Ecology

Now consider a generalized model in which the species interactions are heterogeneous. A natural way of introducing heterogeneity in the system is by having a species diversify into subpopulations with different interaction strengths12,13,14,15. This way of modeling heterogeneity is useful as it can describe different kinds of heterogeneity. For example, the subpopulations could represent polymorphic traits that are genetically determined or result from plastic response to heterogeneous environments. A population could also be divided into local subpopulations in different spatial patches, which can migrate between patches and may face different local predators. We can also model different behavioral modes as subpopulations that, for instance, spend more time foraging for food or hiding from predators. We study several kinds of heterogeneity after we introduce a common mathematical framework. By studying these different scenarios using variants of the model, we show that our main results are not sensitive to the details of the model.We focus on the simple case where only the prey species splits into two types, (C_1) and (C_2), as illustrated in Fig. 1b. The situation is interesting when predator A consumes (C_1) more readily than predator B and B consumes (C_2) more readily than A (i.e., (a_1 / a_0 > b_1 / b_0) and (b_2 / b_0 > a_2 / a_0), which is equivalent to the condition that the nullclines of A and B cross, see section “Resources competition and nullcline analysis”). The arrows between (C_1) and (C_2) in Fig. 1b represent the exchange of individuals between the two subpopulations, which can happen by various mechanisms considered below. Such exchange as well as intraspecific competition between (C_1) and (C_2) result from the fact that the two prey types remain a single species.The system is now described by an enlarged Lotka-Volterra system with four variables, A, B, (C_1), and (C_2): $$begin{aligned} dot{A}&= varepsilon _A ,alpha _{A1} , A , C_1 + alpha _{A2} , A , C_2 – beta _A , A end{aligned}$$
(3a)
$$begin{aligned} dot{B}&= varepsilon _B , alpha _{B1} , B , C_1 + alpha _{B2} , B , C_2 – beta _B , B end{aligned}$$
(3b)
$$begin{aligned} dot{C_1}&= C_1 , (beta _C – alpha _{CC} , C)-alpha _{A1} , C_1 A-alpha _{B1} , C_1 B – sigma _1 , C_1 + sigma _2 , C_2 end{aligned}$$
(3c)
$$begin{aligned} dot{C_2}&= C_2 , (beta _C – alpha _{CC} , C) -alpha _{A2} , C_2 A -alpha _{B2} , C_2 B + sigma _1 , C_1 – sigma _2 , C_2 end{aligned}$$
(3d)
The parameters in these equations and their meanings are listed in Table 1. Here we assume that the prey types (C_1) and (C_2) have the same birth rate and intraspecific competition strength, but different interaction strengths with A and B. Note that (C_1) and (C_2) are connected by the (sigma _i) terms, which represent the exchange of individuals between these subpopulations through mechanisms studied below; these terms indicate a major difference between our model of a prey with intraspecific heterogeneity and other models of two prey species. For the convenience of analysis, we transform the variables (C_1) and (C_2) to another pair of variables C and (lambda), where (C equiv C_1 + C_2) is the total population of C as before, and (lambda equiv C_2 / (C_1 + C_2)) represents the composition of the population (Fig. 1c). After this transformation and rescaling of variables (described in “Methods”), the new dynamical system can be written as: $$begin{aligned} dot{A}&= A , big ( C , (a_1 (1-lambda ) + a_2 lambda ) – a_0 big ) end{aligned}$$
(4a)
$$begin{aligned} dot{B}&= B , big ( C , (b_1 (1-lambda ) + b_2 lambda ) – b_0 big ) end{aligned}$$
(4b)
$$begin{aligned} dot{C}&= C , big ( 1 – C – A (a_1 (1-lambda ) + a_2 lambda ) – B (b_1 (1-lambda ) + b_2 lambda ) big ) end{aligned}$$
(4c)
$$begin{aligned} dot{lambda }&= lambda (1-lambda ) , big ( A (a_1 – a_2) + B (b_1 – b_2) big ) + eta _1 (1-lambda ) – eta _2 lambda end{aligned}$$
(4d)
Here, (a_i) and (b_i) are the (rescaled) feeding rates of the predators on the prey type (C_i); (a_0) and (b_0) are the death rates of the predators as before; (eta _1) and (eta _2) are the exchange rates of the prey types (Table 1). The latter can be functions of other variables, representing different kinds of heterogeneous interactions that we study below. Notice that Eqs. (4a–4c) are equivalent to the homogeneous Eqs. (2a–2c) but with effective interaction strengths (a_text {eff} = (1-lambda ) , a_1 + lambda , a_2) and (b_text {eff} = (1-lambda ) , b_1 + lambda , b_2) that both depend on the prey composition (lambda) (Fig. 1c).Table 1 Model parameters (before/after rescaling) and their meanings.Full size tableThe variable (lambda) can be considered an internal degree of freedom within the C population. In all of the models we study below, (lambda) dynamically stabilizes to a special value (lambda ^*) (a bifurcation point), as shown in Fig. 3a. Accordingly, a new equilibrium point (P_N) appears (on the line (mathscr {L}) in Fig. 2), at which all three species coexist. For comparison, Fig. 3b shows the equilibrium points if (lambda) is held fixed at any other values, which all result in the exclusion of one of the predators. Thus, heterogeneous interactions give rise to a new coexistence phase (see Fig. 4 below) by bringing the prey composition (lambda) to the value (lambda ^*), instead of having to fine-tune the interaction strengths. The exact conditions for this new equilibrium to be stable are detailed in “Methods”.Figure 3(a) Time series of (lambda) for systems with each kind of heterogeneity. All three systems stabilize at the same (lambda ^*) value, which is the bifurcation point in panel (b). (b) Equilibrium population of each species (X = A), B, or C, with (lambda) held fixed at different values. Solid curves represent stable equilibria and dashed curves represent unstable equilibria (see Eq. (9) in “Methods”). The vertical dashed line is where (lambda = lambda ^*), which is also the bifurcation point. Notice that the equilibrium population of C is maximized at this point (for (a_1 > a_2) and (b_2 > b_1)). Parameters used here are ((a_0, a_1, a_2, b_0, b_1, b_2, rho , eta _1, eta _2, kappa ) = (0.25, 0.5, 0.2, 0.4, 0.2, 0.6, 0.5, 0.05, 0.05, 50)).Full size imageInherent heterogeneityWe first consider a scenario where individuals of the prey species are born as one of two types with a fixed ratio, such that a fraction (rho) of the newborns are (C_2) and ((1-rho )) are (C_1). This could describe dimorphic traits, such as the winged and wingless morphs in aphids12 or the horned and hornless morphs in beetles13. We call this “inherent” heterogeneity, because individuals are born with a certain type and cannot change in later stages of life. The prey type given at birth determines the individual’s interaction strength with the predators. This kind of heterogeneity can be described by Eq. (4d) with (eta _1 = rho (1-C)) and (eta _2 = (1-rho ) (1-C)) (see “Methods”).Figure 4Phase diagrams showing regions of parameter space identified by the stable equilibrium points. Yellow region represents (P_C) (predators A, B both extinct), red represents (P_A) (A excludes B), blue represents (P_B) (B excludes A), and green represents (P_N) (A, B coexist). The middle point (black dot) is where the preferences of the two predators are identical, (a_2/a_0=b_2/b_0) and (b_1/b_0=a_1/a_0). The coexistence phase appears in all three kinds of heterogeneity modeled here. (a–d) Inherent heterogeneity: Individuals of the prey population are born in two types with a fixed composition (rho). In the extreme cases of (rho = 0) and 1, the prey is homogeneous and there is no coexistence of the predators. (e–h) Reversible heterogeneity: Individual prey can switch types with fixed switching rates (eta _1) and (eta _2). As the switching rates increase, the coexistence region shrinks because the prey population becomes effectively homogeneous (the occasional green spots are numerical artifacts because the time to reach the equilibrium becomes long in this limit). (i–l) Adaptive heterogeneity: The switching rates (eta _i) dynamically adapt to the predator densities, so as to maximize the growth rate of the prey. As the sharpness (kappa) of the sigmoidal decision function is increased, the prey adapts more optimally and the region of coexistence expands. Parameters used here are ((a_0, a_1, b_0, b_2) = (0.3, 0.5, 0.4, 0.6)).Full size imageThe stable equilibrium of the system can be represented by phase diagrams that show the identities of the species at equilibrium. We plot these phase diagrams by varying the parameters (a_2) and (b_1) while keeping (a_1) and (b_2) constant. As shown in Fig. 4a–d, the equilibrium state depends on the parameter (rho). In the limit (rho = 0) or 1, we recover the homogeneous case because only one type of C is produced. The corresponding phase diagrams (Fig. 4a, d) contain only two phases where either of the predators is excluded, illustrating the competitive exclusion principle. For intermediate values of (rho), however, there is a new phase of coexistence that separates the two exclusion phases (Fig. 4b, c). There are two such regions of coexistence, which touch at a middle point and open toward the bottom left and upper right, respectively. The middle point is at ((a_2/a_0 = b_2/b_0, b_1/b_0 = a_1/a_0)), where the feeding preferences of the two predators are identical (hence their niches fully overlap). Towards the origin and the far upper right, the predators consume one type of C each (hence their niches separate). The coexistence region in the bottom left is where the feeding rates of the predators are the lowest overall. There can be a region (yellow) where both predators go extinct, if their primary prey type alone is not enough to sustain each predator. Increasing the productivity of the system by increasing the birth rate ((beta _C)) of the prey eliminates this extinction region, whereas lowering productivity causes the extinction region to take over the lower coexistence region. Because the existence and identity of the phases is determined by the configuration of the equilibrium points (Fig. 2, see also section “Mathematical methods”), the qualitative shape of the phase diagram is not sensitive to changes of parameter values.The new equilibrium is not only where the predators A and B can coexist, but also where the prey species C grows to a larger density than what is possible for a homogeneous population. This is illustrated in Fig. 3b, which shows the equilibrium population of C if we hold (lambda) fixed at different values. The point (lambda = lambda ^*) is where the system with a dynamic (lambda) is stable, and also where the population of C is maximized (when A and B prefer different prey types). That means the population automatically stabilizes at the optimal composition of prey types. Moreover, the value of (C^*) at this coexistence point can even be larger than the equilibrium population of C when there is only one predator A or B. This is discussed further in section “Multiple-predator effects and emergent promotion of prey”. These results suggest that heterogeneity in interaction strengths can potentially be a strategy for the prey population to leverage the effects of multiple predators against each other to improve survival.Reversible heterogeneityWe next consider a scenario where individual prey can switch their types. This kind of heterogeneity can model reversible changes of phenotypes, i.e., trait changes that affect the prey’s interaction with predators but are not permanent. For example, changes in coat color or camouflage14,16,17, physiological changes such as defense15, and biomass allocation among tissues18,19. One could also think of the prey types as subpopulations within different spatial patches, if each predator hunts at a preferred patch and the prey migrate between the patches20,21. With some generalization, one could even consider heterogeneity in resources, such as nutrients located in different places, that can be reached by primary consumers, such as swimming phytoplankton22. We can model this “reversible” kind of heterogeneity by introducing switching rates from one prey type to the other. In Eq. (4d), (eta _1) and (eta _2) now represent the switching rates per capita from (C_1) to (C_2) and from (C_2) to (C_1), respectively. Here we study the simplest case where both rates are fixed.In the absence of the predators, the natural composition of the prey species given by the switching rates would be (rho equiv eta _1 / (eta _1 + eta _2)), and the rate at which (lambda) relaxes to this natural composition is (gamma equiv eta _1 + eta _2). Compared to the previous scenario where we had only one parameter (rho), here we have an additional parameter (gamma) that modifies the behavior of the system. Fig. 4e–h shows phase diagrams for the system as (rho) is fixed and (gamma) varies. We again find the new equilibrium (P_N) where all three species coexist. When (gamma) is small, the system has a large region of coexistence. As (gamma) is increased, this region is squeezed into a border between the two regions of exclusion, where the slope of the border is given by (eta _1/eta _2) as determined by the parameter (rho). However, this is different from the exclusion we see in the case of inherent heterogeneity, which happens only for (rho rightarrow 0) or 1, where the borders are horizontal or vertical (Fig. 4a,d). Here the predators exclude each other despite having a mixture of prey types in the population.This special limit can be understood as follows. For a large (gamma), (lambda) is effectively set to a constant value equal to (rho), because it has a very fast relaxation rate. In other words, the prey types exchange so often that the population always maintains a constant composition. In this limit, the system behaves as if it were a homogeneous system with effective interaction strengths (a_text {eff} = (1-rho ) , a_1 + rho , a_2) and (b_text {eff} = (1-rho ) , b_1 + rho , b_2). As in a homogeneous system, there is competitive exclusion between the predators instead of coexistence. This demonstrates that having a constant level of heterogeneity is not sufficient to cause coexistence. The overall composition of the population must be able to change dynamically as a result of population growth and consumption by predators.An interesting behavior is seen when we examine a point inside the shrinking coexistence region as (gamma) is increased. Typical trajectories of the system for such parameter values are shown in Fig. 5. As (gamma) increases, the system relaxes to the line (mathscr {L}) quickly, then slowly crawls along it towards the final equilibrium point (P_N). This is because increasing (gamma) increases the speed that (lambda) relaxes to (lambda ^*), and when (lambda rightarrow lambda ^*), (mathscr {L}) becomes marginally stable. Therefore, the attraction to (mathscr {L}) in the perpendicular direction is strong, but the attraction towards the equilibrium point along (mathscr {L}) is weak. This leads to a long transient behavior that makes the system appear to reach no equilibrium in a limited time23,24. It is especially true when there is noise in the dynamics, which causes the system to diffuse along (mathscr {L}) with only a weak drift towards the final equilibrium (Fig. 5). Thus, the introduction of a fast timescale (quick relaxation of (lambda) due to a large (gamma)) actually results in a long transient.Figure 5Trajectories of the system projected in the A-B plane, with parameters inside the coexistence region (by holding the position of (P_N) fixed). As (gamma) increases, the system tends to approach the line (mathscr {L}) quickly and then crawl along it. The grey trajectory is with independent Gaussian white noise ((sim mathscr {N}(0,0.5))) added to each variable’s dynamics. Noise causes the system to diffuse along (mathscr {L}) for a long transient period before coming to the equilibrium point (P_N). Parameters used here are ((a_0, a_1, a_2, b_0, b_1, b_2) = (0.2, 0.8, 0.5, 0.2, 0.6, 0.9)), chosen to place (P_N) away from the middle of (mathscr {L}) to show the trajectory drifting toward the equilibrium.Full size imageAdaptive heterogeneityA third kind of heterogeneity we consider is the change of interactions in time. By this we mean an individual can actively change its interaction strength with others in response to certain conditions. This kind of response is often invoked in models of adaptive foraging behavior, where individuals choose appropriate actions to maximize some form of fitness25,26. For example, we may consider two behaviors, resting and foraging, as our prey types. Different predators may prefer to strike when the prey is doing different things. In response, the prey may choose to do one thing or the other depending on the current abundances of different predators. Such behavioral modulation is seen, for example, in systems of predatory spiders and grasshoppers27. Phenotypic plasticity is also seen in plant tissues in response to consumers28,29,30.This kind of “adaptive” heterogeneity can be modeled by having switching rates (eta _1) and (eta _2) that are time-dependent. Let us assume that the prey species tries to maximize its population growth rate by switching to the more favorable type. From Eq. (4c), we see that the growth rate of C depends linearly on the composition (lambda) with a coefficient (u(A,B) equiv (a_1 – a_2) A + (b_1 – b_2) B). Therefore, when this coefficient is positive, it is favorable for C to increase (lambda) by switching to type (C_2). This can be achieved by having a positive switching rate (eta _2) whenever (u(A,B) > 0). Similarly, whenever (u(A,B) < 0), it is favorable for C to switch to type (C_1) by having a positive (eta _1). In this way, the heterogeneity of the prey population constantly adapts to the predator densities. We model such adaptive switching by making (eta _1) and (eta _2) functions of the coefficient u(A, B), e.g., (eta _1(u) = 1/(1+mathrm {e}^{kappa u})) and (eta _2(u) = 1/(1+mathrm {e}^{-kappa u})). The sigmoidal form of the functions means that the switching rate in the favorable direction for C is turned on quickly, while the other direction is turned off. The parameter (kappa) controls the sharpness of this transition.Phase diagrams for the system with different values of (kappa) are shown in Fig. 4i–l. A larger (kappa) means the prey adapts its composition faster and more optimally, which causes the coexistence region to expand. In the extreme limit, the system changes its dynamics instantaneously whenever it crosses the boundary where (u(A,B) = 0), like in a hybrid system31. Such a system can still reach a stable equilibrium that lies on the boundary, if the flow on each side of the boundary points towards the other side32. This is what happens in our system and, interestingly, the equilibrium is the same three-species coexistence point (P_N) as in the previous scenarios. The region of coexistence turns out to be largest in this limit (Fig. 4l).Our results suggest that the coexistence of the predators can be viewed as a by-product of the prey’s strategy to maximize its own benefit. The time-dependent case studied here represents a strategy that involves the prey evaluating the risk posed by different predators. This is in contrast to the scenarios studied above, where the prey population passively creates phenotypic heterogeneity regardless of the presence of the predators. These two types of behavior are analogous to the two strategies studied for adaptation in varying environments, i.e., sensing and bet-hedging33,34. The former requires accessing information about the current environment to make optimal decisions, whereas the latter relies on maintaining a diverse population to reduce detrimental effects caused by environmental changes. Here the varying abundances of the predators play a similar role as the varying environment. From this point of view, the heterogeneous interactions studied here can be a strategy of the prey species that is evolutionarily favorable. More

Ren, W. et al. Changes of periphyton abundance and biomass driven by factors specific to flooding inflow in a river inlet area in Erhai Lake, China. Front. Environ. Sci. 9, 680718. https://doi.org/10.3389/fenvs.2021.680718 (2021).Article 

Google Scholar 
Woodruff, S. L. et al. The effects of a developing biofilm on chemical changes across the sediment-water interface in a freshwater environment. Int. Rev. Hydrobiol. 84(5), 509–532 (1999).CAS 

Google Scholar 
Muñoz, I., Real, M., Guasch, H., Navarro, E. & Sabater, S. Effects of atrazine on periphyton under grazing pressure. Aquat. Toxicol. 55(3–4), 239–249 (2001).
Google Scholar 
Hoagland, K. D., Roemer, S. C. & Rosowski, J. R. Colonization and community structure of two periphyton assemblages, with emphasis on the diatoms (Bacillariophyceae). Am. J. Bot. 69, 188–213. https://doi.org/10.2307/2443006 (1982).Article 

Google Scholar 
Steinman, A. D. & McIntire, C. D. Effects of current velocity and light energy on the structure of periphyton assemblages in laboratory streams. J. Phycol. 22, 352–361. https://doi.org/10.1111/J.1529-8817.1986.TB00035.X (1986).Article 

Google Scholar 
Tonkin, J. D., Death, R. G. & Barquín, J. Periphyton control on stream invertebrate diversity: Is periphyton architecture more important than biomass?. Mar. Freshw. Res. 65(9), 818–829 (2014).
Google Scholar 
Beck, W. S., Markman, D. W., Oleksy, I. A., Lafferty, M. H. & Poff, N. L. Seasonal shifts in the importance of bottom-up and top-down factors on stream periphyton community structure. Oikos 128, 680–691. https://doi.org/10.1111/oik.05844 (2018).Article 
CAS 

Google Scholar 
Hogsden, K. L. & Harding, J. S. Consequences of acid mine drainage for the structure and function of benthic stream communities: A review. Freshw. Sci. 31, 108–120. https://doi.org/10.1899/11-091.1 (2012).Article 

Google Scholar 
Sofi, M. S., Bhat, S. U., Rashid, I. & Kuniyal, J. C. The natural flow regime: A master variable for maintaining river ecosystem health. Ecohydrology 13(8), e2247. https://doi.org/10.1002/eco.2247 (2020).Article 

Google Scholar 
Biggs, B. J. F. Eutrophication of streams and rivers: Dissolved nutrient-chlorophyllrelationship for benthic algae. J. N. Am. Benthol. Soc. 19, 17–31 (2000).
Google Scholar 
Ormerod, S. J., Dobson, M., Hildrew, A. G. & Townsend, C. Multiple stressors in freshwater ecosystems. Freshw. Biol. 55, 1–4 (2010).
Google Scholar 
Poff, et al. The natural flow regime: A paradigm for river conservation and restoration. Bioscience 47, 769–784 (1997).
Google Scholar 
Naiman, R. J., Décamps, H., & McClain, M. E. Riparia: Ecology, Conservation and Management of Streamside Communities, (Elsevier/Academic Press, 2005).Gleick, P. H. Water use. Annu. Rev. Environ. Resour. 28, 275–314 (2003).
Google Scholar 
Jenkins, K. M. & Boulton, A. J. Connectivity in a dryland river: Short-term aquatic macroinvertebrate recruitment following floodplain inundation. Ecology 84(10), 2708–2723 (2003).
Google Scholar 
Biggs, B. J. F. Patterns in benthic algae of streams. In Algal Ecology in Freshwater Benthic Ecosystems (eds. Stevenson, R. J., Bothwell, M. L., & Lowe, R. L.) 31–56 (Academic Press, 1996).Smolar-Žvanut, N. & Mikoš, M. The impact of flow regulation by hydropower dams on the periphyton community in the Soča River, Slovenia. Hydrol. Sci. J. 59(5), 1032–1045. https://doi.org/10.1080/02626667.2013.834339 (2014).Article 
CAS 

Google Scholar 
Curry, C. J. & Baird, D. J. Habitat type and dispersal ability influence spatial structuring of larval Odonata and Trichoptera assemblages. Freshw. Biol. 60, 2142–2152 (2015).
Google Scholar 
Wu, N., Cai, Q. & Fohrer, N. Contribution of microspatial factors to benthic diatom communities. Hydrobiologia 732, 49–60. https://doi.org/10.1007/s10750-014-1843-3 (2014).Article 
CAS 

Google Scholar 
Mueller, M., Pander, J. & Geist, J. The effects of weirs on structural stream habitat and biological communities. J. Appl. Ecol 48(6), 1450–1461. https://doi.org/10.1111/j.1365-2664.2011.02035.x (2011).Article 

Google Scholar 
Davies, P. M. et al. Flow–ecology relationships: closing the loop on effective environmental flows. Mar. Freshw. Res. 65(2), 133–141 (2013).
Google Scholar 
Jun, Y. C. et al. Spatial distribution of benthic macroinvertebrate assemblages in relation to environmental variables in Korean nationwide streams. Water 8(1), 27. https://doi.org/10.3390/w8010027 (2016).Article 

Google Scholar 
Biggs, B. J. F. & Close, M. E. Periphyton biomass dynamics in gravel bed rivers: The relative effects of flows and nutrients. Freshw. Biol. 22, 209–231 (1989).CAS 

Google Scholar 
Jowett, I. & Biggs, B. J. F. Flood and velocity effects on periphyton and silt accumulation in two New Zealand rivers. N. Zeal. J. Mar. Freshw. Res. 31, 287–300 (1997).
Google Scholar 
Biggs, B. J. F., Goring, D. G. & Nikora, V. I. Subsidy and stress responses of stream periphyton to gradients in water velocity as a function of community growth form. J. Phycol. 34, 598–607 (1998).
Google Scholar 
Malmqvist, B. & Englund, G. Effects of hydropower-induced flow perturbations on mayfly (Ephemeroptera) richness and abundance in north Swedish river rapids. Hydrobiologia 341(2), 145–158 (1996).
Google Scholar 
Poff, N. L. & Ward, J. V. Herbivory under different flow regimes: A field experiment and test of a model with a benthic stream insect. Oikos 72, 179–188 (1995).
Google Scholar 
Poff, L. N., Wellnitz, T. & Monroe, J. B. Redundancy among three herbivorous insects across an experimental current velocity gradient. Oecologia 134, 262–269. https://doi.org/10.1007/s00442-002-1086-2 (2003).Article 

Google Scholar 
Vaughn, C. C. The role of periphyton abundance and quality in the microdistribution of a stream grazer, Helicopsyche borealis (Trichoptera: Helicopsychidae). Freshw. Biol. 16, 485–493 (1986).
Google Scholar 
Francoeur, S. N. Meta-analysis of lotic nutrient amendment experiments: Detecting and quantifying subtle responses. J. N. Am. Benthol. Soc. 20, 358–368 (2001).
Google Scholar 
Elser, J. J. et al. Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecol. Lett. 10, 1135–1142 (2007).
Google Scholar 
Hillebrand, H. Meta-analysis of grazer control of periphyton biomass across aquatic ecosystems. J. Phycol. 45, 798–806 (2009).
Google Scholar 
Lamberti, G. A. The role of periphyton in benthic food webs. In Algal Ecology—Freshwater Benthic Ecosystems, 533–572 (eds. Stevenson, R. J., Bothwell, M. L. & Lowe, R. L.) (Academic Press, 1996).Lamberti, G. A. et al. Influence of grazer type and abundance on plant–herbivore interactions in streams. Hydrobiologia 306, 179–188 (1995).
Google Scholar 
Gregory, S. V. Plant–herbivore interactions in stream systems. In Stream Ecology (eds. Barnes, J. R. & Minshall, G. W.) 157–189 (Plenum, 1983).Lamberti, G. A. & Moore, J. W. Aquatic insects as primary consumers. In The Ecology of Aquatic Insects (eds Resh, V. H. & Rosenberg, D. M.) 164–195 (Praeger, 1984).
Google Scholar 
Sterner, R. W., Elser, J. J. & Hessen, D. O. Stoichiometric relationships among producers, consumers and nutrient cycling in pelagic ecosystems. Biogeochemistry 17, 49–67 (1992).CAS 

Google Scholar 
Kahlert, M. & Baunsgaard, M. T. Nutrient recycling—A strategy of a grazer community to overcome nutrient limitation. J. N. Am. Benthol. Soc. 18, 363–369 (1999).
Google Scholar 
Burkholder, J. M., Wetzel, R. G. & Klomparens, K. L. Direct comparison of phosphate uptake by adnate and loosely attached microalgae within and intact biofilm matrix. Appl. Environ. Microbiol. 56, 2882–2890 (1990).CAS 

Google Scholar 
Steinman, A. D. Effects of grazers on freshwater benthic algae. In Algal Ecology: Freshwater Benthic Ecosystems (eds. Stevenson, R. J., Bothwell & Lowe, R. L.) 341–366 (Academic Press, 1996).Smucker, N. J. & Vis, M. L. Spatial factors contribute to benthic diatom structure in streams across spatial scales: Considerations for biomonitoring. Ecol. Indic. 11, 1191–1203 (2011).
Google Scholar 
Myers, et al. Biodiversity hotspots for conservation priorities. Nature 403, 853–858 (2000).Article 
CAS 

Google Scholar 
Wang, J., Pan, F., Soininen, J., Heino, J. & Shen, J. Nutrient enrichment modifies temperature-biodiversity relationship in large scale field experiments. Nat. Commun. 7, 13 (2016).
Google Scholar 
Wu, et al. Flow regimes filter species traits of benthic diatom communities and modify the functional features of lowland streams-a nationwide scale study. Sci. Total Environ. 651, 357–366 (2019).CAS 

Google Scholar 
Nisar, M. A. Geospatial approach to study environmental characterization of a Kashmir wetland (Anchar) catchment with special reference to land use/land cover and changing climate. Ph.D Thesis, Sher-e-Kashmir University of Agricultural Sciences and Technology, Kashmir. Weblink. http://krishikosh.egranth.ac.in/handle/1/91309 (2012).Bhat, S. U., Sofi, A. H., Yaseen, T., Pandit, A. K. & Yousuf, A. R. Macro invertebrate community from Sonamarg streams of Kashmir Himalaya. Pak. J. Biol. Sci. 14(3), 182–194. https://doi.org/10.3923/pjbs.2011.182.194 (2011).Article 
CAS 

Google Scholar 
Baba, A. I., Sofi, A. H., Bhat, S. U., & Pandit, A. K. Periphytic algae of river Sindh in the Sonamarg area of Kashmir valley. J. Phytol. 3(6) (2011).Sofi, M. S., Rautela, K. S., Bhat, S. U., Rashid, I. & Kuniyal, J. C. Application of geomorphometric approach for the estimation of hydro-sedimentological flows and cation weathering rate: Towards understanding the sustainable land use policy for the Sindh Basin, Kashmir Himalaya. Water Air Soil Pollut. 232(7), 1–11. https://doi.org/10.1007/s11270-021-05217-w (2021).Article 
CAS 

Google Scholar 
Romshoo, S. A., & Fayaz, M. Use of high resolution remote sensing for improving environmental Friendly tourism master planning in the Alpine Himalaya: A case study of Sonamarg tourist resort, Kashmir. J. Himalayan Ecol. Sustain. Dev. 14 (2019).Biggs, B. J. F. & Kilroy, C. Stream periphyton monitoring manual. Published by NIWA for Ministry for the Environment, 226 Christchurch, New Zealand: NIWA (2000).APHA. Standard Methods for Examination of Water and Wastewater, 22nd edn. (American Public Health Association, 2012).Cox, E. J. Identification of Freshwater Diatoms from Live Material. (Chapman and Hall, 1996). https://doi.org/10.1017/S0025315400041023.Krammer, K., & Lange-Bertalot, H. Bacillariophyceae, Part 5. English and French Translation of the Keys. (VEB Gustav Fisher Verlag, 2000).Reichardt, E. A remarkable association of diatoms in a spring habitat in the Grazer Bergland, Austria. In Iconographia Diatomologica (ed. Lange-Bertalot, H.) 419–479 (2004).Żelazna-Wieczorek, J. Diatom flora in springs of Lódz Hills (Central Poland). Biodiversity, taxonomy and temporal changes of epipsammic diatom assemblages in springs affected by human impact, 419. Volume 13 of Diatom monographs. Gantner. https://books.google.co.in/books?id=bdxeewAACAAJ (2011).Stark, J. D., Boothroyd, I. K. G., Harding, J. S., Maxted, J. R. & Scarsbrook, M. R. Protocols for sampling macroinvertebrates in wadeable streams. In New Zealand Macroinvertebrate Working Group Report no. 1. Prepared for the Ministry for the Environment. Sustainable Management Fund Project, 5103 (2001).Winterbourn, M. J. Sampling stream invertebrates. In Biological Monitoring of Freshwaters. Proceedings of the Seminar. Water and Soil Miscellaneous Publication No. 83 (eds. Pridmore, R. D., Cooper, A. B.) 241–258. (National Water and Soil Conservation Authority, 1985).Barbour, M. T., Gerritsen, J., Snyder, B. D., Stribling, J. B. Rapid bioassessment protocols for use in streams and wadeable rivers: periphyton, benthic macroinvertebrates and fish, 339. (United States Environmental Protection Agency, Office of Water, 1999).Malmqvist, B. & Hoffsten, P. O. Macroinvertebrate taxonomic richness, community structure and nestedness in Swedish streams. Fundam. Appl. Limnol. 150(1), 29–54. https://doi.org/10.1127/archiv-hydrobiol/150/2000/29 (2000).Article 

Google Scholar 
Ilmonen, J. & Paasivirta, L. Benthic macrocrustacean and insect assemblages in relation to spring habitat characteristics: Patterns in abundance and diversity. Hydrobiologia 533(1–3), 99–113. https://doi.org/10.1007/s10750-004-2399-4 (2005).Article 

Google Scholar 
Munasinghe, D. S. N., Najim, M. M. M., Quadroni, S. & Musthafa, M. M. Impacts of streamflow alteration on benthic macroinvertebrates by mini-hydro diversion in Sri Lanka. Sci. Rep. 11(1), 546. https://doi.org/10.1038/s41598-020-79576-5 (2021).Article 
CAS 

Google Scholar 
Edmondson, W. T. Fresh-Water Biology, 2nd ed. 1050–1056 (Wiley, 1959).Pennak, R. W. Freshwater Invertebrates of United States. (Wiley, 1978).McCafferty, W. P., Provonsha, A. V. Aquatic entomology: The fishermen’s and ecologists’ Illustrated Guide to Insects and their Relatives. (Jones and Bartlett Publishers, 1983).Borror, D., Triplehorn, C., Johnson, N. An Introduction to the Study of Insects, 6th ed. (Saunders College Publishing, 1989).Ward, J. V. Aquatic Insect Ecology, Biology and Habitat. (Wiley, 1992).Engblom, E. & Lingdell, P.E. Analyses of Benthic Invertebrates (ed. Nyman, L.) (1999).Bouchard, R. W. Guide to Aquatic Invertebrates of the Upper Midwest: Identification Manual for Students (University of Minnesota, 2004).
Google Scholar 
Subramanian, K. A. & Sivaramakrishnan, K. G. Aquatic Insects for Biomonitoring Freshwater Ecosystems—A Methodology Manual. (Ashoka Trust for Ecology and Environment (ATREE), 2007).Thorp, J. H., & Covich, A. P. (eds.) Ecology and Classification of North American Freshwater Invertebrates. (Academic Press, 2009).Allan, J. D. & Castillo, M.M. An introduction to fluvial ecosystems. In Stream Ecology: Structure and Function of Running Waters, 1–12 (2007).Oksanen, et al. Vegan: Community ecology package. In: R package version 2.4-3.McCune, B. & Grace, B. Analysis of Ecological Communities (MjM Software Design, 2016).Clarke, K. R. & Gorley, R. N. Primer v6 Permanova+ (Primer-E Ltd., 2006).
Google Scholar 
Salazar, G. EcolUtils: Utilities for Community Ecology Analysis. R package version 0.1 software (2018).Anderson, M. J., Ellingsen, K. E. & McArdle, B. H. Multivariate dispersion as a measure of beta diversity. Ecol. Lett. 9(6), 683–693 (2006).
Google Scholar 
Gardener, M. Community Ecology: Analytical Methods in Using R and Excel. (Pelagic Publishing, 2014).Chao, A. & Bunge, J. Estimating the number of species in a stochastic abundance model. Biometrics 58, 531–539. https://doi.org/10.1111/j.0006-341X.2002.00531.x (2002).Article 
MATH 

Google Scholar 
Peres-Neto, P. R., Legendre, P., Dray, S. & Borcard, D. Variation partitioning of species data matrices: estimation and comparison of fractions. Ecology 87, 2614–2625 (2006).
Google Scholar 
Meng, X. L. et al. Responses of macroinvertebrates and local environment to short-term commercial sand dredging practices in a flood-plain lake. Sci. Total Environ. 631, 1350–1359 (2018).
Google Scholar 
Core Team, R. R: A Language and Environmental for Statistical Computing. (R Foundation for Statistical Computing, 2017).Wood, P. J. & Armitage, P. D. Biological effects of fine sediment in the lotic environment. Environ. Manag. 21, 203–217 (1997).CAS 

Google Scholar 
Marchant, R. Changes in the benthic invertebrate communities of the Thomson River, southeastern Australia, after dam construction. Regul. Rivers Res. Manag. 4, 71–89 (1989).
Google Scholar 
Gray, L. J. & Ward, J. V. Effects of sediment releases from a reservoir on stream macroinvertebrates. Hydrobiologia 96, 177–184 (1982).
Google Scholar 
Sand-Jensen, K., Moller, J. & Olesen, B. H. Biomass regulation of microbenthic algae in Danish lowland streams. Oikos 53, 332–340 (1988).
Google Scholar 
Lewis, M. A., Weber, D. E., Stanley, R. S. & Moore, J. C. Dredging impact on an urbanized Florida bayou: Effects on benthos and algal-periphyton. Environ. Pollut. 115(2), 161–171 (2001).CAS 

Google Scholar 
Biggs, B. J. Algal ecology in freshwater benthic ecosystems geology and landuse to the habitat template of periphyton in stream ecosystems. Freshw. Biol. 33, 419–438 (1995).
Google Scholar 
Taylor, et al. Can diatom-based pollution indices be used for biomonitoring in South Africa? A case study of the Crocodile West and Marico water management area. Hydrobiologia 592, 455–464 (2007).
Google Scholar 
Porter, et al. Efficacy of algal metrics for assessing nutrient and organic enrichment in flowing waters. Freshw. Biol. 53, 1036–1054 (2008).
Google Scholar 
Wetzel, R. G. & Likens, G. E. Limnological analyses, 3rd ed. In Nitrogen, Phosphorus, and Other Nutrients, 85–113. (Springer, 2000). https://doi.org/10.1007/978-1-4757-3250-4.Wetzel, R. G. Attached algal-substrata interactions: Fact or myth, and when and how? vol. 17. In Periphyton of Freshwater Ecosystems (ed. Wetzel, R.) 207–215 (Springer, 1983). https://doi.org/10.1007/978-94-009-7293-3_28.Krajenbrink, H. J. et al. Diatoms as indicators of the effects of river impoundment at multiple spatial scales. PeerJ 7, e8092. https://doi.org/10.7717/peerj.8092 (2019).Article 

Google Scholar 
Poff, N. L., Voelz, N. J., Ward, J. V. & Lee, R. E. Algal colonization under four experimentally-controlled current regimes in a high mountain stream. J. N. Am. Benthol. Soc. 9, 303–318 (1990).
Google Scholar 
Dodds, W. K. & Marra, J. L. Behaviors of the midge, Cricotopus (Diptera; Chironomidae) related to mutualism with Nostoc parmeloides (Cyanobacteria). Aquat. Insects 11, 201–208 (1989).
Google Scholar 
Tang, T., Niu, S. Q. & Dudgeon, D. Responses of epibenthic algal assemblages to water abstraction in Hong Kong streams. Hydrobiologia 703(1), 225–237. https://doi.org/10.1007/s10750-012-1362-z (2013).Article 
CAS 

Google Scholar 
Maheshwari, K., Vashistha, J., Paulose, P. V. & Agarwal, T. Seasonal changes in phytoplankton community of lake Ramgarh, India. Int. J. Curr. Microbiol. Appl. Sci. 4(11), 318–330 (2015).CAS 

Google Scholar 
Luttenton, M. R., & Baisden, C. The relationships among disturbance, substratum size and periphyton community structure. In Advances in Algal Biology: A Commemoration of the Work of Rex Lowe 111–117. (Springer, 2006).Uehlinger, U. Spatial and temporal variability of periphyton biomass in a prealpine river (Necker, Switzerland). Arch. Fur. Hydrobiol. 123, 219–237 (1991).
Google Scholar 
Hill, W. R. Effects of light. In Algal Ecology in Freshwater Benthic Ecosystems. 121–148 (eds. Stevenson, R. J., Bothwell, M. L., Lowe, R. L.) (Academic Press, 1996).DeNichola, D. M. Periphyton responses to temperature at different ecological levels. In Algal Ecology in Freshwater Benthic Ecosystems. (eds. Stevenson, R. J., Bothwell, M. L., Lowe, R. L.) 149–181 (Academic Press, 1996).O’Reilly, C. M. Seasonal dynamics of periphyton in a large tropical lake. Hydrobiologia 553, 293–301. https://doi.org/10.1007/s10750-005-0878-x (2006).Article 

Google Scholar 
Borduqui, M. & Ferragut, C. Factors determining periphytic algae succession in a tropical hypereutrophic reservoir. Hydrobiologia 683, 109–122. https://doi.org/10.1007/s10750-011-0943-6 (2012).Article 
CAS 

Google Scholar 
De Souza, M. L., Pellegrini, B. G. & Ferragut, C. Periphytic algal community structure in relation to seasonal variation and macrophyte richness in a shallow tropical reservoir. Hydrobiologia 755, 183–196. https://doi.org/10.1007/s10750-015-2232-2 (2015).Article 

Google Scholar 
Prowse, T. D. River-ice hydrology. In Encyclopedia of Hydrological Sciences, vol. 4 (ed. Anderson, M. G.). (Wiley, 2005).Rusanov, A. G., Stanislavskaya, E. V. & Ács, É. Periphytic algal assemblages along environmental gradients in the rivers of the Lake Ladoga basin, Northwestern Russia: Implication for the water quality assessment. Hydrobiologia 695(1), 305–327 (2012).CAS 

Google Scholar 
Sofi, M. S., Hamid, A., Bhat, S. U., Rashid, I. & Kuniyal, J. C. Impact evaluation of the run-of-river hydropower projects on the water quality dynamics of the Sindh River in the Northwestern Himalayas. Environ. Monit. Assess. 194(9), 1–6 (2022).
Google Scholar 
MCCormick, P. V. Resource competition and species coexistence in freshwater algal assemblages. In Algal ecology—Freshwater Benthic Ecosystems (eds. Stevenson, R. J., Bothwell, M. L., Lowe, R. L.) 229–252 (Academic, 1996).Hillebrand, H., Worm, B. & Lotze, H. K. Marine microbenthic community structure regulated by nitrogen loading and grazing pressure. Mar. Ecol. Prog. Ser. 204, 27–38 (2000).CAS 

Google Scholar  More

Data were analyzed from samples collected, processed, and published previously [21, 25, 29] and have been summarized here. The present analysis, which consisted of sequence data processing, the calculation of taxon-specific isotopic signatures, and subsequent analyses, reflects original work.Sample collection and isotope incubationTo generate experimental data, three replicate soil samples were collected from the top 10 cm of plant-free patches in four ecosystems along the C. Hart Merriam elevation gradient in Northern Arizona. From low to high elevation, these sites are located in the following environments: desert grassland (GL; 1760 m), piñon-pine juniper woodland (PJ; 2020 m), ponderosa pine forest (PP; 2344 m), and mixed conifer forest (MC; 2620 m). Soil samples were air-dried for 24 h at room temperature, homogenized, and passed through a 2 mm sieve before being stored at 4 °C for another 24 h. This produced three distinct but homogenous soil samples from each of the four ecosystems that were subject to experimental treatments. Three treatments were applied to bring soils to 70% water-holding capacity: water alone (control), water with glucose (C treatment; 1000 µg C g−1 dry soil), or water with glucose and a nitrogen source (CN treatment; [NH4]2SO4 at 100 µg N g−1 dry soil). To track growth through isotope assimilation, both 18O-enriched water (97 atom %) and 13C-enriched glucose (99 atom %) were used. In all treatments isotopically heavy samples were paired with matching “light” samples that received water with a natural abundance isotope signatures. For 18O incubations, this design resulted in three soil samples per ecosystem per treatment (across four ecosystems and three treatments, n = 36) while 13C incubations were limited to only C and CN treatments (n = 24). Previous analyses suggest that three replicates is sufficient to detect growth of 10 atom % 18O in microbial DNA with a power of 0.6 and a growth of 5 atom % 18O with a power of 0.3 (12 and 6 atom % respectively for 13C) [30]. All soils were incubated in the dark for one week. Following incubation, soils were frozen at −80 °C for one week prior to DNA extraction.Quantitative stable isotope probingThe procedure of qSIP (quantitative stable isotope probing) is described here but has been applied to these samples as previously published [17, 21, 25]. DNA extraction was performed on soils using a DNeasy PowerSoil HTP 96 Kit (MoBio Laboratories, Carlsbad, CA, USA) and following manufacturer’s protocol. Briefly, 0.25 g of soils from each sample were carefully added to deep, 96-well plates containing zirconium dioxide beads and a cell lysis solution with sodium dodecyl sulfate (SDS) and shaken for 20 min. Following cell lysis, supernatant was collected and centrifuged three times in fresh 96-well plates with reagents separating DNA from non-DNA organic and inorganic materials. Lastly, DNA samples were collected on silica filter plates, rinsed with ethanol and eluted into 100 µL of a 10 mM Tris buffer in clean 96-well plates. To quantify the degree of 18O or 13C isotope incorporation into bacterial DNA (excess atom fraction or EAF), the qSIP protocol [31] was used, though modified slightly as reported previously [21, 24, 32]. Briefly, microbial growth was quantified as the change in DNA buoyant density due to incorporation of the 18O or 13C isotopes through the method of density fractionation by adding 1 µg of DNA to 2.6 mL of saturated CsCl solution in combination with a gradient buffer (200 mM Tris, 200 mM KCL, 2 mM EDTA) in a 3.3 mL OptiSeal ultracentrifuge tube (Beckman Coulter, Fullerton, CA, USA). The solution was centrifuged to produce a gradient of increasingly labeled (heavier) DNA in an Optima Max bench top ultracentrifuge (Beckman Coulter, Brea, CA, USA) with a Beckman TLN-100 rotor (127,000 × g for 72 h) at 18 °C. Each post-incubation sample was thus converted from a continuous gradient into approximately 20 fractions (150 µL) using a modified fraction recovery system (Beckman Coulter). The density of each fraction was measured with a Reichart AR200 digital refractometer (Reichert Analytical Instruments, Depew, NY, USA). Fractions with densities between 1.640 and 1.735 g cm−3 were retained as densities outside this range generally did not contain DNA. In all retained fractions, DNA was cleaned and purified using isopropanol precipitation and the abundance of bacterial 16S rRNA gene copies was quantified with qPCR using primers specific to bacterial 16S rRNA genes (Eub 515F: AAT GAT ACG GCG ACC ACC GAG TGC CAG CMG CCG CGG TAA, 806R: CAA GCA GAA GAC GGC ATA CGA GGA CTA CVS GGG TAT CTA AT). Triplicate reactions were 8 µL consisting of 0.2 mM of each primer, 0.01 U µL−1 Phusion HotStart II Polymerase (Thermo Fisher Scientific, Waltham, MA), 1× Phusion HF buffer (Thermo Fisher Scientific), 3.0 mM MgCl2, 6% glycerol, and 200 µL of dNTPs. Reactions were performed on a CFX384 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) under the following cycling conditions: 95 °C at 1 min and 44 cycles at 95 °C (30 s), 64.5 °C (30 s), and 72 °C (1 min). Separate from qPCR, retained sample-fractions were subject to a similar amplification step of the 16S rRNA gene V4 region (515F: GTG YCA GCM GCC GCG GTA A, 806R: GGA CTA CNV GGG TWT CTA AT) in preparation for sequencing with the same reaction mix but differing cycle conditions – 95 °C for 2 min followed by 15 cycles at 95 °C (30 s), 55 °C (30 s), and 60 °C (4 min). The resulting 16S rRNA gene V4 amplicons were sequenced on a MiSeq sequencing platform (Illumina, Inc., San Diego, CA, USA). DNA sequence data and sample metadata have been deposited in the NCBI Sequence Read Archive under the project ID PRJNA521534.Sequence processing and qSIP analysisIndependently from previous publications, we processed raw sequence data of forward and reverse reads (FASTQ) within the QIIME2 environment [33] (release 2018.6) and denoised sequences within QIIME2 using the DADA2 pipeline [34]. We clustered the remaining sequences into amplicon sequence variants (ASVs, at 100% sequence identity) against the SILVA 138 database [35] using a pre-trained open-reference Naïve Bayes feature classifier [36]. We removed samples with less than 3000 sequence reads, non-bacterial lineages, and global singletons and doubletons. We converted ASV sequencing abundances in each fraction to the number of 16S rRNA gene copies per gram dry soil based on qPCR abundances and the known amount of dry soil equivalent added to the initial extraction. This allowed us to express absolute population densities, rather than relative abundances. Across all replicates, we identified 114 543 unique bacterial ASVs.We calculated the 18O and 13C excess atom fraction (EAF) for each bacterial ASV using R version 4.0.3 [37] and data.table [38] with custom scripts available at https://www.github.com/bramstone/. Negative enrichment values were corrected using previously published methods [17]. ASVs that appeared in less than two of the three replicates of an ecosystem-treatment combination (n = 3) and less than three density fractions within those two replicates were removed to avoid assigning spurious estimates of isotope enrichment to infrequent taxa. Any ASVs filtered out of one ecosystem-treatment group were allowed to be present in another if they met the frequency threshold. Applying these filtering criteria, we limited our analysis towards 3759 unique bacterial ASVs which accounted for a small proportion of the total diversity but represented 68.0% of all sequence reads, and encompassed most major bacterial groups (Supplementary Fig. 1).Analysis of life history strategies and nutrient responseAll statistical tests were conducted in R version 4.0.3 [37]. We assessed the ability of phylum-level assignment of life history strategy to predict growth in response to C and N addition, as proxied by the incorporation of heavy isotope during DNA replication [39, 40]. Phylum-level assignments (Table 1) were based on the most frequently observed behavior of lineages with a representative phylum (or subphylum) as compiled previously [23]. We averaged 18O EAF values of bacterial taxa for each treatment and ecosystem and then subtracted the values in control soils from values in C-amended soils to determine C response (∆18O EAFC) and from the 18O EAF of bacteria in CN-amended soils to determine C and N response (Δ18O EAFCN). Because an ASV must have a measurable EAF in both the control and treatment for a valid Δ18O EAF to be calculated, we were only able to resolve the nutrient response for 2044 bacterial ASVs – 1906 in response to C addition and 1427 in response to CN addition.We used Gaussian finite mixture modeling, as implemented by the mclust R package [41], to demarcate plausible multi-isotopic signatures for oligotrophs and copiotrophs. For each treatment, we calculated average per-taxon 13C and 18O EAF values. To compare both isotopes directly, we divided 18O EAF values by 0.6 based on the estimate that this value (designated as µ) represents the fraction of oxygen atoms in DNA derived from the 18O-water, rather than from 16O within available C sources [42]. Two mixture components, corresponding to oligotrophic and copiotrophic growth modes, were defined using the Mclust function using ellipsoids of equal volume and shape. We observed several microorganisms with high 18O enrichment but comparatively low 13C enrichment, potentially indicating growth following the depletion of the added glucose, and that were reasonably clustered as oligotrophs in our mixture model.We tested how frequently mixture model clustering of each microorganism’s growth (based on average 18O–13C EAF in a treatment) could predict its growth across replicates (n = 12 in each treatment—although individual). We applied the treatment-level mixture models defined above to the per-taxon isotope values in each replicate, recording when a microorganism’s life history strategy in a replicate agreed with the treatment-level cluster, and when it didn’t. We used exact binomial tests to test whether the number of “successes” (defined as a microorganism being grouped in the same life history category as its treatment-level cluster) was statistically significant. To account for type I error across all individual tests (one per ASV per treatment), we adjusted P values in each treatment using the false-discovery rate (FDR) method [43].To determine the extent that life history categorizations may be appropriately applied at finer levels of taxonomic resolution, we constructed several hierarchical linear models using the lmer function in the nlme package version 3.1-149 [44]. To condense growth information from both isotopes into a single analysis, 18O and 13C EAF values were combined into a single variable using principal components analysis separately for each treatment. Across the C and CN treatments, the first principal component (PC1) was able to explain – respectively – 86% and 91% of joint variation of 18O and 13C EAF values. In all cases, we applied PC1 as the response variable and treated taxonomy and ecosystem as random model terms to limit the potential of pseudo-replication to bias significance values. We used likelihood ratio analysis and Akaike information criterion (AIC) values to compare models where life history strategy was determined based on observed nutrient responses at different taxonomic levels (Eq. 1) against a model with the same random terms but without any life history strategy data (Eq. 2). Separate models were applied to each treatment. To reduce model overfitting, we removed families represented by fewer than three bacterial ASVs as well as phyla represented by only one order. In addition, we removed bacterial ASVs with unknown taxonomic assignments (following Morrissey et al. [21]). This limited our analysis to 1 049 ASVs in the C amendment and 984 in the CN amendment.$${{{{{rm{PC}}}}}}{1}_{{18{{{{{rm{O}}}}}} – 13{{{{{rm{C}}}}}}}}sim {{{{{rm{strategy}}}}}} + 1|{{{{{rm{phylum}}}}}}/{{{{{rm{class}}}}}}/{{{{{rm{order}}}}}}/{{{{{rm{family}}}}}}/{{{{{rm{genus}}}}}}/{{{{{rm{eco}}}}}}$$
(1)
$${{{{{rm{PC}}}}}}{1}_{{18{{{{{rm{O}}}}}} – 13{{{{{rm{C}}}}}}}}sim 1 + 1|{{{{{rm{phylum}}}}}}/{{{{{rm{class}}}}}}/{{{{{rm{order}}}}}}/{{{{{rm{family}}}}}}/{{{{{rm{genus}}}}}}/{{{{{rm{eco}}}}}}$$
(2)
Here, life history strategy was defined at each taxonomic level using the mixture models above and based on the mean 18O and 13C EAF values of each bacterial lineage (Supplemental Fig. 2). We compared these models with the no-strategy model (Eq. 2) directly using likelihood ratio testing. 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

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

All the materials followed relevant institutional and national guidelines and legislation.MitesWe used a T. kanzawai population collected from trifoliate orange trees (Poncirus trifoliata [L.] Raf.) in 2018 in Kyoto, Japan, and a T. urticae population collected from chrysanthemum plants (Chrysanthemum morifolium Ramat.) in 1998 in Nara, Japan. These populations were reared on adaxial surfaces of kidney bean (Phaseolus vulgaris L.) primary leaves, which were pressed onto water-saturated cotton in Petri dishes (90 mm diameter, 14 mm depth). The water-saturated cotton served as a barrier to prevent mites from escaping. The dishes were maintained at 25 °C, 50% relative humidity, and a 16L:8D photoperiod. All experiments were conducted under these conditions. We only used mated adult females (i.e., the dispersal stage) of T. kanzawai or T. urticae mites.CaterpillarsWe used caterpillars of four lepidopteran species: Bombyx mori L., P. Xuthus, Spodoptera litura Fabricius and T. oldenlandiae. We collected eggs and larvae of T. oldenlandiae from C. japonica in 2021 in Kyoto, Japan, and reared them on C. japonica leaves until pupation. Theretra oldenlandiae shares Vitaceae host plants with T. kanzawai and T. urticae8,15. We collected eggs and larvae of P. xuthus from Ptelea trifoliata in 2021 in Kyoto, Japan, and reared them on Citrus unshiu Markov. leaves until pupation. Papilio. xuthus and T. kanzawai share P. trifoliata as a host plant in Kyoto (Kinto, personal observation).We obtained commercial populations of the B. mori Kinshu × Showa strain (Ueda-sanshu Co., Ltd, Nagano, Japan) or the w1-pnd strain. We reared B. mori larvae on an artificial diet produced at the Kyoto Institute of Technology. Although T. kanzawai use Morus alba, a food plant for the B. mori strain, the mite and the strain never encounter one another in the wild, because the B. mori strain has been domesticated for hundreds of years.We obtained a sub-cultured population of S. litura from the Kyoto Institute of Technology. We reared first to fourth instars of S. litura on an artificial diet (Insecta LFM, Nosan Insect Materials, Kanagawa, Japan), while final instars were fed P. vulgaris leaves. Because S. litura feeds on various wild and cultivated plants22,23, it may share some host plants with T. kanzawai and T. urticae, both of which also feed on many host plant species8,9,10.We reared caterpillars of T. oldenlandiae, P. xuthus, and S. litura in 900 mL transparent plastic cups and caterpillars of B. mori in transparent plastic containers (140 × 220 × 35 mm). All caterpillars were maintained under the same laboratory conditions described above.PlantsWe used several parts of P. vulgaris plants in the following experiments. This species is a preferred food for both mite species16,17 and S. litura24, but the other three caterpillar species do not feed on it (Kinto, personal observation). We thus used P. vulgaris rather than shared host plants, because some caterpillars and mites (T. urticae and P. xuthus, for example) do not share any host plant.Avoidance of caterpillar traces on leaf surfaces by spider mitesTo examine whether spider mites avoid settling on host plant surfaces bearing caterpillar traces, we conducted dual-choice tests using paired adjacent leaf squares with and without caterpillar traces. We did not use whole plants because, in practice, it was difficult to induce caterpillar traces on whole plants. We used two spider mite species (T. kanzawai and T. urticae) and four caterpillar species (T. oldenlandiae, P. xuthus, B. mori, and S. litura). We cut a 10 × 20 mm leaf piece from a fully expanded primary kidney bean leaf and then cut the piece into two equal squares (10 × 10 mm). To introduce caterpillar traces to one square, we arranged them on a separate piece of paper towel on water-saturated cotton. This procedure was necessary because the caterpillars used were larger than individual leaf squares. Then we placed a fourth or final instar caterpillar on the squares and induced the caterpillar to walk across every leaf square three times (Fig. 1a). We carefully removed all caterpillar-produced silk threads from the squares. Within 30 min, we arranged the square (trace +) to touch against the other square (trace −) on water-saturated cotton in a Petri dish. Subsequently, a 2- to 4-day-old mated adult female of T. kanzawai or T. urticae was introduced onto a pointed piece of Parafilm in contact with both leaf edges using a fine brush (Fig. 1a). We recorded the leaf square onto which the mite had settled at 2 h after its introduction, as preliminary observations confirmed that all females would settle on a particular leaf within that period. Each female mite and pair of leaf squares were used only once. All tests described below were conducted between 13:00 and 17:00 h, when adult female spider mites actively disperse by walking. There were 14 replicates using traces of T. oldenlandiae, 48 of P. xuthus, 20 of B. mori, and 26 of S. litura for T. kanzawai, as well as 18, 32, 16, and 47, respectively, for T. urticae. Data were subjected to two-tailed binomial tests with the common null hypothesis that a spider mite would settle on the two squares with equal probability (i.e., 0.5).Figure 1(a) Procedure used to observe avoidance of caterpillar traces by spider mites. (b) Experimental setup used to observe avoidance of B. mori traces on plant stems by T. kanzawai. (c) Experimental setup used to observe avoidance of B. mori trace extracts by T. kanzawai.Full size imageDuration of B. mori trace avoidance by T. kanzawai
To examine whether the effects of caterpillar traces on spider mite avoidance decline over time, we used T. kanzawai mites and B. mori caterpillars. We used B. mori because populations can be easily maintained over many generations. We prepared bean leaf squares with B. mori traces in the same manner descried above and preserved the traced square on water-saturated cotton for 0 h (n = 30), 24 h (n = 29), 48 h (n = 28), or 72 h (n = 28). Then we arranged the square (trace +) to lie in close proximity to the control square (trace −) that had been preserved for the same periods of time. Then we compared the avoidance response of T. kanzawai females in the same manner described above.Avoidance of B. mori traces on plant stems by T. kanzawai
To examine whether T. kanzawai females avoid walking along plant stems bearing caterpillar traces, we used Y-shaped kidney bean stems (Fig. 1b). We cut symmetric bean plants ca. 15 days after sowing from their base and inserted them perpendicularly into a 5 mL glass bottle filled with water and wet cotton. To induce caterpillar traces on one branch of the stem, we allowed a silkworm to crawl from the branching point to the far end of one branch three times for each stem (n = 20). Then we introduced a T. kanzawai adult female at a release point 35 mm below the branch point (Fig. 1b). We recorded the branch along which the female walked to the far end. Each female mite and each Y-shaped stem were used only once. The numbers of females were compared using binomial tests in the same manner described above.Avoidance of B. mori trace extracts by T. kanzawai
To extract chemical traces of caterpillar, we introduced 10 third instar B. mori to a glass Petri dish (120 mm diameter, 60 mm depth). After 1 h, we removed all caterpillars and washed the inside bottom of the dish with 1.0 mL acetone. We replicated the procedure twice using different individuals to combine all extracts and to acquire enough extract for the following experiment.To examine avoidance of B. mori trace extracts by T. kanzawai females, we conducted dual-choice experiments using T-shaped pathways of filter paper (35 × 35 mm; width, 2 mm; Fig. 1c). Using disposable micropipettes (Drummond Scientific Co., PA, USA), 1.75 caterpillar equivalents (i.e., 60 µL) of acetone extract were applied to an alternately selected branch (17.5 mm long) of each pathway (i.e., 0.10 caterpillar equivalent/mm), with control acetone applied to the other branch. We applied each solution dropwise at the junction point to minimize mixing. After evaporating the solvent from those pathways, we perpendicularly suspended them (Fig. 1c) and introduced an adult female mite at 2 days post-maturation onto the bottom of each pathway using a fine brush and recorded the branch along which the female first walked to the far end. Each female mite and each T-shaped filter paper were used only once, with 19 replicates. Each female mite made a choice within 10 min. The avoidance response of T. kanzawai was analysed in the same manner described above.Indirect effects of B. mori traces on T. kanzawai via plantsTo determine whether B. mori traces on plants indirectly affect the performance of T. kanzawai on plants, we introduced 70–80 randomly selected quiescent female deutonymphs of T. kanzawai onto kidney bean leaf disks. Immediately after synchronized adult emergence, we introduced the same number of adult males to allow mating; the detailed procedure is described elsewhere25. After 24 h, we transferred the females singly onto 10 × 10 mm bean leaf squares with or without B. mori traces prepared as described above. Because the number of eggs laid within a certain period is considered the most sensitive performance index of spider mite females26,27, any plant-mediated indirect interaction, such as defence induction in response to caterpillar traces, should result in lower egg numbers laid by the test females. We counted the eggs laid on the leaf squares 24 h after their introduction. One female that laid no eggs during the 24 h period (n = 1, trace +) was excluded from the analysis. We obtained 33 and 36 replicates for the trail+ and trail– conditions, respectively. We compared the numbers of eggs laid on leaves with and without B. mori traces using a generalized linear model with a Poisson error distribution using the SAS 9.22 software (SAS Institute Inc., Cary, NC, USA).EthicsThis article does not contain any studies with human participants or animals. 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

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