
Chapman, D. et al. Improving knowledge of Xylella fastidiosa vector ecology: modelling vector occurrence and abundance in the wider landscape in Scotland. Project Final Report. PHC2020/04, Scotland’s Centre of Expertise for Plant Health (PHC) https://doi.org/10.5281/zenodo.6523478 (2022).Saponari, M., Giampetruzzi, A., Loconsole, G., Boscia, D. & Saldarelli, P. Xylella fastidiosa in olive in Apulia: Where we stand. Phytopathology 109, 175–186. https://doi.org/10.1094/PHYTO-08-18-0319-FI (2019).Article 
CAS 

Google Scholar 
López-Mercadal, J. et al. Collection of data and information in Balearic Islands on biology of vectors and potential vectors of Xylella fastidiosa (GP/EFSA/ALPHA/017/01). EFSA Supp. Publ. 18, 10. https://doi.org/10.2903/sp.efsa.2021.EN-6925 (2021).Article 

Google Scholar  More

Himalayan marmot habitat analysisThe environmental factors including temperature, vegetation and elevation are the key drivers for the wildlife in alpine ecosystems32. Specific landform attributes such as slope and elevation and vegetation cover affect the population and burrowing of rodents33. For example, rodent burrows in the Western Usambara Mountains in Tanzania were only found at an elevation of above 1600 m33. However, the Himalayan marmot seems to prefer to inhabit areas with low elevation and high land surface temperature34. In this study, the data showed that 76.25% of the Himalayan marmots were found in areas with elevation values of 3400–4600 m. The majority of marmots were found in areas with slopes of 5–20° and vegetation cover higher than 60%. Most marmots were found in alpine meadows, a few were found in temperate grasslands and alpine grasslands, and none were found in other grassland types.Preliminary statistical analysis of vegetation cover, grass type, vegetation type, and Himalayan marmot distribution sample sites obtained using spatial geographic information technology revealed that the meadow grassland areas with lush grass growth, more dominant plants, and abundant food had more marmots. When the vegetation cover reached 0.60–1.00, the number of marmot distribution sample sites was the highest. Dense grass is an ideal habitat and provides concealment for Himalayan marmots, and the abundant plant types provide sufficient food for marmots. In contrast, no marmots were distributed in the alpine scrub, coniferous forest, and alpine snow/ice covered areas where vegetation growth was poor, vegetation cover was low, and food was relatively scarce. Moreover, 70.24% of Himalayan marmots were found in alpine meadows with a wide variety of plant species, including Poaceae, Cyperaceae, and grasses. This finding indicated that alpine meadows are more suitable for Himalayan marmots and have more advantageous habitat conditions compared with other grassland types. The elevation of alpine meadows is 3236–5126 m, and the vegetation is mainly meadows with simple vegetation structure, substantial vegetation cover and dense vegetation growth, and a wide variety of plants, rich food, soft grass, and good palatability. Therefore, alpine meadows provide good natural habitats and foraging sites for marmots.Habitat selection of large rodents is influenced by a combination of vegetation cover availability, food availability, and population density35. Vegetation cover is an important parameter that describes vegetation communities and ecosystems and is closely related to vegetation quantity and productivity. The quality of habitat vegetation is an important factor that affects the spatial distribution of plateau rodents. Both feeding and concealment depend on vegetation, and the height and cover of edible plants and vegetation suitable for concealment determine the choice of vegetation type by marmots. Thus, vegetation cover becomes an important factor for habitat selection by marmots. Different grassland types determine different plant conditions, and selection of different vegetation conditions can increase the chances of survival and improve the reproductive success of marmots; therefore, grassland type is an important ecological factor in habitat selection by marmots. A study showed that the ecological factors affecting habitat selection of Himalayan marmots are mainly topography, anthropogenic disturbance, and vegetation8. Another study concluded that habitat selection by Himalayan marmots is closely related to elements such as topography, landform, temperature, precipitation, and vegetation24.The functions of burrows’ physical parameters is to protect the Himalayan marmots from natural enemies and bad weather36. There is clearly influence of slope on habitat selection by marmots. When the slope is large, wind is strong, and burrows are not well hidden; this makes them difficult to defend against enemies, unsafe for survival, and not conducive to hibernation during winter. In addition, Himalayan marmots prefer to burrow on sunny aspect, because the temperature is suitable and the vegetation is lush, which is suitable for marmots to breed. Therefore, the number of marmot burrows gradually decreases with increasing slope and ubac. Although flat and low-lying areas with small slopes are good for marmots to create dens, rainwater will easily flow into the dens during summer rainfall, which will kill marmots. Therefore, a suitable slope and sunny aspect are also very important for habitat selection by marmots.Application of the predictive spatial distribution map of Himalayan marmots in Qinghai provincePlague surveillance is the main measure used for plague prevention and control in China. Although we have made many improvements in plague surveillance, the traditional method of dragnet surveillance still consumes a lot of human and material resources, is inefficient. The pasture area of Qinghai province is approximately 380,000 km2, and the identified natural plague focus is approximately 180,000 km2; therefore, there is still 200,000 km2 of pasture where the distribution of Himalayan marmots and plague have not been identified. Currently, RS technology is widely used in the fields of mapping and ecological surveillance18,19,21,22,37.Applications of RS technology in areas such as malaria, dengue, schistosomiasis and plague have been previously reported27,37. Using GIS combined with remotely sensed data, Proches Hieronimo et al. found that the presence of small mammals was positively influenced by elevation, whereas the presence of fleas was clearly influenced by land management features, and thus these observations have positive implications for plague surveillance27. In this study, RS technology combined with field validations were used to determine the distribution and areas of different types of grasslands in Qinghai province, and the average density of Himalayan marmot distribution in different types of grasslands. The high-, low-, and very low-density areas of Himalayan marmot distribution were identified. The soil map, vegetation map, administrative map, and marmot density statistics were merged to form the spatial data and attribute data basis for the information system to map the distribution of Himalayan marmot and determine the area of Himalayan marmot distribution. Generally speaking, the occurrence of human plague epidemic is closely related to the local animal plague epidemic2. However, a large part of the high-density distribution of Himalayan marmots is located in uninhabited areas and the areas are generally sparsely populated, which also indicates that we should reasonably allocate plague prevention and control resources to areas where human plague is most likely to occur to prevent the occurrence of human plague epidemics.Field validation for verificationThrough field validation and information from local farmers and herdsmen, we confirmed that Himalayan marmots inhabited 68 sample sites in Tongde, Zeku, Guinan, Xunhua, Haiyan, Ulan, Qilian, Hualong, and Huzhu counties. Among them, Tongde, Zeku, Guinan, Xunhua, Haiyan, Ulan, and Qilian counties have all historically experienced marmot plague outbreaks and can be considered as reliable natural plague foci38. The data from this field validation are consistent with the previous survey data and the epidemic history of the counties in Qinghai province39.MAE can better reflect the actual number of errors in prediction values; the smaller the MAE value, the higher the prediction accuracy. The MAE derived from the field validation data was 0.1331 and the prediction accuracy was 0.8669. The accuracy of the predicted Himalayan marmot spatial distribution reached 87%, which indicated that the predicted probability map of the Himalayan marmot spatial distribution can better predict the potential marmot distribution.The predicted spatial distribution map of Himalayan marmot in Qinghai province was then compared with environmental information such as elevation, vegetation, grass type, slope, and aspect of 352 field survey sites. The obtained RS data showed that the prediction results were excellent, and the predicted spatial distribution map of Himalayan marmot in Qinghai province was drawn with high accuracy. The prediction map visually reflects the different density distribution of Himalayan marmots; this allows us to optimize the settings and reasonable spatial layout of animal plague surveillance sites and improve surveillance efficiency.Application of marmot information collection system V3.0Marmot information collection system V3.0 was developed based on the “3S” technology standardizing the collection of surveillance data, and makes the management and analysis of information more convenient and faster. This study revolutionized the traditional method of considering plague-stricken counties as the plague foci, and effectively reduces the work intensity of operators and improves the data collection efficiency. In 2016 and 2017, we applied this system to the animal plague surveillance tasks in the plague-stricken counties of Haidong, Hainan, and Haibei in Qinghai province, and standardized the collection of provincial geographic location data of animal plague surveillance (data not shown). In 2018, we also applied this system in Wulan County, which frequently experiences plague, and achieved a good application effect (data not shown).In the next step, we will expand the pilot areas (mainly national and provincial plague surveillance sites), collect surveillance data from each surveillance site, continuously optimize and update the system, improve the efficiency of data analysis and utilization, detect the plague epidemic in marmot in a timely and accurate manner, correctly determine the epidemic trend of plague in marmots, and attempt to strictly prevent the plague from spreading to humans. We plan to use a new model of drone surveillance to create a multidimensional, three-dimensional, real-time big data plague surveillance information reporting system to enhance early plague warnings and prediction in Qinghai province and even in the country, which will be of positive practical significance to serve and guarantee the Belt and Road Initiative. These approaches are expected to provide new technical means for plague investigation and research, and to provide references for setting up plague surveillance programs and prediction for the natural Himalayan marmot plague focus in Qinghai province and the QTP. More

No plants were collected or harmed during this study, and all research involving plants followed relevant national, and international guidelines and legislation.Study areaThe study site encloses a wetland area bordering Ramalhete Channel, in the western part of the Ria Formosa lagoon, a mesotidal system located in southern Portugal (Fig. 1). Lunar tides are semi-diurnal, with a mean tidal range of about 2 m that can reach up to 3.5 m during spring tides. Offshore waves have no major propagation inside the lagoon33,34. Water circulation inside the lagoon is mostly driven by tides. The lagoon extends over 55 km along the coast and is connected to the ocean through six tidal inlets35. The three westmost inlets of the system (Ancão, Faro-Olhão, and Armona), which together capture ca. 90% of the total prism, are highly interconnected, with a strong residual circulation from Faro-Olhão Inlet directed towards Ancão and Armona inlets (located in Fig. 1), during both spring and neap tides36. The tidal currents in Ramalhete Channel, connecting the Faro-Olhão and Ancão Inlet, have high tidal asymmetry and shifts in tidal dominance, from flood to ebb. There are no significant fluvial inputs into the lagoon, with a yearly average terrestrial sediment influx of around 2 × 105 m3/yr37, reaching the system through small streams. The main sediment delivery to the system is through the inlets, though there are few studies assessing related fluxes. The net sediment entry through the stabilized Faro-Olhão Inlet is estimated at 1.4 × 105 m3/year38. Recent sedimentation rates in the marsh of the westmost edge of the lagoon were estimated at 1.1 ± 0.1 mm/yr39.The lagoon system is composed of large salt marsh patches, tidal flats and a complex net of natural, and partially dredged tidal channels. The tidal flats (vegetated and non-vegetated) and salt marshes represent more than 2/3 of the total lagoon area. The salt marshes comprise silt and fine sand40, while coarser (sand to shingle) shell-rich sediment, of marine provenance, is found on tidal channels and the lower domain of intertidal flats41. The dominant intertidal species are Spartina maritima and the seagrass Zostera noltei, the latter occupying an estimated area of 1304 ha, which represent 45% of the total intertidal area42.Figure 1Location of the field site in the Ria Formosa lagoon western sector over a satellite image collected in 2019 (South Portugal; upper panel); zoom to monitoring stations S1 to S4 (left lower panel); and field view of the studied site (right lower panel). Map generated with ArcGIS 10.8 (http://www.esri.com) and Adobe Illustrator 2022. Map data: Google Earth 7.3, image Landsat / Copernicus.Full size imageExperimental setup and data analysisAn experimental setup was deployed in the study area to assess dominant local topography, hydrodynamics (water levels and current velocities), Suspended Sediment Concentrations (SSCs), Deposition Rates (DRs), vegetation characteristics, and bed sediment grain size and organic matter content. Measurements were made during a full tide cycle, on a spring tide (tidal range = 3.2 m), and on a neap tide (tidal range = 1.8 m). Sampling was conducted in four wetland stations: S1 and S2 in a vegetated tidal flat comprising Zostera noltei; S3 in the low marsh comprising Spartina maritima; and S4 in the mid-upper marsh with the most abundant species of Sarcocornia perennis and Atriplex portucaloides (see S1 to S4, Fig. 1); the tidal flat is interrupted by a small oblique secondary tidal creek that flows near S2 station.Stations of sediment sampling and equipment deployment along the transect are illustrated in Fig. 2. During neap tide there was no data collection in S4, since the inundation time of the station was very short. The profile elevation was measured using Real Time Kinematic Differential Global Positioning System (RTK-DGPS, Trimble R6; vertical error in the order of few centimetres), and the slope of each habitat within a transect was calculated and expressed in percentage (%). Vegetation at each point was characterized by the canopy height, calculated as the average shoot length.Suspended Sediment Samplers (SSSs) were installed during low tide in the monitored stations using siphon samplers (Fig. 2) and recovered in the next low tide. These samplers consist of 0.5 L bottles with two holes on the cap, one for water intake and the other for air exhaust, according to the method described in13. Each intake tube is adjusted to form a siphon (i.e., inverse U), allowing to control the water level at which intake starts. Siphons were aligned at the same elevation along the transect for spring and neap tides, which means that all SSSs were collecting at the same time within the tidal cycle. During spring tide, in S1 and S2 at the tidal flat, SSSs were sampling at 0.1, 0.9, and 1.2 m from the bed, while at S3 SSSs were sampling at 0.7 and 1.0 m from the bed, and at S4 the SSS was sampling at 0.1 m from the bed (Fig. 2). During neap tide, in S1 and S2, SSSs were sampling at 0.1 and 0.9 m from the bed, while at S3 the SSS was sampling at 0.7 m from the bed.Surficial sediment samples were collected in each habitat to characterize the sediment grain size (d50) and content of organic matter (% OM). Sediment traps were installed in 3 replicates, during low tide, at each sampling point to measure the short-term sediment deposition rate (i.e., deposition over a tidal cycle, following procedures of43). Traps consisted of 3 cm diameter pre-labeled cylindrical tubes (Falcon® tubes, 50 ml). Traps and sediment samples were transported to the laboratory and maintained in a fridge. The sediment content was washed, and both the inorganic and organic weights were determined.The measured inorganic DR (g/m2/hr) was calculated as:$${text{DR}} = {raise0.7exhbox{${{text{W}}_{{{text{DS}}}} }$} !mathord{left/ {vphantom {{{text{W}}_{{{text{DS}}}} } {{text{A}} cdot {text{T}}}}}right.kern-0pt} !lower0.7exhbox{${{text{A}} cdot {text{T}}}$}}$$
(1)
where WDS is the weight of deposited sediment (in grams), A is the area of the sediment trap opening (m2), and T is in hours. Two different tide durations were considered to compute DRs, one assuming T equal to the hydroperiod in each station, and one assuming T equal to the entire tide duration (~ 12.4 h). These measured DRs are hereon mentioned as flood and tide DRs (DRflood and DRtide, respectively). The former is an expression of the actual deposition rate within the flood phase, during the period in which each station is inundated (and therefore active deposition can take place). The latter is the value used to compare with DRs in literature, which typically corresponds to values averaged over multiple tidal cycles (thus accounting for the entire tide duration).Tide levels were measured in the field using pressure sensors (PT, InSitu Inc. Level TROLL; ~ 2 cm from the bed), deployed from S2 towards S4 (Fig. 2). Velocity currents were measured at 20 cm from the bed, using an electromagnetic current meter (EMCM; Infinity Series JFE Advantech Co., Ltd; in S2 to S4; Fig. 2), and raw data (recording interval: 30 s) were filtered using a 10 min moving average for cross-shore and longshore components. To identify tidal asymmetry and assess the related phase dominance, tidal current skewness was calculated through the formula described in44 by which:
$$Sk_{U} = frac{{frac{1}{N – 1}mathop sum nolimits_{t = 1}^{N} left( {U_{t} – overline{U}} right)^{3} }}{{left( {frac{1}{N – 1}mathop sum nolimits_{t = 1}^{N} left( {U_{t} – overline{U}} right)^{2} } right)^{{{raise0.7exhbox{$3$} !mathord{left/ {vphantom {3 2}}right.kern-0pt} !lower0.7exhbox{$2$}}}} }}$$
(2)
where N is the number of recordings, Ut is the input velocity signal and (overline{U}) is the mean velocity. Positive/negative skewness indicates flood/ebb dominance (assuming that flood currents are positive).Figure 2Deployment of the sediment traps, SSSs and devices (electromagnetic current meter EMCM; pressure transducer PT) in the stations (S1 to S4) during spring tide (sketch is exaggerated in the vertical).Full size imageComplementary to the measured DRs, theoretical DRs were also determined from the data, allowing us to link the sediment and flow data collected, and validate the deposition patterns from the traps. The theoretical deposition rate was determined based on45 formula:$${text{DR}} = left{ {begin{array}{*{20}c} {{text{C}}_{{text{b}}} cdot {text{w}}_{{text{s}}} cdot left( {1 – frac{{{uptau }_{{text{b}}} }}{{{uptau }_{{{text{cd}}}} }}} right)} & {{uptau }_{{text{b}}} < {uptau }_{{{text{cd}}}} } \ 0 & {{uptau }_{{text{b}}} ge {uptau }_{{{text{cd}}}} } \ end{array} } right.$$ (3) where Cb is the SSC at the bed, ws is the flock settling velocity, τb is the bed shear stress and τcd is the corresponding critical value for deposition.To determine the settling rate of the flocculates, the modified Stokes’ velocity for cohesive sediment was used, taking shape factors α and β (α = β = 1 for perfectly spherical particles):$${text{w}}_{{text{s}}} = frac{{upalpha }}{{upbeta }} cdot frac{{left( {{uprho }_{{text{s}}} - {uprho }_{{text{w}}} } right) cdot {text{g}} cdot {text{D}}_{50}^{2} }}{{{uprho }_{{text{w}}} cdot 18 cdot {upnu }}}$$ (4) where ρw and ρs are the densities of the water and sediment, respectively and ν is the kinematic viscosity of water (~ 106 m2/s).The bed shear stress τb was calculated from the measured current magnitude, |U| using the law of the wall:$$begin{array}{*{20}c} \ {{uptau }_{{text{b}}} = {uprho }_{{text{w}}} cdot {text{u}}_{*}^{2} , {text{u}}_{*} = frac{left| U right| cdot kappa }{{ln left( {{raise0.7exhbox{$z$} !mathord{left/ {vphantom {z {z_{0} }}}right.kern-0pt} !lower0.7exhbox{${z_{0} }$}}} right)}} } \ end{array} { }$$ (5) where κ is the von Kármán constant (~ 0.4) and z0 is the roughness length. For Zostera noltei, the roughness length was estimated at 5 mm46, value that was also used in the other stations, in lack of related estimate for marsh plants.The critical shear for deposition, τcd, was calculated using the formula47:$$sqrt {frac{{{uptau }_{{{text{cd}}}} }}{{{uprho }_{{text{w}}} }}} = left{ {begin{array}{*{20}c} {0.008} & {{text{w}}_{{text{s}}} le 5 cdot 10^{ - 5} {text{m}}/{text{s}}} \ {0.094 + 0.02 cdot {text{log}}_{10} left( {{text{w}}_{{text{s}}} } right)} & {3 cdot 10^{ - 4} le {text{w}}_{{text{s}}} le 5 cdot 10^{ - 5} {text{m}}/{text{s}}} \ {0.023} & {{text{w}}_{{text{s}}} ge 3 cdot 10^{ - 4} {text{m}}/{text{s}}} \ end{array} } right.$$ (6) Theoretical values of minimum SSCs needed for these DRs were also calculated, assuming that there is constant deposition (i.e., setting τb = 0), and compared with the field results. More

Study areaThe Qinghai–Tibet Plateau (26°00′-39°47′N, 73°19′-104°47′E), one of the most important pastoral areas in the world, straddles the southwest regions of China, and it includes 244 counties, which belong to six provinces: Tibet, Qinghai, Xinjiang, Gansu, Sichuan, and Yunnan. It is characterized by rich natural grassland resources, including desert steppes, alpine steppes, and alpine meadows (Fig. 1a). The grassland areas account for over 56% of this region34. The grassland plays a vital role in providing regional and national animal husbandry products and fodder35, which enables the local herders to obtain almost all of the resources required for survival36. The grazing density distribution is extremely unbalanced (Fig. 1a) owing to the high spatial heterogeneity of economic development (Fig. 1b-1) and grassland production (Fig. 1b-2), resulting from the differences in resources and environmental factors37. Over the past few decades, there has been a significant change in the number of livestock animals, and the number of sheep exceeded 160 million by 2020. Therefore, it is urgent to obtain a high-resolution gridded grazing dataset for its evaluating spatiotemporal changes and coordinating the relationship between human beings and the grassland ecosystem.Fig. 1Location of the Qinghai–Tibet Plateau: (a) grassland type and distribution, and grazing density (GD) in 244 counties; (b) spatial heterogeneity of economic development (ED) and grassland production (GP) in 244 counties. GD, ED, and GP are represented by sheep unit per grassland area per county (SU/hm2), human footprint index per pixel (HF/pixel) per county, and net primary production per grassland area per county (gC/m2), respectively.Full size imageFig. 2Methodological framework for grazing spatialization.Full size imageMethodological frameworkWe developed a methodological framework for high-resolution gridded grazing dataset mapping. The framework mainly includes four parts: (i) identifying features affecting grazing, (ii) extracting theoretical suitable grazing areas, (iii) building grazing spatialization model, and (iv) correcting the grazing spatialization dataset. Each step is explained in more detail below (Fig. 2).Step 1: Identifying features affecting grazingGrazing activities are affected by the spatial heterogeneity of resources and environmental factors, regulated by the grazing behavior of herders and the foraging behavior of herds, and restricted by ecological protection policies. Therefore, the specific implications of the 14 influencing factors from the above four aspects are presented in Table 1. These factors are necessary for spatializing the county-level grazing data.Table 1 The identified features affecting grazing.Full size tableStep 2: Extracting theoretical suitable grazing areasThe decision tree approach38 was adopted to extract the theoretical suitable grazing areas for further grazing spatialization (step 2 in Fig. 2). First, the potential grazing area was identified according to the boundary of the grassland ecosystem, because grazing behavior only occurs in the grassland. Then, the unsuitable areas for grazing, i.e., extremely-high-altitude areas and areas adjacent to towns, were removed from the potential grazing area stepwise. The areas strictly prohibited for grazing, i.e., the core areas of national nature reserves39 within grassland areas, were also deemed unsuitable for grazing. Finally, the extracted areas were the theoretically suitable grazing areas.Step 3: Building grazing spatialization model(i) Extracting cross-scale feature (CSFs)In the traditional method, the spatial resolution of the training data (i.e., the average value at the administrative level) differs from that of the predicting data (i.e., the value at the pixel level), and the trained model can only capture the characteristics within the training data. However, the extreme value of the predicting data inevitably exceeds the range of the training data, which can result in underestimation in these parts40. To reduce these mismatches, we built an improved method for CSFs extraction (Fig. 2, first part of step 3).First, the census grazing data are simply distributed from county level to pixel level using the weight of the absolute disturbance (AD) index as Eq. (1). The AD index is measured by Mahalanobis distance using Eq. (2), which is calculated according to the deviation between the potential and observed normalized difference vegetation index (NDVI) values22. Second, the distributed grazing data are graded via the hierarchical clustering method, and the optimal number of the group can be determined using the Davies–Bouldin index (DBI)41 as Eq. (3), an index for evaluating the quality of clustering algorithm. The smaller the DBI, the smaller the distance within each group. Therefore, the DBI can be used to select the best similar values to minimize the deviation within each group. Finally, we can obtain the scope of the groups within each county using the above two steps and obtain the average value of all independent variables and the dependent variable accordingly. As expected, we can decompose the average value at the county level (traditional features in Fig. 2) into the average value at the group level (improved features in Fig. 2).$$S{U}_{i}=S{U}_{j}^{C}frac{{w}_{A{D}_{i}}}{{w}_{A{D}_{j}}}$$
(1)
where SUi and (S{U}_{j}^{C}) are the grazing value for pixel i and the census grazing value for county j; ({w}_{A{D}_{i}}) is the weight of the AD index for pixel i and ({w}_{A{D}_{j}}) represents the summed weight of the AD index values for all pixels in county j.$$begin{array}{cll}A{D}_{i} & = & sqrt{{({D}_{i}-u)}^{T}co{v}^{-1}({D}_{i}-u)}\ {D}_{i} & = & NDV{I}_{i}^{T}-NDV{I}_{i}^{P}end{array}$$
(2)
where ADi is the AD index for pixel i; the vector composed of its observed NDVI (left(NDV{I}_{i}^{T}right)) and potential NDVI (left(NDV{I}_{i}^{P}right)) time-series data could be considered as two points in the feature space for pixel i, and Di and u are the difference and the mean value of the vector, respectively; cov is the covariance matrix.$$DB{I}_{k}=frac{1}{k}{sum }_{x=1}^{k}ma{x}_{yne x}left(frac{overline{{a}_{x}}+overline{{a}_{y}}}{left|{delta }_{x}-{delta }_{y}right|}right)$$
(3)
where DBIk is the DBI coefficient when the cluster number is k; (overline{{a}_{x}}) and (overline{{a}_{y}}) are the average distances of the group xth and the group yth, respectively; δx and δy are the center distance of the group xth and the group yth, respectively.Different from the traditional method, our method can decompose features into multiple features using the grading AD index. The differences among counties will not be easily averaged out. Moreover, our method is less affected by scale mismatch and can be transferred to cross-scale modeling26.(ii) Building RF model with partitioningA single model cannot accurately obtain the variation information of the Qinghai–Tibet Plateau with high spatial heterogeneity. The partition model, a widely used method for estimating population distribution and others42,43, can be incorporated into the proposed model to improve its performance. The thresholds (0.43, 0.35 and 0.21 SU/hm2), determined according to the theoretical livestock carrying capacity (equation S1), were calculated and used to separate independent variables and dependent variable for each grassland types: alpine meadow, alpine steppe and alpine desert steppe (see Section 6.1 for details). Then, the RF models were established, and the training and testing samples were randomly divided in the proportion of 3:1. It is notable that transforming the response variable using natural log prior to RF model fitting is necessary to achieve higher prediction accuracies44. Finally, the independent variables at the pixel level were inputted into the two trained RF models, and the corresponding grid grazing dataset was output by combining the two results (Fig. 2, second part of step 3).(iii) Validating the accuracy of the methodsThe performance of the grazing spatialization model was evaluated through a comparison of the predicted value with census value26. Accuracy validation indexes, including coefficients of determination (R2), root mean square error (RMSE), and mean absolute error (MAE), were used to evaluate the performances of the proposed RF-based models (Table 2), as presented in Eq. (4).$$begin{array}{ccc}{R}^{2} & = & 1-frac{{sum }_{j=1}^{N}{left(S{U}_{j}^{C}-S{U}_{j}^{P}right)}^{2}}{{sum }_{j=1}^{N}{left(S{U}_{j}^{C}-overline{S{U}^{C}}right)}^{2}}\ RMSE & = & sqrt{frac{{sum }_{j=1}^{N}{left(S{U}_{j}^{C}-S{U}_{j}^{P}right)}^{2}}{N}}\ MAE & = & frac{{sum }_{j=1}^{N}| S{U}_{j}^{C}-S{U}_{j}^{P}| }{N}end{array}$$
(4)
where (S{U}_{j}^{C}) and (S{U}_{j}^{P}) are the census grazing value and the predicted grazing value for county j, respectively; (overline{S{U}^{C}}) is the average census data for all counties; and N is the number of all counties.Table 2 The proposed methods and their descriptions.Full size tableStep 4: Correcting grazing spatialization dataset(i) Correcting residuals of datasetCorrecting residuals is necessary to obtain datasets with higher accuracy45,46, because propagating the cross-scale relationship in the RF models will inevitably generate errors47. The residuals, calculated by the difference between the average census grazing and predicted grazing values at the administrative level, were used to calibrate the errors related to all pixels within this county. The revised dataset after residual correction is the final product provided in this study. The residual correction method is expressed by Eq. (5), and the process is shown in the fourth step in Fig. 2.$$S{U}_{i}^{RP}=S{U}_{i}^{P}+{R}_{j}$$
(5)
where (S{U}_{i}^{RP}) denotes the predicted grazing value revised by the residuals for pixel i, (S{U}_{i}^{P}) denotes the predicted grazing for pixel i, and Rj denotes the residuals calculated from the difference between census grazing and predicted grazing data for county j.(ii) Validating the accuracy of datasetTwo goodness-of-fit indexes were used to validate the consistency of spatial distribution and the temporal trend between predicted grazing data and census grazing data. Generally, the coefficient of determination (R2), defined in Eq. (4), is used to verify the consistency of spatial distribution, and the Nash–Sutcliffe efficiency (NSE, Eq. (6)) is used to verify the consistency of temporal trend. An index value closer to 1 corresponds to a more accurate dataset. Meanwhile, we also collected field grazing data from 56 sites to further validate the spatial accuracy of the dataset, and it measured using the R2 in Eq. (4).$$NSE=1-frac{{sum }_{t=1}^{T}{left(S{U}_{t}^{RP}-S{U}_{t}^{C}right)}^{2}}{{sum }_{t=1}^{T}{left(S{U}_{t}^{C}-overline{S{U}^{{C}^{{prime} }}}right)}^{2}}$$
(6)
where (S{U}_{t}^{RP}) and (S{U}_{t}^{C}) are the predicted grazing value revised by residuals and the census grazing value of all counties in year t, respectively; (overline{S{U}^{{C}^{{prime} }}}) is the average census grazing value of all years; and T is the number of time steps.(iii) Identifying uncertainties associated with datasetThe uncertainties associated with the dataset originate from the following two aspects: First, the unreasonableness of our method, owing to the errors related to cross-scale modeling or the inappropriate selection of influencing factors, is an important source of uncertainties. Second, the incompleteness of auxiliary variables also introduces uncertainties. In this instance, grassland-free areas are not accurately identified in some counties, but livestock animals are raised in these counties. These counties have no effective value for livestock density prediction. Overall, the uncertainties can be identified in terms of the mean relative error (MRE) in Eq. (7).$$MRE=frac{{sum }_{j=1}^{N}left|frac{S{U}_{j}^{C}-S{U}_{j}^{RP}}{S{U}_{j}^{C}}right|}{N}ast 100 % $$
(7)
where (S{U}_{j}^{C}) is the census grazing value for county j, (S{U}_{j}^{RP}) is the predicted grazing value revised by residuals for county j, and N is the number of counties.Data sourceCensus grazing data at county levelEight types of livestock, namely cattle, yaks, horses, donkeys, mules, camels, goats, and sheep, were considered according to the regional characteristics, and livestock stocking quantity at the end of year for each county can be determined from statistical yearbooks. However, the numbers of livestock at the county level for some years between 1982 and 2015 were not recorded. The missing data were indirectly approximated from city- or provincial-level data (e.g., interpolation using their temporal trends). Each type of livestock stocking quantity was converted into standard sheep unit (SU) according to the national standards using Eq. (8)48, namely the calculation of rangeland carrying capacity (NY/T 635-2015). Of the 244 counties of the Qinghai–Tibet Plateau, only 242 counties were considered, as the census grazing data for the other 2 counties were unavailable. The unit of grazing statistics data at the county level is defined as SU per county per year (SU·county−1·year−1).$$begin{array}{l}SU={N}_{sheep}+0.8times {N}_{goats}+5times {N}_{cattle}+5times {N}_{yaks+}+\ 6times {N}_{horses}+3times {N}_{donkeys}+6times {N}_{mules}+7times {N}_{camels}end{array}$$
(8)
where Nsheep, Ngoats, Ncattle, Nyaks, Nhorses, Ndonkeys, Nmules, Ncamels are the number of sheep, goats, cattle, yaks, horses, donkeys, mules, and camels at the year-end, respectively. SU denotes the standard sheep unit (SU·county−1·year−1).Data of grazing influencing factors at pixel levelThe types of features affecting grazing were obtained from the first step described in Methods, and the detailed information, such as original spatiotemporal resolution, format, and source, is shown in Table 3. The format (i.e., GeoTIFF), spatial resolution (i.e., 0.083°), and the number of rows and columns of the gridded features were leveraged to further produce a high-resolution grazing dataset.Table 3 Data source of grazing influence factors.Full size table More

Arunrat, N., Sereenonchai, S., Chaowiwat, W. & Wang, C. Climate change impact on major crop yield and water footprint under CMIP6 climate projections in repeated drought and flood areas in Thailand. Sci. Total Environ. 807, 150741 (2022).ADS 
CAS 

Google Scholar 
Chandio, A. A., Shah, M. I., Sethi, N. & Mushtaq, Z. Assessing the effect of climate change and financial development on agricultural production in ASEAN-4: the role of renewable energy, institutional quality, and human capital as moderators. Environ. Sci. Pollut. Res. 29, 13211–13225 (2022).
Google Scholar 
Masood, N., Akram, R., Fatima, M., Mubeen, M., Hussain, S., Shakeel, M., Khan, N., Adnan, M., Wahid, A., Shah, A. N. and Ihsan, M. Z. (2022) Insect pest management under climate change. In Building climate resilience in agriculture. Springer, ChamOzdemir, D. The impact of climate change on agricultural productivity in Asian countries: A heterogeneous panel data approach. Environ. Sci. Pollut. Res. 29, 8205–8217 (2022).
Google Scholar 
Aidoo, O. F. et al. Climate-induced range shifts of invasive species (Diaphorina citri Kuwayama). Pest Manag. Sci. 78, 2534–2549 (2022).CAS 

Google Scholar 
Hebbar, K. B. et al. Predicting the Potential Suitable Climate for Coconut (Cocos nucifera L.) Cultivation in India under Climate Change Scenarios Using the MaxEnt Model. Plants. 11, 731 (2022).
Google Scholar 
Martín-Vélez, V. & Abellán, P. Effects of climate change on the distribution of threatened invertebrates in a Mediterranean hotspot. Insect Conserv. Divers. 15, 370–379 (2022).
Google Scholar 
Williams, J. J., Freeman, R., Spooner, F. & Newbold, T. Vertebrate population trends are influenced by interactions between land use, climatic position, habitat loss and climate change. Glob. Chang. Biol. 28, 797–815 (2022).CAS 

Google Scholar 
Aidoo, O. F. et al. Lethal yellowing disease: insights from predicting potential distribution under different climate change scenarios. J. Plant Dis. Prot. 128, 1313–1325 (2021).
Google Scholar 
Sofaer, H. R. et al. Development and delivery of species distribution models to inform decision-making. Bioscience 69, 544–557 (2019).
Google Scholar 
Mead FW, The Asiatic citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae). Florida Department of Agriculture Conservation Service, Division of Plant Industry Entomological Circular No. 180.Bové, J. M. Huanglongbing: A destructive, newly-emerging, century-old disease of citrus. Plant Pathol. J. 1, 7–37 (2006).
Google Scholar 
Li, S., Wu, F., Duan, Y., Singerman, A. & Guan, Z. Citrus greening: Management strategies and their economic impact. HortScience 55, 604–612 (2020).
Google Scholar 
Jia, H. et al. Genome editing of the disease susceptibility gene Cs LOB 1 in citrus confers resistance to citrus canker. Plant Biotechnol. J. 15, 817–823 (2017).CAS 

Google Scholar 
Ehsani, R., Dewdney, M. & Johnson, E. Controlling HLB with thermotherapy: What have we learned so far?. Citrus Ind. News 9, 26–28 (2016).
Google Scholar 
Spreen, T. H., Baldwin, J. P. & Futch, S. H. An economic assessment of the impact of Huanglongbing on citrus tree plantings in Florida. J. Hortic. Sci. 49, 1052–1055 (2014).
Google Scholar 
Djeddour, D., Pratt, C., Constantine, K., Rwomushana, I. and Day, R., (2021) The Asian citrus greening disease (Huanglongbing). Evidence note on invasiveness and potential economic impacts for East Africa. CABI Working Paper, 24, 94Hu, J., Jiang, J. & Wang, N. Control of citrus Huanglongbing via trunk injection of plant defense activators and antibiotics. Phytopathology 108, 186–195 (2018).CAS 

Google Scholar 
Fan, G. C. et al. Evaluation of thermotherapy against Huanglongbing (citrus greening) in the greenhouse. J. Integr. Agric. 15, 111–119 (2016).
Google Scholar 
Nguyen, V. A., Bartels, D. & Gilligan, C. Modelling the spread and mitigation of an emerging vector-borne pathogen: citrus greening in the US. Biorxiv https://doi.org/10.1101/2022.05.04.490566 (2022).Article 

Google Scholar 
Milosavljević, I. et al. Post-release evaluation of Diaphorencyrtus aligarhensis (Hymenoptera: Encyrtidae) and Tamarixia radiata (Hymenoptera: Eulophidae) for biological control of Diaphorina citri (Hemiptera: Liviidae) in Urban California, USA. Agronomy 12, 583 (2022).
Google Scholar 
Maluta, N., Castro, T. & Lopes, J. R. S. Entomopathogenic fungus disrupts the phloem-probing behavior of Diaphorina citri and may be an important biological control tool in citrus. Sci. Rep. 12, 1–10 (2022).
Google Scholar 
Hall, D. G., Richardson, M. L., Ammar, E. D. & Halbert, S. E. Asian citrus psyllid, Diaphorina citri, vector of citrus huanglongbing disease. Entomol. Exp. Appl. 146, 207–223 (2013).
Google Scholar 
Vázquez-García, M. et al. Insecticide resistance in adult Diaphorina citri Kuwayama1 from lime orchards in central west Mexico. Southwest. Entomol. 38, 579–596 (2013).
Google Scholar 
Naeem, A., Freed, S., Jin, F. L., Akmal, M. & Mehmood, M. Monitoring of insecticide resistance in Diaphorina citri Kuwayama (Hemiptera: Psyllidae) from citrus groves of Punjab Pakistan. Crop Prot. 86, 62–68 (2016).CAS 

Google Scholar 
Hulme, P. E. et al. Grasping at the routes of biological invasions: A framework for integrating pathways into policy. J. Appl. Ecol. 45, 403–414 (2008).
Google Scholar 
Oke, A. O., Oladigbolu, A. A., Kunta, M., Alabi, O. J. & Sétamou, M. First report of the occurrence of Asian citrus psyllid Diaphorina citri (Hemiptera: Liviidae), an invasive species in Nigeria. West Africa. Sci. Rep. 10, 1–8 (2020).
Google Scholar 
Tang, Y.Q. (1990) On the parasite complex of Diaphorina citri Kuwayama (Homoptera: Psyllidae) in Asian-Pacific and other areas. In proceedings 4th international conference on citrus rehabilitation, Chiang Mai, Thailand. 4: 240 245Chien, C. C., Chiu, S. C. & Ku, S. C. Biological control of Diaphorina citri in Taiwan. Fruits 44, 401–407 (1989).
Google Scholar 
Hoddle, M. S. Foreign exploration for natural enemies of Asian citrus psyllid, Diaphorina citri (Hemiptera: Psyllidae), in the Punjab of Pakistan for use in a classical biological control program in California USA. Pakistan Entomol. 34, 1–5 (2012).
Google Scholar 
Étienne, J., Quilici, S., Marival, D., Franck, A. & Gonzalez Fernandez, C. Biological control of Diaphorina citri (Hemiptera: Psyllidae) in Guadeloupe by imported Tamarixia radiata (Hymenoptera: Eulophidae). Fruits 56, 307–315 (2001).
Google Scholar 
Qureshi, J. A., Rogers, M. E., Hall, D. G. & Stansly, P. A. Incidence of invasive Diaphorina citri (Hemiptera: Psyllidae) and its introduced parasitoid Tamarixia radiata (Hymenoptera: Eulophidae) in Florida citrus. J. Econ. Entomol. 102, 247–256 (2009).
Google Scholar 
Chen, X., Triana, M. & Stansly, P. A. Optimizing production of Tamarixia radiata (Hymenoptera: Eulophidae), a parasitoid of the citrus greening disease vector Diaphorina citri (Hemiptera: Psylloidea). Biol. Control. 105, 13–18. https://doi.org/10.1016/j.biocontrol.2016.10.010 (2017).Article 

Google Scholar 
Kistner, E. J., Amrich, R., Castillo, M., Strode, V. & Hoddle, M. S. Phenology of Asian citrus psyllid (Hemiptera: Liviidae), with special reference to biological control by Tamarixia radiata, in the residential landscape of southern California. J. Econ. Entomol. 109, 1047–1057. https://doi.org/10.1093/jee/tow021 (2016).Article 

Google Scholar 
Ramos Aguila, L. C. et al. Temperature-dependent biological control effectiveness of Tamarixia radiata (Hymenoptera: Eulophidea) under laboratory conditions. J. Econ. Entomol. 114, 2009–2017 (2021).
Google Scholar 
Ramos Aguila, L. C. et al. Temperature-dependent demography and population projection of Tamarixia radiata (Hymenoptera: Eulophidea) reared on Diaphorina citri (Hemiptera: Liviidae). J. Econ. Entomol. 113, 55–63 (2020).
Google Scholar 
Ashraf, H. J. et al. Comparative microbiome analysis of Diaphorina citri and its associated parasitoids Tamarixia radiata and Diaphorencyrtus aligarhensis reveals Wolbachia as a dominant endosymbiont. Environ. Microbiol. 24, 1638–1652 (2022).CAS 

Google Scholar 
Chow, A. & Sétamou, M. Parasitism of Diaphorina citri (Hemiptera: Liviidae) by Tamarixia radiata (Hymenoptera: Eulophidae) on residential citrus in Texas: Importance of colony size and instar composition. Biol. Control 165, 104796 (2022).
Google Scholar 
Ajene, I. J. et al. Habitat suitability and distribution potential of Liberibacter species (“Candidatus Liberibacter asiaticus” and “Candidatus Liberibacter africanus”) associated with citrus greening disease. Environ. Microbiol. 26, 575–588 (2020).
Google Scholar 
Shabani, F., Kumar, L. & Ahmadi, M. A comparison of absolute performance of different correlative and mechanistic species distribution models in an independent area. Ecol. Evol. 6, 5973–5986 (2016).
Google Scholar 
Kearney, M. & Porter, W. Mechanistic niche modelling: Combining physiological and spatial data to predict species’ ranges. Ecol 12, 334–350 (2009).
Google Scholar 
Byeon, D. H., Jung, S. & Lee, W. H. Review of CLIMEX and MaxEnt for studying species distribution in South Korea. J. Asia-Pac. Biodivers. 1, 325–333 (2018).
Google Scholar 
Kriticos, D. J., Yonow, T. & McFadyen, R. E. The potential distribution of Chromolaena odorata (Siam weed) in relation to climate. Weed Res 45, 246–254 (2005).
Google Scholar 
Wharton, T. N. & Kriticos, D. J. The fundamental and realized niche of the Monterey pine aphid, Essigella californica (Essig) (Hemiptera: Aphididae): implications for managing softwood plantations in Australia. Divers. Distrib. 10, 253–262 (2004).
Google Scholar 
Sutherst, R., Maywald, G. and Kriticos, D., CLIMEX version 3: user’s guide. (2007).Ramirez-Cabral, N. Y., Kumar, L. & Shabani, F. Global alterations in areas of suitability for maize production from climate change and using a mechanistic species distribution model (CLIMEX). Sci. Rep. 7, 1–3 (2017).CAS 

Google Scholar 
McCalla, K. A., Keçeci, M., Milosavljević, I., Ratkowsky, D. A. & Hoddle, M. S. The influence of temperature variation on life history parameters and thermal performance curves of Tamarixia radiata (Hymenoptera: Eulophidae), a parasitoid of the Asian citrus psyllid (Hemiptera: Liviidae). J. Econ. Entomol. 112, 1560–1574 (2019).
Google Scholar 
Gonzalez-Cabrera, J., Moreno-Carrillo, G., Sanchez-Gonzalez, J. A. & Bernal, H. C. Natural and augmented parasitism of tamarixia radiata (Hymenoptera Eulophidae) in Urban Areas of western Mexico. Entomol. Sci. 53, 486–492. https://doi.org/10.18474/JES17-112.1 (2018).Article 

Google Scholar 
Chavez, Y. et al. Tamarixia radiata (Waterston) and Cheilomenes sexmaculata (Fabricius) as biological control agents of Diaphorina citri Kuwayama in Ecuador. Chil. J. Agric. Res. 77, 180–184. https://doi.org/10.4067/S0718-58392017000200180 (2017).Article 

Google Scholar 
Flores, D. & Ciomperlik, M. Biological control using the ectoparasitoid, Tamarixia radiata, against the Asian citrus psyllid, Diaphorina citri, in the lower Rio Grande valley of Texas. Southwest. Entomol. 42, 49–59. https://doi.org/10.3958/059.042.0105 (2017).Article 

Google Scholar 
Parra, J. R., Alves, G. R., Diniz, A. J. & Vieira, J. M. Tamarixia radiata (Hymenoptera: Eulophidae) × Diaphorina citri (Hemiptera: Liviidae): Mass rearing and potential use of the parasitoid in Brazil. J. Integr. Pest. Manag. https://doi.org/10.1093/jipm/pmw003 (2016).Article 

Google Scholar 
Diniz, A. J. F., Otimização da criação de Diaphorina citri Kuwayama, 1908 (Hemiptera: Liviidae) e de Tamarixia radiata (Waterston, 1922) (Hymenoptera: Eulophidae), visando a produção em larga escala do parasitoide e avalliação do seu estabelecimento em campo. Tese (Doutorado em Entomologia)—Escola Superior de Agricultura “Luiz de Queiroz”, Universidade de São Paulo, São Paulo. (2013)Hoddle, M. S. & Pandey, R. Host range testing of Tamarixia radiata (Hymenoptera: Eulophidae) sourced from the Punjab of Pakistan for classical biological control of Diaphorina citri (Hemiptera: Liviidae: Euphyllurinae: Diaphorinini) in California. J. Econ. Entomol. 107, 125–136. https://doi.org/10.1603/EC13318 (2014).Article 

Google Scholar 
Gómez-Torres, M. L., Nava, D. E. & Parra, J. R. Thermal hygrometric requirements for the rearing and release of Tamarixia radiata (Waterston) (Hymenoptera, Eulophidae). Rev. Bras. Entomol. 58, 291–295. https://doi.org/10.1590/S0085-56262014000300011 (2014).Article 

Google Scholar 
Gómez-Torres, M. L., Nava, D. E. & Parra, J. R. Life table of Tamarixia radiata (Hymenoptera: Eulophidae) on Diaphorina citri (Hemiptera: Psyllidae) at different temperatures. J. Econ. Entomol. 105, 338–343 (2012).
Google Scholar 
Chong, J. H., Roda, A. L. & Mannion, C. M. Density and natural enemies of the Asian Citrus Psyllid, Diaphorina citri (Hemiptera: Psyllidae), in the residential landscape of Southern Florida. J. Agric. Urban Entomol. 27, 33–49. https://doi.org/10.3954/11-05.1 (2010).Article 

Google Scholar 
Pluke, R. W., Qureshi, J. A. & Stansly, P. A. Citrus flushing patterns, Diaphorina citri (Hemiptera: Psyllidae) populations and parasitism by Tamarixia radiata (Hymenoptera: Eulophidae) in Puerto Rico. Florida Entomol. 91, 36–42 (2008).
Google Scholar 
Ashraf, H. J. et al. Genetic diversity of Tamarixia radiata populations and their associated endosymbiont Wolbachia species from China. Agronomy 11, 2018 (2021).CAS 

Google Scholar 
Jung, J. M., Lee, W. H. & Jung, S. Insect distribution in response to climate change based on a model: Review of function and use of CLIMEX. Entomol. Res. 46, 223–235 (2016).
Google Scholar 
Kriticos, D. J. et al. CLIMEX Version 4, 184p (2015).
Google Scholar 
Gomez-Marco, F., Gebiola, M., Baker, B. G., Stouthamer, R. & Simmons, G. S. Impact of the temperature on the phenology of Diaphorina citri (Hemiptera: Liviidae) and on the establishment of Tamarixia radiata (Hymenoptera: Eulophidae) in urban areas in the lower Colorado Desert in Arizona. Environ. Entomol. 48, 514–523 (2019).
Google Scholar 
Vieira, J. M. Biologia em temperaturas alternantes e exigências térmicas de Diaphorina citri Kuwayama, 1908 (Hemiptera: Liviidae) e Tamarixia radiata (Waterston, 1922) (Hymenoptera: Eulophidae) visando ao seu zoneamento em regiões citrícolas do estado (Doctoral dissertation, Universidade de São Paulo).Castillo, J., Jacas, J. A., Peña, J. E., Ulmer, B. J. & Hall, D. G. Effect of temperature on life history of Quadrastichus haitiensis (Hymenoptera: Eulophidae), an endoparasitoid of Diaprepes abbreviatus (Coleoptera: Curculionidae). Biol. Control. 36, 189–196 (2006).
Google Scholar 
McFarland, C. D. & Hoy, M. A. Survival of Diaphorina citri (Homoptera: Psyllidae), and its two parasitoids, Tamarixia radiata (Hymenoptera: Eulophidae) and Diaphorencyrtus aligarhensis (Hymenoptera: Encyrtidae), under different relative humidities and temperature regimes. Fla. Entomol. 84, 227–233 (2001).
Google Scholar 
Fauvergue, X. & Quilici, S. Etude de certains parametres de la biologie de Tamarixia radiata (Waterston, 1992)(Hymenoptera: Eulophidae), ectoparasitoide primaire de Diaphorina citri Kuwayama (Hemiptera: Psyllidae) vecteur du greening des agrumes. Paris Fruits 46, 179–179 (1991).
Google Scholar 
Araújo, F. H. et al. Modelling climate suitability for Striga asiatica, a potential invasive weed of cereal crops. Crop Prot. 1(160), 106050 (2022).
Google Scholar 
Silva, D. A. & RS, Kumar L, Shabani F and Picanço MC,. Potential risk levels of invasive Neoleucinodes elegantalis (small tomato borer) in areas optimal for open-field Solanum lycopersicum (tomato) cultivation in the present and under predicted climate change. Pest Manag. Sci 73, 616–627 (2017).
Google Scholar 
Kumar, S., Neven, L. G. & Yee, W. L. Evaluating correlative and mechanistic niche models for assessing the risk of pest establishment. Ecosphere 5, 1–23. https://doi.org/10.1890/ES14-00050.1 (2014).Article 
CAS 

Google Scholar 
Kriticos, D. J. et al. CliMond: global high-resolution historical and future scenario climate surfaces for bioclimatic modelling. Methods Ecol. Evol. 1, 53–64 (2012).
Google Scholar 
Santana Júnior PA, Worldwide spatial distribution of Tuta absoluta (Lepidoptera: Gelechiidae) and its natural enemies under current and future climatic change conditions through modelling. 136 f 2019 (Tese (Doutorado em Fitotecnia) – Universidade Federal de Viçosa, 2019).
Google Scholar 
Kriticos, D. J., Maywald, G. F., Yonow, T., Zurcher, E. J., Herrmann, N. I. and Sutherst, R. W., CLIMEX Version 4: Exploring the effects of climate on plants, animals and diseases. CSIRO, Canberra.156, (2015)Ramos Aguila, L. C. et al. Temperature-dependent demography and population projection of Tamarixia radiata (Hymenoptera: Eulophidea) reared on Diaphorina citri (Hemiptera: Liviidae). J. Econ. Entomol. 113, 55–63 (2019).
Google Scholar 
Oliveira, R. C., Modelagem de nicho ecológico para Helicoverpa punctigera (Wallengren, 1860) (Lepidoptera: Noctuidae) no mundo: Potencial invasão e riscos diante das mudanças climáticas. (2021). http://www.repositorio.ufc.br/handle/riufc/61961Bazzocchi, G. G., Lanzoni, A., Burgio, G. & Fiacconi, M. R. Effects of temperature and host on the pre-imaginal development of the parasitoid Diglyphus isaea (Hymenoptera: Eulophidae). Biol. Control 26, 74–82 (2003).
Google Scholar 
Hondo, T., Koike, A. & Sugimoto, T. Comparison of thermal tolerance of seven native species of parasitoids (Hymenoptera: Eulophidae) as biological control agents against Liriomyza trifolii (Diptera: Agromyzidae) in Japan. Appl. Entomol. Zool. 41, 73–82 (2006).
Google Scholar 
Duale, A. Effect of temperature and relative humidity on the biology of the stem borer parasitoid Pediobius furvus (Gahan) (Hymenoptera: Eulophidae) for the management of stem borers. Environ. Entomol. 34, 1–5 (2005).
Google Scholar 
Ashraf, H. J. et al. Comparative transcriptome analysis of Tamarixia radiata (Hymenoptera: Eulophidae) reveals differentially expressed genes upon heat shock. Comp. Biochem. Physiol. D: Genom. Proteom. 41, 100940 (2022).CAS 

Google Scholar 
van Doan, C. et al. Natural enemies of herbivores maintain their biological control potential under short-term exposure to future CO2, temperature, and precipitation patterns. Ecol. Evol. 11, 4182–4192 (2021).
Google Scholar 
Thomson, L. J., Macfadyen, S. & Hoffmann, A. A. Predicting the effects of climate change on natural enemies of agricultural pests. Biol. Control. 52, 296–306 (2010).
Google Scholar 
Rosenblatt, A. E. & Schmitz, O. J. Climate change, nutrition, and bottom-up and top-down food web processes. Trends Ecol. Evol. 31, 965–975 (2016).
Google Scholar 
Aidoo, O. F. et al. A machine learning algorithm-based approach (MaxEnt) for predicting invasive potential of Trioza erytreae on a global scale. Ecol. Inform. 71, 101792 (2022).
Google Scholar 
Aidoo, O. F. et al. The Impact of Climate Change on Potential Invasion Risk of Oryctes monoceros Worldwide. Front. Ecol. Evol. https://doi.org/10.3389/fevo.2022.895906 (2022).Article 

Google Scholar 
Hao, M. et al. Global potential distribution of Oryctes rhinoceros, as predicted by Boosted Regression Tree model. Glob. Ecol. Conserv. 1(37), e02175 (2022).
Google Scholar 
Aidoo, O. F. et al. Model-based prediction of the potential geographical distribution of the invasive coconut mite, Aceria guerreronis Keifer (Acari: Eriophyidae) based on MaxEnt. Agric. For. Entomol. 24, 390–404 (2022).
Google Scholar  More

During the COVID-19 pandemic, behavioral restrictions were imposed, after which various health problems were reported in many countries45,46. The pandemic has also increased food insecurity worldwide; consequently, panic buying has been observed in many countries, including Japan47. However, even in such situations, we found that diversity in local food access, ranging from self-cultivation to direct-to-consumer sales, was significantly associated with health and food security variables. Specifically, our results revealed the following five key discussion points.Urban agriculture in walkable neighborhoods bore fruit for health and food system resilience. However, the magnitude of its contribution differed depending on the type of urban agricultureThe results of this study showed that those who grew food by themselves at allotment farms and home gardens had significantly better subjective well-being and physical activity levels than those who did not. This result is in line with previous studies conducted during times free from the impact of infectious disease pandemics38,39,40. The use of direct sales was not related to subjective well-being but was significantly associated with physical activity. The reason might be that farm stand users tend to live in areas with farmland and travel to purchase fruits and vegetables at farm stands on foot or by bicycle. This result is consistent with that of a previous study demonstrating that the food environment in neighborhoods is an important component in promoting physical activity17.Our results also showed that those who grew food by themselves at allotment farms and those who purchased local foods at farm stands were significantly less anxious about the availability of fresh food both during the state of emergency and in the future than their counterparts. In contrast, home garden users showed significant differences only for the state of emergency. This result might be due to the differences in the size and yield of cultivation at allotment farms and home gardens. One lot in allotment farms in Tokyo can produce as much as or more than the average annual vegetable consumption per household in Japan48. However, home gardens are generally smaller and produce limited fresh foods for consumption, which may have influenced food security concerns.As in other countries, Japan imports much food from overseas and is deeply integrated into the large-scale global food system. However, as shown in this study, urban agriculture in Japanese suburbs forms small-scale, decentralized, and community-based local food systems. This multilayered food system can complement the disruptions and shortages of the global system when various problems occur for climatic, sociopolitical, or other reasons, such as pandemics. In fact, our empirical evidence suggests that urban agriculture in walkable neighborhoods, particularly allotment farms and direct-to-consumer sales at farm stands, contributed to the mitigation of food security concerns in neighborhood communities. This means that urban agriculture could enhance the resilience of the urban food system at a time when the global food system has been disrupted due to a pandemic. This validates recent discussions about the potential of urban agriculture to facilitate food system resilience10. Furthermore, our findings imply that the types of urban agriculture employed matter in determining the degree of contribution to food system resilience.To summarize the overall results, urban agriculture in walkable neighborhoods bore fruit for health and food system resilience during the COVID-19 pandemic. However, different types of urban agriculture exhibited varying associations with health and resilience. Allotment farms were positively related to all of the following: subjective well-being, physical activity, and food security concerns, both during the state of emergency and in the future. Home gardens were positively related to subjective well-being, physical activity, and food security concerns only during the state of emergency. Farm stands were positively related to physical activity and food security concerns both during the state of emergency and in the future.These differences may be due to the characteristics of the respective spaces. It is suggested that this diversity of urban agriculture has led to different types of people benefiting from various kinds of urban agriculture. Allotment farms were found to be associated with high subjective well-being, physical activity, and food security, but they may not be feasible for those who do not have enough physical strength because users are responsible for cultivating their lots, which measure 10–30 square meters40. In contrast, home gardens can be created even by those who are not confident in their physical strength. In fact, our study showed that women and older people engaged in home gardening more than men and younger people. In addition, direct-to-consumer sales at farm stands are the easiest way to obtain local fresh foods for those who do not have the time and space for allotment farms and home gardens. The need for urban agriculture has been argued in many countries2,3. However, little attention has been paid to its scale, accessibility, and diversity. Our study suggests that it is worthwhile to create diverse food production spaces within walkable neighborhoods while considering the diversity of people who access these spaces.Compared to other urban greenery and food retailers, the benefits of urban agriculture on subjective well-being and food security could be greaterCompared to the use of other urban green spaces, including urban parks, our results indicated that self-cultivation at allotment farms and home gardens was more strongly associated with subjective well-being. Previous studies have offered limited perspectives on the differences among various types of urban green spaces33. Our study further suggests that urban parks, allotment farms, and home gardens are differently associated with human health. However, as the reason was not determined, further research is needed.Furthermore, compared to other food retailers, such as supermarkets, convenience stores, and co-op deliveries, allotment farms and farm stands were more strongly associated with less anxiety about fresh food availability in the future. The availability of local fresh foods within walkable neighborhoods might have mitigated food security concerns because residents could grow food by themselves or directly observe farmers’ production processes, which may have made the difference from purchasing at places where the food systems were not visible.Flexibility in work style might promote urban agriculture in walkable neighborhoodsThere was an association between work style—working from home—and access to local food. According to the Ministry of Health, Labor and Welfare (https://www.mhlw.go.jp/english), 52% of Tokyo office workers worked from home during the first emergency declaration. Long commute times and high train congestion rates have been a problem in Tokyo suburbs, but remote workers have gained more time at and around their homes by reducing their commute times, increasing their opportunities to access local food in their walkable neighborhoods. Those who worked from home sought outdoor activities for refreshment and exercise and used a variety of urban green spaces during the pandemic49. Allotment farms and home gardens might be used as such urban green spaces. This result is consistent with previous studies assessing the characteristics of Canadian gardeners during the COVID-19 pandemic28,30.Until now, urban planners and policymakers have rarely taken work style into account. However, the flexibility of work styles and work hours may bring new insights; for example, those who work from home may become important players in urban agriculture. It has been pointed out that cities have a large hidden potential for urban agriculture by cultivating underused lands50. Our study suggests that such underused lands could be converted into productive urban landscapes for remote workers to engage in farming or gardening in between jobs as a hobby or as a side business.Food equity might be improved by urban agriculture in walkable neighborhoodsLocal fresh food is generally considered more expensive than junk food in high-income countries, creating social issues of food inequity. Therefore, past discussions on urban agriculture and food security have focused primarily on low-income households in socioeconomically disadvantaged areas24,25,26.In contrast, our study covered people from all income groups and found no statistically significant relationship between access to local food and income. This finding might be due to two urban cultural backgrounds regarding local food in Tokyo, that is, accessibility and affordability. First, residential segregation by income levels is not noteworthy in Tokyo and people from various income brackets live mixed in the same neighborhoods51. Therefore, most urban residents living in the suburbs have geographically equitable opportunities to access local foods. Second, local foods sold at farm stands are affordable. Prices are almost the same or cheaper than buying food at food retailers. While prices increase because of middleman margins related to shipping in the wholesale market, such increases are unnecessary when selling directly to consumers at farm stands. In addition, the allotment farm lots are not expensive to rent, particularly those operated by local municipalities (Supplementary Note 1).These two backgrounds make local fresh food physically and economically accessible to consumers of all income levels, resulting in food equity. This is particularly important because the concept of food system resilience includes the equitability perspective27.The integration of urban agriculture into walkable neighborhoods is a fruitful wayWhile the current discussion on walkable neighborhoods does not emphasize urban agriculture, our evidence indicated its effectiveness. The concept of walkable neighborhoods (e.g., the 15-min city model) stresses the decarbonization benefit of limiting vehicle travel, as well as the health benefits of promoting walking and cycling13,14,15,16. In addition, our research indicated that urban agriculture in walkable neighborhoods benefited health and well-being by increasing recreational outdoor opportunities to neighborhood communities, including remote workers. It also contributed to food system resilience by providing local foods to all people, including low-income households, when the global food system was disrupted due to the pandemic. Furthermore, recent studies on urban agriculture reported the decarbonization benefit of reducing carbon footprints in food production and distribution7,8. Small-scale and community-based urban agriculture in walkable neighborhoods might especially bring this benefit because neighborhood communities travel to farms on foot or by bicycle, which means almost no emission by distribution. While urban green spaces have various health benefits32,33,34,35, urban agriculture also contributes to food system resilience as well as carbon emission reduction, which makes it unique.Urban agriculture was once considered a failure of urban planning in Japan because it symbolized uncontrolled sprawl. This is analogous to the Western view, as urban agriculture was once considered the ultimate oxymoron1. However, our empirical evidence suggests that the urban‒rural mixture at neighborhood scales is a reasonable urban form that contributes to the resilience of the urban food system and to the health and well-being of neighborhood communities. It is no longer a failure of urban planning but a legacy of urban sprawl in the current urban context.Our study showed that integrating urban agriculture into walkable neighborhoods is a fruitful way of creating healthier cities and developing more resilient urban food systems during times of uncertainty. In cities where there is no farmland in intraurban areas, it would be considered effective to utilize underused spaces such as vacant lots and rooftops as productive urban landscapes. In growing cities where urban areas are still expanding, it would be advantageous to conserve agricultural landscapes within their urban fabrics. Our study could provide referential insights and robust evidence for urban policy to integrate urban agriculture into walkable neighborhoods.This study has potential limitations, including the timing of the survey and the measurement method that was utilized. We conducted the survey between June 4 and 8, 2020, just after the end of the first declaration of a state of emergency by the Japanese government. During this period, the main cultivation activities were planting and growing, and the harvest was just beginning. This seasonal constraint may have influenced the results. Because the survey was conducted during the pandemic, we used subjective methods to measure health and well-being status. However, the results might be different using objective methods52, thus further research is necessary. In addition, a longitudinal study is needed to determine whether the trends observed in this study were specific to the emergency period or whether they will persist after the COVID-19 pandemic. More

Anna Lena Bercht interviewed fishers in Lofoten, Norway, to assess how climate change was affecting their livelihoods.Credit: Anna Lena Bercht

I remember February 2011, when, in the Chinese megacity of Guangzhou, an older man finally overcame his scepticism about being interviewed and invited me to sit down next to him on a stone bench under a shady tree. I held my notebook on my lap, and we sat on either side of a translator and talked about his life and world for more than two hours. It was one of the most informative and revealing interviews that I had done during my fieldwork in the city.
Making it in the megacity
One of the most fundamental challenges in qualitative fieldwork is gaining access to research participants. This is often time-consuming and labour-intensive, particularly when the topic requires in-depth methods and addresses a sensitive subject.Advice that goes beyond the usual recommendations of establishing relationships with gatekeepers, ensuring anonymity for interviewees and relying on the snowball sampling technique (in which one research participant suggests further ones) is rare. In this light, I’m happy to share some simple, but often neglected, examples from my qualitative fieldwork in the lively Guangzhou (where I worked for 12 months)1 and on the remote, Arctic island chain of Lofoten, Norway (done over 4 months)2, that might offer some inspiration and encouragement.I have a background in human geography, and did my PhD on experiences of stress, coping and resilience among the Chinese population of Guangzhou in the face of the city’s rapid urbanization. I travelled there five times to help to establish research cooperation with Chinese scholars, make field observations, select a case-study site and interview locals. I, together with other PhD students, stayed in a typical Chinese high-rise apartment in a neighbourhood that wasn’t a common choice for expatriates. Living side-by-side with the locals gave us a perfect opportunity to experience genuine everyday life and Chinese culture.My first postdoctoral project after my PhD brought me to Lofoten, where I looked at psychological barriers to climate adaptation in small-scale coastal fisheries. I went to Lofoten twice. On my first visit, I travelled across the whole archipelago by bus for one month to get a profound overview of the fishing villages and local living conditions, and to conduct first interviews. During my second visit, I stayed for a total of three months in rental locations near fishing harbours, and conducted more extensive interviews.In both China and Norway, I used in-depth interviews to learn about the challenges that people face. I asked people about unemployment, about the possibility of being forced to move elsewhere and about how climate change might affect their livelihoods. This required a sensitive and thoughtful approach to ‘getting invited’ into people’s lives. In Guangzhou, German- and English-speaking Chinese students assisted me as translators (and interpreters, when needed). On Lofoten, I conducted the interviews myself in English.There are two ways to access research participants: physical access, which refers to the ability of the researcher to get in direct face-to-face contact with people, and mental access. Successful mental access means that interlocutors open up about why they think, feel and behave as they do. Physical access is a necessary condition for mental access; however, in my experience, both are equally valuable.

Chinese interviewees in Guangzou shared their feelings about the rapid urbanization of their city.Credit: Anna Lena Bercht

Compared with Lofoten, it took longer to get physical access to local inhabitants in China. Presumably, this was because of the language barrier and reliance on translators, as well as cultural differences. Trust is considered a central tenet in Chinese relationships, and time and effort are needed to let it grow. During my time in Guangzhou, I occasionally benefited from being a foreigner: people were touched that someone from abroad showed genuine interest in their well-being. In Lofoten, fishers appreciated talking to a social scientist instead of a natural scientist who would have mainly asked questions about fishing quotas and catch volume.My advice for other social scientists hoping to gain access to research participants falls into those two categories.How to get good physical accessUse local public transport. Using local public transport creates many unexpected opportunities to bump into people, get into conversations and gain relevant information. For example, while waiting at a bus stop in Lofoten, I came across an art-gallery owner from a fishing village. He wondered why I was travelling out of the peak tourism season. I ended up with an invitation to his gallery, where he introduced me to two retired fishers whom he had also invited. Without the gallerist and his proactive networking, I probably would not have been given the chance to interview these two very informative and engaging fishers.In a metro station in Guangzhou, a toddler kept staring at me and tried to touch my light hair. This small interaction led me to chat to the toddler’s father, who recommended that I talk to a local teacher to learn more about the area’s history. His advice opened up important insights into urban-restructuring processes that I would have missed otherwise.
Nine ‘brain food’ tips for researchers
Use local media. In Norway, a journalist was at the harbour to get first-hand information on the year’s cod catch, when he saw me interviewing fishers. He became curious and eager to learn more about my work. In the end, he wrote an article about my research, which was published a few days later across Lofoten. His article was a door-opener for me.People recognized me from my photo in the article and contacted me to tell me about their lives and the cod fisheries. They also invited me on their vessels and put me in touch with other key informants.Change your workplace. During fieldwork, a workplace is often needed for interview transcription, literature research and interim data analysis. Moving the workplace outside wherever you are staying during a field trip allows you to immerse yourself in the daily lives of local people and interact with them more easily. For me, such agile ‘mini-office’ locations were cafes, public libraries and picnic tables. In this way, I was able to recruit interview partners on the spot.How to create deeper mental accessWear appropriate outfits. First impressions count, always. Researchers are judged not only on what they say and how they say it, but also on how they look. Certain clothes, such as those with a political slogan or religious symbol, have certain meanings and connotations. Depending on the context and whom you talk to, your appearance could promote or impede making connections and building rapport. For instance, whereas my practical ‘outdoorsy’ get-dirty outfit was appropriate for interviews on fishing vessels, a modest appearance (non-branded clothes and a simple style) was useful in rural areas of Guangzhou.Show respect. Just like in any other relationship, respect and humility play a crucial part in building a trustworthy interviewer–interviewee relationship. Showing respect can be subtly embedded in conversations in many ways, including in the content of questions and the manner in which they are asked. When interviewees started to close down when asked about painful issues, such as underemployment or loss of identity, I upheld their privacy, comfort and security by not probing when given an evasive answer. Instead, I changed the interview focus and, when appropriate, cautiously reapproached the sensitive issue by using interview techniques such as roleplaying. Interviewees were asked to put themselves in the position of someone else, such as a spatial planner or politician, and assess the issue at hand from this perspective. Taking such an imaginary role can help to make the interviewees feel more secure and face pain more openly.Be humble. Having a modest view of yourself is essential to communicate at eye level with people. As a scientist, you can easily fall into the trap of thinking that your thoughts and concepts are somehow more valuable because you are well-educated and established. However, you are the one asking questions — and the interviewees, whether they are fishers, farmers or homeless people, often know more about many things than you do. Being aware of this is an expression of humility. I let the interviewees know that they were the local experts and I was the foreign learner.Use small talk. Small talk — including non-verbal communication, such as smiling, or connective gestures, for example handing out a handkerchief or offering some tea — has an essential bonding function. Talking about ‘safe’ topics can help the interviewee to overcome the feelings of otherness, newness and discomfort that can emerge in an interview, and fosters social cohesiveness. This can help to counteract the asymmetrical power relationship between the researcher (who asks) and the researched (who answers). For example, before substantive questioning, I created shared experiences by talking about last night’s storm or the world cod-fishing championship, which takes place every year in Lofoten. This took the relationship to a greater level of intimacy and togetherness — which small talk after finishing the interview can strengthen. I remember joking about my stamina for eating properly with chopsticks to one interviewee.Use self-disclosure. Revealing selected information about yourself and sharing your own thoughts with interlocutors can help to create and reaffirm a sphere of confidentiality and trust. Fishers in Norway would, for instance, often ask “What interested you in Lofoten coastal fisheries?” or “Why do you ask me and not the scientists from Tromsø University?” I answered such questions honestly, which assisted in creating a more balanced relationship, encouraging the interviewees to address sensitive subjects more openly and readily.Change interview sites. In several interviews, I found that the answers given tended to depend on where the interview was held and which identity that site evoked for the interviewee. For example, a fisher did not talk about climate-change concerns on his fishing vessel (any concern was masked by his existential fear of losing his livelihood as a coastal fisher), but he later that day freely discussed his worries in his home. Changing the interview site can be a helpful technique to access hidden thoughts and feelings.Above all, be realistic. You will probably make mistakes; I regretted not dressing warmly enough on a fishing vessel in Arctic weather. Locals will find you amusing, weird or impolite. They will keep out of your way, and you will never know why. And they will terminate interviews prematurely with no excuse. And that’s all right. In the end, fieldwork is a combination of planning, resources, time, skills, hard work, commitment, headache, joy — and luck. Learn from your mistakes, and accept the things you cannot change. More

Analytical method validationThe results of the precision study with relative standard deviation (RSD), and accuracy are shown in Table 1. Through the precision study we found the value of RSD as less than 5%. Moreover, accuracy was done with percent recovery experiments. The results showed that the percentage recoveries for spiked samples were in the range of 95.7–103.7%.Table 1 Shows percent (%) recovery and relative standard deviation.Full size tablePhysicochemical properties and water quality indexThe investigations of the water quality properties of the Narora channel are shown in Table 2. The temperature, TDS, turbidity, and alkalinity were within the standards of the country18 and WHO19 (taken from UNEPGEMS). While pH and dissolved oxygen (D.O) were above the recommended standards indicating poor water quality. Moreover, the detected heavy metals were in the following order Ni  > Fe  > Cd  > Zn  > Cr  > Cu  > Mn. Among these heavy metals Mn, Cu, and Zn were within the recommended limits whereas Cr, Fe, Ni, and Cd were crossing the limits18 contributing to the poor quality. Furthermore, the WQI calculation will give more insights into the overall quality of water as it explains the combined effect of several physicochemical properties12. Its calculation is done simply by converting numerous variables of water quality into a single number12,20. In addition to this, WQI simplifies all the data and helps in clarifying water quality issues by combining the complex data and producing a score that shows the status of water quality2,12,21. The WQI classifies water quality status into five groups such as if WQI  Cu  > Zn  > Fe  > Zn  > Ni  > Cr from root to stalk; and Mn  > Cd  > Zn  > Cu  > Fe  > Ni  > Cr from stalk to leaves.Table 5 Heavy metal concentrations in Eichhornia crassipes (mg/kg.dw).Full size tableFigure 3MPI values in E. crassipes.Full size imageTable 6 Bioaccumulation factor (BAF), transfer factor (TF), and mobility factor (MF) in plant E. crassipes.Full size tableThese factors BAF, TF, and MF are utilized to monitor the level of anthropogenic pollution in plants and their surrounding medium2,15,32,34,35. BAF shows the concentrations of heavy metals bioaccumulated by plants from the water. If the BAF  > 1 it indicates hyperaccumulation36. So, in the present study, all the concerned heavy metals were hyperaccumulated in the plant. The TF elucidates the capability of the plant to translocate the accumulated metals to its other parts. The roots of E. crassipes showed the highest translocation capacity for Ni (1.57) as well as Zn (1.30) to other parts. If the value of TF exceeds 1, then it represents the high accumulation efficiency37,38, therefore, plants will be considered as the hyperaccumulators for the Ni and Zn. Although the Cd was the highest accumulated metal in the plant, it could have been because of its may be because of its low TF. Whereas, TF values lower than 1 for Cr, Mn, Fe, Cu, and Cd pointed out that this plant’s roots act as a non-hyperaccumulator for these heavy metals. Furthermore, the highest MF values were depicted for Mn in both cases which reflects that E. crassipes can suitably be used for phytoextraction of Mn as well as for Cd, Zn, Fe, Ni, and Cu. The BAF, TF, and MF of Cr are low in the present study, which implies that roots are limiting the Cr. Moreover, if the BAF ≤ 1.00 then it shows the capability of absorption only rather than accumulation36,37. In addition, if the values of BAF, TF, and MF exceed 1, plants can also work for phytoextraction. Furthermore, if the BAF  > 1 and TF  More