Ecology

Lehtonen12 presents three simple models with the same broad structure: a single mutant individual with divergent mating behaviour arises in a population of ‘residents’ that all play the same strategy, and the success of that mutant is then followed (Figs. 1, 2). Specifically, Lehtonen investigates the fitness benefits of increased mating for mutant males in comparison to mutant females. Two important parameters can be varied: (i) the degree of anisogamy (defined here as the ratio of sperm number to egg number), which captures how divergent males and females are in the size (and thus number) of gametes they produce, and (ii) the efficiency of fertilisation, which determines how easily gametes can find and fuse with each other. If fertilisation is highly efficient, then gametes of the less numerous type will achieve nearly full fertilisation; on the other hand, inefficient fertilisation can result in gametes of both sexes going unfertilised.Fig. 2: Structure of the three models of Lehtonen12, showing differences in mating behaviour between resident males (green), resident females (blue) and mutant males and females (both yellow).For illustration, we suppose that females produce four eggs each and males produce eight sperm (the anisogamy ratio in nature is typically much higher). In Model 1, resident individuals spawn monogamously in a ‘nest’ (black outline), whereas mutant males and females can bring additional partners to their nest to spawn in a group. In Model 2, resident individuals divide their gametes equally among m spawning groups, each consisting of m individuals of each sex (shown here with m = 2). Mutant males and females instead divide their gametes among a larger or smaller number of groups, mmutant (shown here with mmutant = 4). In Model 3, there is a further sex asymmetry in addition to anisogamy: Fertilisation takes place inside the female’s body. Resident individuals mate with m partners (shown here with m = 2), whereas mutant males and females mate with a larger or smaller number of partners, mmutant (shown here with mmutant = 4).Full size imageIn the first two models, fertilisation is external and no assumptions are made about pre-existing differences between the sexes apart from the number of gametes they produce. In other words, males and females are identical except that males produce sperm in greater numbers than females produce eggs. In Model 1, resident individuals are assumed to mate monogamously, whereas a mutant can monopolise multiple partners of the opposite sex (Fig. 2). Importantly, both male and female mutants can bring additional partners back to their ‘nest’ to spawn in a group. When fertilisation is highly efficient, females can fertilise all of their eggs by bringing back a single male, and there is simply no benefit (in this model) of seeking further partners (Fig. 1A). In contrast, anisogamy means that males always produce at least some gametes in excess, and thus can benefit from seeking additional mates. When fertilisation is inefficient, however, both sexes benefit from increasing the concentration of opposite-sex gametes at their ‘nest’ (Fig. 1B). This latter benefit is sex-symmetric, whereas the former continues to apply only to males. As a consequence, the Bateman gradients are always steeper for males than for females (Fig. 1A, B), confirming Bateman’s argument.Model 2 similarly assumes external fertilisation, but in this case the resident males and females meet in groups consisting of m individuals of each sex (Fig. 2). Fertilisation occurs via group spawning. It is assumed that each resident individual divides its gametes evenly across M groups, whereas mutant individuals can instead spread their gametes over a larger or smaller number of groups (note that the author assumes that M = m, but this assumption could be relaxed without undermining the core argument). Spreading gametes out across a larger number of spawning groups does not increase the concentration of opposite-sex gametes they encounter (Fig. 2). However, a mutant that spreads its gametes more widely reduces the density of its own gametes across those groups in which it spawns. This in turn results in there being more opposite-sex gametes for each gamete of the mutant’s sex in those groups. For example, in Fig. 2, mutant males spawn in twice as many groups as resident males and thereby halve the density of their own sperm in each group. The resulting egg-to-sperm ratio of (frac{4}{6}=frac{2}{3}) is more favourable than the ratio of (frac{4}{8}=frac{1}{2}) that the resident males experience. Mutant females can similarly increase local sperm-to-egg ratios by spreading their eggs over more groups. However, in contrast to males, this only leads to fitness benefit if fertilisation is inefficient, and even then the benefit to females is very modest (scarcely perceptible in Fig. 1D). Gamete spreading reduces wasteful competition among the mutants’ own gametes for fertilisation. Such ‘local’ gamete competition, like gamete competition more generally, is stronger among sperm than among eggs because sperm are more numerous under anisogamy13,14. Consequently, as in Model 1, Bateman gradients are always steeper in males (Fig. 1C, D). Recall that the results of the above models emerge in the absence of any assumptions beyond the sex difference in the number of gametes produced.The third and final model allows for a further pre-existing difference between the sexes in addition to anisogamy: internal fertilisation, which is common and widespread in animals (Fig. 2)15. Each female is assumed to mate with m males, while each male divides his gametes evenly among m females. As in the previous two models, males benefit more than females from additional matings under most conditions. However, in the particular case where fertilisation is highly inefficient and the ratio of sperm to eggs is not too large, the pattern can theoretically reverse, such that female Bateman gradients exceed their male counterparts (Fig. 1F). The reason is that the effects of gamete concentration are asymmetric under internal fertilisation: Multiple mating by a female increases the local concentration of sperm its eggs experience, whereas a male’s multiple mating does not increase the concentration of eggs around its sperm (Fig. 2). Under conditions of severe sperm limitation—due to both weak anisogamy and highly inefficient fertilisation—this can lead to females benefitting more from additional matings than males (Fig. 1F). Although intriguing, it is unclear whether this finding has any empirical relevance, as sperm limitation is probably rarely severe in internal fertilisers. Under more realistic conditions of moderate to high fertilisation rates, sex differences in the degree of local gamete competition once again become decisive, and male Bateman gradients exceed their female counterparts (Fig. 1E). More

Hauer, M. E. et al. Sea-level rise and human migration. Nat. Rev. Earth Environ. 1, 28–39 (2020).ADS 
Article 

Google Scholar 
Duy, P., Chapman, L., Tight, M., Thuong, L. & Linh, P. Urban resilience to floods in coastal cities: Challenges and opportunities for Ho Chi Minh city and other emerging cities in southeast Asia. J. Urban Plan. Dev. 144, 05017018 (2018).Article 

Google Scholar 
Magno, R. et al. Semi-automatic operational service for drought monitoring and forecasting in the Tuscany region. Geosciences 8, 49 (2018).ADS 
Article 

Google Scholar 
Rico, A., Olcina, J., Baños, C., Garcia, X. & Sauri, D. Declining water consumption in the hotel industry of mass tourism resorts: Contrasting evidence for Benidorm, Spain. Curr. Issues Tour. 23, 770–783 (2020).Article 

Google Scholar 
Hasnat, G. T., Kabir, M. A. & Hossain, M. A. Major environmental issues and problems of South Asia, particularly Bangladesh. Handb. Environ. Mater. Manag., 1–40 (2018).Neumann, B., Vafeidis, A. T., Zimmermann, J. & Nicholls, R. J. Future coastal population growth and exposure to sea-level rise and coastal flooding—A global assessment. PLoS One 10, e0118571 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Cao, A. et al. Future of Asian Deltaic Megacities under sea level rise and land subsidence: Current adaptation pathways for Tokyo, Jakarta, Manila, and Ho Chi Minh City. Curr. Opin. Environ. Sustain. 50, 87–97 (2021).Article 

Google Scholar 
Rahmasary, A. N. et al. Overcoming the challenges of water, waste and climate change in Asian cities. Environ. Manag. 63, 520–535 (2019).ADS 
Article 

Google Scholar 
Smol, M., Adam, C. & Preisner, M. Circular economy model framework in the European water and wastewater sector. J. Mater. Cycles Waste Manag. 22, 682–697 (2020).Article 

Google Scholar 
Islam, M. F., Bhattacharya, B. & Popescu, I. Flood risk assessment due to cyclone-induced dike breaching in coastal areas of Bangladesh. Nat. Hazards Earth Syst. Sci. 19, 353–368 (2019).ADS 
Article 

Google Scholar 
Salim, M. A. & Siswanto, A. B. Kajian Penanganan Dampak Banjir Kabupaten Pekalongan. Rang Tek. J. 4, 295–303 (2021).Article 

Google Scholar 
Endo, A. et al. Describing and visualizing a water–energy–food nexus system. Water 10, 1245 (2018).Article 

Google Scholar 
Gurdak, J. J., Geyer, G. E., Nanus, L., Taniguchi, M. & Corona, C. R. Scale dependence of controls on groundwater vulnerability in the water–energy–food nexus, California Coastal Basin aquifer system. J. Hydrol. Reg. Stud. 11, 126–138 (2017).Article 

Google Scholar 
Lu, J., Lin, Y., Wu, J. & Zhang, C. Continental-scale spatial distribution, sources, and health risks of heavy metals in seafood: Challenge for the water-food-energy nexus sustainability in coastal regions?. Environ. Sci. Pollut. Res. 28, 63815–63828 (2021).CAS 
Article 

Google Scholar 
Miller-Robbie, L., Ramaswami, A. & Amerasinghe, P. Wastewater treatment and reuse in urban agriculture: Exploring the food, energy, water, and health nexus in Hyderabad, India. Environ. Res. Lett. 12, 075005 (2017).ADS 
Article 
CAS 

Google Scholar 
Taniguchi, M., Endo, A., Gurdak, J. J. & Swarzenski, P. Water-energy-food nexus in the Asia-Pacific region. J. Hydrol. 11, 1–8 (2017).
Google Scholar 
Bahri, M. Analysis of the water, energy, food and land nexus using the system archetypes: A case study in the Jatiluhur reservoir, West Java, Indonesia. Sci. Total Environ. 716, 137025 (2020).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Lubis, R., Delinom, R., Martosuparno, S. & Bakti, H. Water-Food Nexus in Citarum Watershed, Indonesia Vol. 118, 012023 (IOP Publishing, 2018).
Google Scholar 
Pawitan, H., Delinom, R. & Taniguchi, M. The human–environment sustainability in Indonesia: The case of the Citarum basin Vol. 23 (UNESCO-IHP, 2015).Carmichael, L. et al. Urban planning as an enabler of urban health: Challenges and good practice in England following the 2012 planning and public health reforms. Land Use Policy 84, 154–162 (2019).Article 

Google Scholar 
World Health Organization. Addressing the Social Determinants of Health: The Urban Dimension and the Role of Local Government (World Health Organization, 2012).
Google Scholar 
Trencher, G. & Karvonen, A. Stretching, “smart”: Advancing health and well-being through the smart city agenda. Local Environ. 24, 610–627 (2019).Article 

Google Scholar 
Yang, L. et al. Can an island economy be more sustainable? A comparative study of Indonesia, Malaysia, and the Philippines. J. Clean. Prod. 242, 118572 (2020).Article 

Google Scholar 
Choirunisa, A. K. & Giyarsih, S. R. Kajian Kerentanan Fisik, Sosial, dan Ekonomi Pesisir Samas Kabupaten Bantul Terhadap Erosi Pantai. J. Bumi Indones. 5 (2016).Gumay, A. Validity and reliability maritime English seafarers proficiency test. INFERENCE J. Engl. Lang. Teach. 3, 64–69 (2021).Article 

Google Scholar 
Tarigan, M. S. Perubahan garis pantai di wilayah pesisir perairan Cisadane, Provinsi Banten. Makara J. Sci. (2010).Pruss-Ustun, A., Corvalán, C. F., World Health Organization. Preventing Disease Through Healthy Environments: Towards an Estimate of the Environmental Burden of Disease (World Health Organization, 2006).
Google Scholar 
Baasanjargal, T., Soon-Joo, A. & Mi-Jeong, K. Comparative analysis of Indonesian Batik traditional patterns: Focused on patterns of Yogyakarta and Pekalongan in Java Island. 한복문화 22, 75–91 (2019).Rismawati, S. D., Sofiani, T. & Rahmawati, D. R. Legal culture of religious capitalism on Batik business (a case study in Pekalongan Indonesia). JL Pol. Glob. 33, 107 (2015).
Google Scholar 
Pekalongan, B. K. Kota Pekalongan dalam Angka 2021 (2021).Google Maps. Pekalongan, Central Java (2022).Sunarjo, W. A., Ilmiani, A. & Ardianingsih, A. Analisis SWOT Sebagai Pengembangan UMKM Berbasis Ekonomi Kreatif Destinasi Pariwisata Batik Kota Pekalongan. Pena J. Ilmu Pengetah. Dan Teknol. 33, 34–43 (2019).Article 

Google Scholar 
Perpustakaan Provinsi Jawa Tengah. Museum Batik Pekalongan (2017).Brzezina, N. et al. Development of organic farming in Europe at the crossroads: Looking for the way forward through system archetypes lenses. Sustainability 9, 821 (2017).Article 

Google Scholar 
Gillies, A. & Maliapen, M. Using healthcare system archetypes to help hospitals become learning organisations. J. Model. Manag. (2008).Braun, W. The System Archetypes. The Systems Modeling Workbook, 1–26 (2002).Sterman, J. System Dynamics: Systems thinking and modeling for a complex world (2002).Islam, M. & Raja, D. R. Waterlogging risk assessment: An undervalued disaster risk in coastal urban community of Chattogram, Bangladesh. Earth 2, 151–173 (2021).Article 

Google Scholar 
Brzezina, N., Kopainsky, B. & Mathijs, E. Can organic farming reduce vulnerabilities and enhance the resilience of the European food system? A critical assessment using system dynamics structural thinking tools. Sustainability 8, 971 (2016).Article 

Google Scholar 
Nguyen, N. C. & Bosch, O. J. A systems thinking approach to identify leverage points for sustainability: A case study in the Cat Ba Biosphere Reserve, Vietnam. Syst. Res. Behav. Sci. 30, 104–115 (2013).Article 

Google Scholar 
Maani, K. E. & Cavana, R. Y. Systems Thinking, System Dynamics: Managing Change and Complexity (Pearson Prentice Hall, 2007).
Google Scholar 
Braun, W. The System Archetypes-the Systems Modeling Workbook. Available Wwwu Uniklu Ac Atgossimitpapsdwbsysarch Pdf (2002).Senge, P. M. The Fifth Discipline: The Art and Practice of the Learning Organization (Currency, 2006).
Google Scholar 
Bahri, M. et al. Deliverable 3.3: Integrated model with ad-hoc systems model of urban water supply (2018).Pekalongan, B. P. P. D. K. Pekalongan dalam Angka (2021).Fajar, M., Mediani, A. & Finesa, Y. Analisis Peranan IPAL dalam Strategi Penanganan Limbah Industri Batik di Kota Pekalongan. in Prosiding Seminar Nasional Geografi UMS X 2019 (2019).Kartika, F. D. S. & Helmi, M. Meta-analysis of Community’s Adaptation Pattern with Tidal Flood in Pekalongan City, Central Java, Indonesia Vol. 125, 09001 (EDP Sciences, 2019).
Google Scholar 
Kartika, F. D. S., Helmi, M. & Amirudin, A. Analisis Perubahan Penggunaan Lahan di Wilayah Pesisir Kota Pekalongan Menggunakan Citra Lansat 8, vol. 1 (2019).Damayanti, M. & Latifah, L. Strategi Kota Pekalongan dalam pengembangan wisata kreatif berbasis industri batik. J. Pengemb. Kota 3, 100–111 (2017).Article 

Google Scholar 
Pekalongan, B. K. Kota Pekalongan dalam Angka 2002 (2002).Andreas, H., Abidin, H. Z., Sarsito, D. A. & Pradipta, D. Adaptation of ‘early climate change disaster’ to the Northern coast of Java Island Indonesia. Eng. J. 22, 207–219 (2018).Article 

Google Scholar 
Marfai, M. A. et al. The impact of tidal flooding on a coastal community in Semarang, Indonesia. Environmentalist 28, 237–248 (2008).Article 

Google Scholar 
Chaussard, E., Amelung, F., Abidin, H. & Hong, S.-H. Sinking cities in Indonesia: ALOS PALSAR detects rapid subsidence due to groundwater and gas extraction. Remote Sens. Environ. 128, 150–161 (2013).ADS 
Article 

Google Scholar 
Andreas, H., Abidin, H. Z., Sarsito, D. A. & Pradipta, D. Remotes Sensing Capabilities on Land Subsidence and Coastal Water Hazard and Disaster Studies Vol. 500, 012036 (IOP Publishing, 2020).
Google Scholar 
Shofiana, R., Subardjo, P. & Pratikto, I. Analisis perubahan penggunaan lahan di wilayah pesisir Kota pekalongan menggunakan data landsat 7 etm+. J. Mar. Res. 2, 35–43 (2013).
Google Scholar 
Wijaya, A. Analisis Dinamika Pola Spasial Penggunaan Lahan Pada Wilayah Terdampak Kenaikan Muka Air Laut di Kota Pekalongan (2017).El-Fath, D. D. I., Atmodjo, W., Helmi, M., Widada, S. & Rochaddi, B. Analisis Spasial Area Genangan Banjir Rob Setelah Pembangunan Tanggul di Kabupaten Pekalongan, Jawa Tengah. Indones. J. Oceanogr. 4, 96–110 (2022).
Google Scholar 
Novita, M. G., Helmi, M., Widiaratih, R., Hariyadi, H. & Wirasatriya, A. Mengkaji Area Genangan Banjir Pasang Terhadap Penggunaan Lahan Pesisir Tahun 2020 Menggunakan Metode Geospasial di Kabupaten Pekalongan, Provinsi Ja. Indones. J. Oceanogr. 3, 14–26 (2021).Article 

Google Scholar 
Salim, M. A. Penanganan Banjir dan Rob di Wilayah Pekalongan. J. Tek. Sipil 11, 15–23 (2018).
Google Scholar 
Jumatiningrum, N. & Indrayati, A. Strategi Adaptasi Masyarakat Kelurahan Bandengan Kecamatan Pekalongan Utara dalam Menghadapi Banjir Pasang Air Laut (Rob). Edu Geogr. 9, 136–143 (2021).
Google Scholar 
BNPB. Data Kebencanaan Nasional (BNPB, 2021).
Google Scholar 
Giampietro, M., Aspinall, R. J., Ramos-Martin, J. & Bukkens, S. G. Resource Accounting for Sustainability Assessment: The Nexus Between Energy, Food, Water and Land Use (Routledge, 2014).Book 

Google Scholar 
Meadows, D. H., Randers, J. & Meadows, D. L. The Limits to Growth (1972) (Yale University Press, 2013).MATH 

Google Scholar 
Albrecht, T., Crootof, A. & Scott, C. Trends in the development of water–energy–food nexus methods (2017).Leck, H., Fitzpatrick, D. & Burchell, K. Energy, water and food: Towards a critical nexus approach. in Handbook on the Geographies of Energy (Edward Elgar Publishing, 2017).Scott, C. A., Kurian, M. & Wescoat, J. L. The water–energy–food nexus: Enhancing adaptive capacity to complex global challenges. in Governing the Nexus 15–38 (Springer, 2015).Wanty, E. E. Analisis Produksi Batik Cap Dari UKM Batik Kota Pekalongan (Studi Pada Sentra Batik Kota Pekalongan-Jawa Tengah, 2006).
Google Scholar 
Mankiw, N. G. Macroeconomics Vol. 41 (Worth Publishers, 2003).
Google Scholar 
Shen, J. & Kee, G. Development and Planning in Seven Major Coastal Cities in Southern and Eastern China (Springer, 2017).Book 

Google Scholar 
Xu, C., Haase, D., Su, M. & Yang, Z. The impact of urban compactness on energy-related greenhouse gas emissions across EU member states: Population density vs physical compactness. Appl. Energy 254, 113671 (2019).Article 

Google Scholar 
Marfai, M. A. & Cahyadi, A. Dampak bencana banjir pesisir dan adaptasi masyarakat terhadapnya di kabupaten Pekalongan (2017).Wartadesa.net. Tiga hari banjir rendam Pekalongan (2018).Google Maps. A dike in Pekalongan (n.d).Anindita, R. M., Susilowati, I. & Muhammad, F. Analisis Efektifitas Tanggul Laut di Pesisir Pekalongan Terhadap Penurunan Intensitas Banjir, vol. 2 80–88 (2020).Taniguchi, M. Groundwater and Subsurface Environments: Human Impacts in Asian Coastal Cities (Springer Science & Business Media, 2011).Book 

Google Scholar 
Baños, C. J., Hernández, M., Rico, A. M. & Olcina, J. The hydrosocial cycle in coastal tourist destinations in Alicante, Spain: Increasing resilience to drought. Sustainability 11, 4494 (2019).Article 

Google Scholar 
Sauda, R. H. & Nugraha, A. L. Kajian pemetaan kerentanan banjir rob di kabupaten pekalongan. J. Geod. Undip 8, 466–474 (2019).
Google Scholar 
Wartadesa.net. Ratusan warga Sragi masih mengungsi (2022).Buchori, I. et al. Adaptation to coastal flooding and inundation: Mitigations and migration pattern in Semarang City, Indonesia. Ocean Coast. Manag. 163, 445–455 (2018).Article 

Google Scholar 
Setiadi, R. & Nalau, J. Can urban regeneration improve health resilience in a changing climate? (2015).Isham, A., Mair, S. & Jackson, T. Wellbeing and productivity: A review of the literature (2020).Banson, K. E., Nguyen, N. C. & Bosch, O. J. Using system archetypes to identify drivers and barriers for sustainable agriculture in Africa: A case study in Ghana. Syst. Res. Behav. Sci. 33, 79–99 (2016).Article 

Google Scholar 
Lavrnić, S., Zapater-Pereyra, M. & Mancini, M. Water scarcity and wastewater reuse standards in Southern Europe: Focus on agriculture. Water. Air Soil Pollut. 228, 1–12 (2017).Article 
CAS 

Google Scholar 
Tortajada, C. & Nam Ong, C. Reused water policies for potable use (2016).Murali, R. M., Riyas, M., Reshma, K. & Kumar, S. S. Climate change impact and vulnerability assessment of Mumbai city, India. Nat. Hazards 102, 575–589 (2020).Article 

Google Scholar 
Abdullah, A. Y. M. et al. Spatio-temporal patterns of land use/land cover change in the heterogeneous coastal region of Bangladesh between 1990 and 2017. Remote Sens. 11, 790 (2019).ADS 
Article 

Google Scholar 
Ginanjar, A., Rezagama, A. & Handayani, D. S. Rencana Induk Sistem Penyediaan Air Minum Kota Pekalongan (2015).Reiblich, J., Hartge, E., Wedding, L., Killian, S. & Verutes, G. Bridging climate science, law, and policy to advance coastal adaptation planning. Mar. Policy 104, 125–134 (2019).Article 

Google Scholar 
Cook, B. I. et al. Revisiting the leading drivers of Pacific coastal drought variability in the contiguous United States. J. Clim. 31, 25–43 (2018).ADS 
Article 

Google Scholar 
Jodar-Abellan, A., Valdes-Abellan, J., Pla, C. & Gomariz-Castillo, F. Impact of land use changes on flash flood prediction using a sub-daily SWAT model in five Mediterranean ungauged watersheds (SE Spain). Sci. Total Environ. 657, 1578–1591 (2019).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Thanvisitthpon, N., Shrestha, S. & Pal, I. Urban flooding and climate change: A case study of Bangkok, Thailand. Environ. Urban. Asia 9, 86–100 (2018).Article 

Google Scholar 
Laksmi, G. S. Dampak Alih Fungsi Lahan dan Curah Hujan terhadap Banjir di Kota Pekalongan, Jawa Tengah, 382–391 (2020).Dhiman, R., VishnuRadhan, R., Eldho, T. & Inamdar, A. Flood risk and adaptation in Indian coastal cities: Recent scenarios. Appl. Water Sci. 9, 1–16 (2019).ADS 
Article 

Google Scholar 
Bahri, M. & Cremades, R. The Urban Drought Nexus Tool. Zenodo (2021). More

Eco-evolutionary model on spatial graphsWe establish an individual-based model (IBM) where individuals are structured over a trait space and a graph representing a landscape. For the sake of simplicity, we consider the case of asexual reproduction and haploid genetics29. Individuals die, reproduce, mutate and migrate in a stochastic fashion, which together results in macroscopic properties. The formulation of the stochastic IBM allows an analytical description of the dynamics at the population level, which links emergent properties to the elementary processes that generate them.The trait space ({{{{{{{mathcal{X}}}}}}}}subseteq {{mathbb{R}}}^{d}) is continuous and can be split into a neutral trait space ({{{{{{{mathcal{U}}}}}}}}) and an adaptive trait space ({{{{{{{mathcal{S}}}}}}}}). We refer to neutral traits (uin {{{{{{{mathcal{U}}}}}}}}) as traits that are not under selection, in contrast to adaptive traits (sin {{{{{{{mathcal{S}}}}}}}}), which experience selection. The graph denoted by G is composed of a set of vertices {v1,v2,…,vM} that correspond to habitat patches (suitable geographical areas), and a set of edges that constrain the movement of individuals between the habitat patches. We use the original measure of genetic differentiation for quantitative traits QST (standing for Q-statistics) in the case of haploid populations45,46. We denote the neutral trait value of the kth individual on vi as ({u}_{k}^{(i)}), the number of individuals on vi as N(i), the mean neutral trait on vi as ({overline{u}}^{(i)}), and the mean neutral trait in the metapopulation as (overline{u}). It follows that we quantify neutral differentiation QST,u as$${Q}_{ST,u}={sigma }_{B,u}^{2}/({sigma }_{B,u}^{2}+{sigma }_{W,u}^{2})$$
(1)
where ({sigma }_{B,u}^{2}={mathbb{E}}[frac{1}{M}{sum }_{i}{left({overline{u}}^{(i)}-overline{u}right)}^{2}]) denotes the expected neutral trait variance between the vertices and ({sigma }_{W,u}^{2}=frac{1}{M}mathop{sum }nolimits_{i}^{M}{mathbb{E}}left[frac{1}{{N}^{(i)}}{sum }_{k}{left({u}_{k}^{(i)}-{overline{u}}^{(i)}right)}^{2}right]) denotes the average expected neutral trait variance within vertices. We similarly quantify adaptive differentiation QST,s.Following the Gillespie update rule47, individuals with trait ({x}_{k}in {{{{{{{mathcal{X}}}}}}}}) on vertex vi are randomly selected to give birth at rate b(i)(xk) and die at rate d(N(i)) = N(i)/K, where K is the local carrying capacity. The definition of d therefore captures competition, which is proportional to the number of individuals on a vertex and does not depend on the individuals’ traits (we relax this assumption later on). The offspring resulting from a birth event inherits the parental traits, which can independently be affected by mutations with probability μ. A mutated trait differs from the parental trait by a random change that follows a normal distribution with variance ({sigma }_{mu }^{2}) (corresponding to the continuum of alleles model48). The offspring can further migrate to neighbouring vertices by executing a simple random walk on G with probability m. A schematic overview of the two different settings considered is provided in Fig. 1. Under the setting with no selection, individuals are only characterised by neutral traits so that ({{{{{{{mathcal{X}}}}}}}}={{{{{{{mathcal{U}}}}}}}}). For individuals on a vertex with trait xk ≡ uk we define b(i)(xk) ≡ b, so that the birth rate is constant. This ensures that neutral traits do not provide any selective advantage. Under the setting with heterogeneous selection, each vertex of the graph vi is labelled by a habitat type with environmental condition Θi that specifies the optimal adaptive trait value on vi. It follows that, for individuals with traits ({x}_{k}=({u}_{k},{s}_{k})in {{{{{{{mathcal{U}}}}}}}}times {{{{{{{mathcal{S}}}}}}}}) on vi, we define$${b}^{(i)}({x}_{k})equiv {b}^{(i)}({s}_{k})=b(1-p{({s}_{k}-{{{Theta }}}_{i})}^{2})$$
(2)
where p is the selection strength41. This ensures that the maximum birth rate on vi is attained for sk = Θi, which results in a differential advantage that acts as an evolutionary stabilising force. In the following we consider two habitat types denoted by I and II with symmetric environmental conditions θI and θII, so that Θi ∈ {θI, θII} and θII = − θI = θ, where θ can be viewed as the habitat heterogeneity41.Fig. 1: Graphical representation of the structure of individuals in the eco-evolutionary model.a Setting with no selection, where individuals are characterised by a set of neutral traits (uin {{{{{{{mathcal{U}}}}}}}}). The scatter plots represent a projection of the first two components of u for the individuals present on the designated vertices at time t = 1000, obtained from one simulation of the IBM. b Setting with heterogeneous selection. In this setting, individuals are additionally characterised by adaptive traits (sin {{{{{{{mathcal{S}}}}}}}}). Blue vertices favour the optimal adaptive trait value θI, while red vertices favour θII. The scatter plots represent a projection of the first component of u and s for the individuals present on the designated vertices at time t = 1000, obtained from one simulation. The majority of individuals are locally well-adapted and have an adaptive trait close to the optimal value, but some maladaptive individuals originating from neighbouring vertices are also present. m = 0.05.Full size imageDeterministic approximation of the population dynamics under no selectionThe model can be formulated as a measure-valued point process (30 and Supplementary Note). Under this formalism, we demonstrate in the Supplementary Note how the population size and the trait dynamics show a deterministic behaviour when a stabilising force dampens the stochastic fluctuations. This makes it possible to express the dynamics of the macroscopic properties with deterministic differential equations, connecting emergent patterns to the processes that generate them. In particular, in the setting of no selection, competition stabilises the population size fluctuations, and the dynamics can be considered deterministic and expressed as$${partial }_{t}{N}_{t}^{(i)}={N}_{t}^{(i)}left[b(1-m)-frac{{N}_{t}^{(i)}}{K}right]+mbmathop{sum}limits_{jne i}frac{{a}_{i,j}}{{d}_{j}}{N}_{t}^{(j)}$$
(3)
where (A={({a}_{i,j})}_{1le i,jle M}) is the adjacency matrix of the graph G and D = (d1,d2,…,dM) is a vector containing the degree of each vertex (number of edges incident to the vertex). The first term on the right-hand side corresponds to logistic growth, which accounts for birth and death events of non-migrating individuals. The second term captures the gains due to migrations, which depend on the graph topology. Assuming that all vertices with the same degree have an equivalent position on the graph, corresponding to a “mean field” approach (see Methods), one can obtain a closed-form solution from Eq. (3) (see Eq. (12)), which shows that the average population size (overline{N}) scales with ({langle sqrt{k}rangle }^{2}/langle krangle), where 〈k〉 is the average vertex degree and (langle sqrt{k}rangle) is the average square-rooted vertex degree. The quantity ({langle sqrt{k}rangle }^{2}/langle krangle), denoted as hd, relates to the homogeneity in vertex degree of the graph and can therefore be viewed as a measure negatively associated with heterogeneity in connectivity. Simulations of the IBM illustrate that hd can explain differences in population size for complex graph topologies with varying migration regimes (Fig. 2a for graphs with M = 7 vertices and Supplementary Fig. 1a for M = 9). This analytical result is connected to theoretical work on reaction-diffusion processes49 and highlights that irregular graphs (graphs whose vertices do not have the same degree) result in unbalanced migration fluxes that affect the ecological balance between births and deaths. Highly connected vertices present an oversaturated carrying capacity (N(i)  > bK, see Methods), increasing local competition and lowering total population size compared with regular graphs (Fig. 2a). Because populations with small sizes experience more drift (31 and Supplementary Fig. 2), this result indicates that graph topology affects neutral differentiation not only through population isolation, but also by affecting population dynamics.Fig. 2: Effect of and hd on average population size (overline{N}) and neutral differentiation QST,u in the setting with no selection.a Response of (overline{N}) to homogeneity in degree ({h}_{d}={langle sqrt{k}rangle }^{2}/langle krangle) for all undirected connected graphs with M = 7 vertices and m = 0.5. b Response of QST,u to average path length for similar simulations obtained with m = 0.01. c Response of QST,u to homogeneity in degree hd for the same data. In a, b, and c, each dot represents average results from 5 replicate simulations of the IBM, the colour scale corresponds to the proportion of the graphs with similar x and y-axis values (graph density), and the blue line corresponds to a linear fit. d Standardized effect of hd and on QST,u, obtained from multivariate regression models independently fitted on similar data obtained for m = 0.01 and m = 0.5. The contributions of and hd to QST,u are alike for low migration regimes. Error bars show 95% confidence intervals. Analogous results on graphs with M = 9 vertices are presented in Supplementary Fig. 1 and all regression details can be found in Supplementary Table 2.Full size imageNonetheless, the stochasticity of the processes at the individual level can propagate to the population level and substantially affect the macroscopic properties. In particular, neutral differentiation emerges from the stochastic fluctuations of the populations’ neutral trait distribution. These fluctuations complicate an analytical underpinning of the dynamics, and in this case simulations of IBM offer a straightforward approach to evaluate the level of neutral differentiation.Effect of graph topology on neutral differentiation under no selectionWe study a setting with no selection and investigate the effect of the graph topology on neutral differentiation. When migration is limited, individuals’ traits are coherent on each vertex but stochastic drift at the population level generates neutral differentiation between the vertices. Migration attenuates neutral differentiation because it has a correlative effect on local trait distributions. Following21,22,26, we expect that the intensity of the correlative effect depends on the average path length of the graph 〈l〉, defined as the average shortest path between all pairs of vertices50. For a constant number of vertices, 〈l〉 is strictly related to the mean betweenness centrality and quantifies the graph connectivity50. High 〈l〉 implies low connectivity and greater isolation of populations, and hence we expect that graphs with high 〈l〉 are associated with high differentiation levels. We consider various graphs with an identical number of vertices and run simulations of the IBM to obtain the neutral differentiation level QST,u attained after a time long enough to discard transient dynamics (see Methods). We then interpret the discrepancies in QST,u across the simulations by relating them to the underlying graph topologies.We observe strong differences in QST,u across graphs for varying m, and find that 〈l〉 explains at least 55% of the variation in QST,u across all graphs with M = 7 vertices for (Fig. 2b). Nonetheless, some specific graphs, such as the star graph, present higher levels of QST,u than expected by their average path length. To explain this discrepancy, we explore the effect of homogeneity in vertex degree hd, as we showed in Eq. (12) that it decreases population size, which should in turn increase QST,u by intensifying stochastic drift. We find that hd explains 57% of the variation for low m (Fig. 2c). However, the fit remains similar after correcting for differences in population size (see Supplementary Table 1), indicating that irregular graphs structurally amplify the isolation of populations. Unbalanced migration fluxes lead central vertices to host more individuals than allowed by their carrying capacity. This causes increased competition that results in a higher death rate, so that migrants have a lower probability of further spreading their trait. Highly connected vertices therefore behave as bottlenecks, increasing the isolation of peripheral vertices and consequently amplifying QST,u.We then evaluate the concurrent effect of 〈l〉 and hd on QST,u with a multivariate regression model that we fit independently for low and high migration regimes (Fig. 2d). The multivariate regression model explains at least 70% of the variation in QST,u for the migration regimes considered and for graphs with M = 7 vertices (see Supplementary Table 2 for details). Moreover, we find that 〈l〉 and hd have akin contributions to neutral differentiation for low m, but the effect of 〈l〉 increases for higher migration regimes while the effect of hd decreases. To ensure that these conclusions can be generalised to larger graphs, we conduct the same analysis on a subset of graphs with M = 9 vertices and find congruent results (Supplementary Fig. 1). In the absence of selection and with competitive interactions, graphs with a high average path length 〈l〉 and low homogeneity in vertex degree hd, or similarly graphs with low connectivity and high heterogeneity in connectivity, show high levels of neutral differentiation.Deterministic approximation of the population dynamics and adaptation under heterogeneous selectionWe next consider heterogeneous selection and investigate the response of adaptive differentiation to the spatial distribution of habitat types, denoted as the Θ-spatial distribution. Adaptive differentiation emerges from local adaptation, but migration destabilises adaptation as a result of the influx of maladaptive migrants. We expect that higher connectivity between vertices of similar habitat type increases the level of adaptive differentiation, because it increases the proportion of well-adapted migrants. Local adaptation can be investigated by approximating the stochastic dynamics of the trait distribution with a deterministic partial differential equation (PDE). We demonstrate under mean-field assumption how the deterministic approximation can be reduced to an equivalent two-habitat model. We analyse the reduced model with the theory of adaptive dynamics36,41 and find a critical migration threshold m⋆ that determines local adaptation. m⋆ depends on a quantity coined the habitat assortativity rΘ, and we demonstrate with numerical simulations that rΘ determines the overall adaptive differentiation level QST,s reached at steady state in the deterministic approximation.Heterogeneous selection, captured by the dependence of the birth rate on Θi, generates a stabilising force that dampens the stochastic fluctuations of the adaptive trait distribution. The dynamics of the adaptive trait distribution consequently shows a deterministic behavior and we demonstrate in the Supplementary Note and Supplementary Figs. 3 and 4 that the number of individuals on vi with traits (sin {{Omega }}subset {{{{{{{mathcal{S}}}}}}}}) can be approximated by the quantity ∫Ωn(i)(s)ds, where n(i) is a continuous function solution of the PDE$${partial }_{t}{n}_{t}^{(i)}(s)= , {n}_{t}^{(i)}(s)left[{b}^{(i)}(s)(1-m)-frac{1}{K}{int}_{{{{{{{{mathcal{S}}}}}}}}}{n}_{t}^{(i)}({{{{{{{bf{s}}}}}}}})d{{{{{{{bf{s}}}}}}}}right]\ +mmathop{sum}limits_{jne i}{b}_{j}(s)frac{{a}_{i,j}}{{d}_{j}}{n}_{t}^{(j)}(s)+frac{1}{2}mu {sigma }_{mu }^{2}{{{Delta }}}_{s}left[{b}^{(i)}(s){n}_{t}^{(i)}(s)right]$$
(4)
Equation (4) is similar to Eq. (3), except that it incorporates an additional term corresponding to mutation processes and that the birth rate is trait-dependent. We show how Eq. (4) can be reduced to an equivalent two-habitat model under mean-field assumption. The mean-field approach differs slightly from the setting with no selection because vertices are labelled with Θi. Here we assume that vertices with similar habitat types have an equivalent position on the graph (see Supplementary Fig. 5 for a graphical representation), so that all vertices with habitat type I are characterised by the identical adaptive trait distribution that we denote by ({overline{n}}^{{{{{{{{bf{I}}}}}}}}}), and are associated with the birth rate ({b}^{{{{{{{{bf{I}}}}}}}}}(s)=b(1-p{(s-{theta }_{{{{{{{{bf{I}}}}}}}}})}^{2})). Let P(I, II) denote the proportion of edges connecting a vertex vi of type II to a vertex vj of type I, and let P(I) denote the proportion of vertices vi of type I. By further assuming that habitats are homogeneously distributed on the graph so that (P({{{{{{{bf{I}}}}}}}})=P({{{{{{{bf{II}}}}}}}})=frac{1}{2}), Eq. (4) transforms into$${partial }_{t}{overline{n}}_{t}^{{{{{{{{bf{I}}}}}}}}}(s)= ,{overline{n}}_{t}^{{{{{{{{bf{I}}}}}}}}}(s)left[{b}^{{{{{{{{bf{I}}}}}}}}}(s)(1-m)-frac{1}{K}{int}_{{{{{{{{mathcal{S}}}}}}}}}{overline{n}}_{t}^{{{{{{{{bf{I}}}}}}}}}({{{{{{{bf{s}}}}}}}})d{{{{{{{bf{s}}}}}}}}right]+frac{1}{2}mu {sigma }_{mu }^{2}({{{Delta }}}_{s}{b}^{{{{{{{{bf{I}}}}}}}}}{overline{n}}_{t}^{{{{{{{{bf{I}}}}}}}}})(s)\ +frac{m}{2},[(1-{r}_{{{Theta }}}){b}^{{{{{{{{bf{II}}}}}}}}}(s){overline{n}}_{t}^{{{{{{{{bf{II}}}}}}}}}(s)+(1+{r}_{{{Theta }}}){b}^{{{{{{{{bf{I}}}}}}}}}(s){overline{n}}_{t}^{{{{{{{{bf{I}}}}}}}}}(t)]$$
(5)
(see Methods), where we define$${r}_{{{Theta }}}=2left(P({{{{{{{bf{I}}}}}}}},{{{{{{{bf{I}}}}}}}})-P({{{{{{{bf{I}}}}}}}},{{{{{{{bf{II}}}}}}}})right)$$
(6)
as the habitat assortativity of the graph, which ranges from −1 to 1. When rΘ = − 1, all edges connect dissimilar habitat types (disassortative graph), while as rΘ tends towards 1 the graph is composed of two clusters of vertices with identical habitat types (assortative graph). Eq. (5) can be analysed with the theory of adaptive dynamics36,38,41, a mathematical framework that provides analytical insights by assuming a “trait substitution process”. Following this assumption, the mutation term in Eq. (5) is omitted and the phenotypic distribution results in a collection of discrete individual types that are gradually replaced by others until evolutionary stability is reached (see Methods and36,38,41 for details). By applying the theory of adaptive dynamics, we find a critical migration rate m⋆$${m}^{star }=frac{1}{(1-{r}_{{{Theta }}})}frac{4p{theta }^{2}}{(1+3p{theta }^{2})}$$
(7)
so that when m  > m⋆, a single type of individual exists with adaptive trait ({s}^{* }=left({theta }_{{{{{{{{bf{II}}}}}}}}}+{theta }_{{{{{{{{bf{I}}}}}}}}}right)/2=0) in the steady-state (see Methods for the derivation of Eq. (7)). In this case, adaptive differentiation QST,s is nil and the average population size is given by (overline{N}=bK{(1-ptheta )}^{2}). In contrast, when m = 0 and/or rΘ = 1, all individuals are locally well-adapted with trait Θi on vi, and it follows that the average population size is higher and equal to (overline{N}=bK), while adaptive differentiation is maximal and equal to ({Q}_{ST,s}={{{{{{{rm{Var}}}}}}}}({{Theta }})/left({{{{{{{rm{Var}}}}}}}}({{Theta }})+0right)=1). When 0  m⋆, implying that individuals become equally fit in all habitats. In this case, the isolation effect of heterogeneous selection is lost and QST,u reaches a similar level as in the setting with no selection for m  > m⋆ (Fig. 5a), although QST,u is slightly higher in the setting with heterogeneous selection due to lower population size ((overline{N}=bK(1-ptheta )) vs. (overline{N}=bK), see section above and Methods). This suggests that rΘ reinforces QST,u, as assortative graphs sustain higher levels of adaptive differentiation (Figs. 3 and 4). Simulations on the path graph with varying Θ-spatial distribution support this conclusion for high migration regimes, but show the opposite relationship under low migration regimes, where the habitat assortativity rΘ decreases QST,u (Fig. 5b). Assortative graphs are composed of large clusters of vertices with similar habitats, within which migrants can circulate without fitness losses. Local neutral trait distributions become more correlated within these clusters, resulting in a decline in QST,u for assortative graphs compared with disassortative graphs. Figure 5b therefore highlights the ambivalent effect of rΘ on QST,u. rΘ reinforces QST,u by favouring adaptive differentiation, but also decreases QST,u by decreasing population isolation within clusters of vertices with the same habitat type.We compare the effect of rΘ on QST,u to the effect of the topology metrics 〈l〉 and hd found in the setting with no selection using multivariate regression analysis on simulation results obtained for different graphs with varying Θ-spatial distribution (Fig. 5d for graphs with M = 7 vertices and Supplementary Fig. 7b for M = 9). The multivariate model explains the discrepancies in QST,u across the simulations for low and high migration regimes (see Supplementary Table 3 for details), and we find that rΘ, 〈l〉, and hd contribute similarly to neutral differentiation. Hence, the effects of rΘ and the topology metrics 〈l〉 and hd add up under heterogeneous selection. A change in sign of the standardized effect of rΘ on QST,s for low and high migration regimes verifies that the ambivalent effect of rΘ on QST,u found on the path graph holds for general graph ensembles. Simulations with trait-dependent competition and simulations on realistic graphs with a continuum of habitat types equally confirm the ambivalent effect of rΘ and further support the complementary effect of 〈l〉 and hd on QST,u (see Supplementary Fig. 8). 〈l〉 and hd therefore drive neutral differentiation with and without heterogeneous selection. rΘ becomes an additional determinant of neutral differentiation under heterogeneous selection. In contrast to the non-ambivalent, positive effect of habitat assortativity on adaptive differentiation, rΘ can amplify or depress neutral differentiation depending on the migration regime considered. More

The following section assesses our main results in terms of the growth in fenced areas over time relative to 1) types of protection, 2) administrative boundaries, and 3) other fences.Fencing relative to land governanceAcross the Greater Mara, a general growth in fenced areas can be observed throughout the 00 s but in particular over the last decade (Fig. 1). Based on satellite images, 35,067 ha were fenced in 1985, corresponding to c. 5%. In the following 25 years there was only an insignificant increase in fenced plots. However, from 2010, the number of fences suddenly grew rapidly, and in the following period (2015–2020) the fenced area increased even more radically, in an exponential manner (Fig. 2). For example, in 2015 there was 63,112 ha of fenced land; in 2016 this number rose to c. 75,176 ha, corresponding to a c. 20% annual increase. From 2010 to 2020, the ha fenced area increased by 170%. This corresponds to a roughly four times increase in the area enclosed by fences during the study period (1985–2020).Figure 2Conservative estimate of the fenced area of the entire Greater Mara, Kenya (1985–2020) expressed in hectares.Full size imageIn almost all regions, the number of fences continued to increase in 2019–20 (Fig. 2). The result is a total of 130,277 ha of fenced land in 2020, corresponding to 19% of the Greater Mara.Hence, there appears to be a building momentum in the expansion of fences in the Greater Mara: those regions that had many fences in 2016 ( > 1,000 ha) continue to experience an increase in the area enclosed by fences, with fences spreading almost everywhere in 2020 in particular. Those regions with the fewest fences in 2016 ( More

Ceballos, G. & Ehrlich, P. R. Mammal population losses and the extinction crisis. Science 296, 904–907 (2002).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Barnosky, A. D. et al. Has the Earth’s sixth mass extinction already arrived?. Nature 471, 51–57 (2011).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Ripple, W. J. et al. Status and ecological effects of the world’s largest carnivores. Science 343, 1241484 (2014).PubMed 
Article 
CAS 

Google Scholar 
Carbone, C. & Gittleman, J. L. A common rule for the scaling of carnivore density. Science 295, 2273–2276 (2014).ADS 
Article 

Google Scholar 
Durant, S. M. et al. The global decline of cheetah Acinonyx jubatus and what it means for conservation. Proc. Natl. Acad. Sci. 114, 528–533 (2017).CAS 
PubMed 
Article 

Google Scholar 
Jowkar, H. et al. Acinonyx jubatus ssp. venaticus. The IUCN Red List of Threatened Species 2008: e.T220A13035342. (2008).Khalatbari, L., Yusefi, G. H., Martínez-Freiría, F., Jowkar, H. & Brito, J. C. Availability of prey and natural habitats are related with temporal dynamics in range and habitat suitability for Asiatic Cheetah. Hystrix 29, 145–151 (2018).
Google Scholar 
Asadi, H. The Environmental Limitations and Future of the Asiatic Cheetah in Iran. (1997).CACP. Annual Report. (2014).Khalatbari, L., Jowkar, H., Yusefi, G. H., Brito, J. C. & Ostrowski, S. The current status of Asiatic cheetah in Iran. Cat News 66, 10–13 (2017).
Google Scholar 
Marker, L. L. et al. Ecology of free-ranging cheetahs. in Cheetahs: Biology and Conservation (eds. Marker, L. L., Boast, L. K. & Schmidt-Kuntzel, A.) 107–119 (Elsevier, 2017). doi:https://doi.org/10.1016/B978-0-12-804088-1.00008-3Hayward, M. W., Hofmeyr, M., O’Brian, J. & Kerley, G. I. H. Prey preferences of the cheetah (Acinonyx jubatus) (Felidae: Carnivora): morphological limitations or the need to capture rapidly consumable prey before kleptoparasites arrive?. J. Zool. 270, 615–627 (2006).Article 

Google Scholar 
Mills, M. G. L., Broomhall, L. S. & Toit, J. T. Cheetah Acinonyx jubatus feeding ecology in the Kruger National Park and a comparison across African savanna habitats: is the cheetah only a successful hunter on open grassland plains?. Wildlife Biol. 10, 177–186 (2004).Article 

Google Scholar 
Wachter, B., Jauernig, O. & Breitenmoser, U. Determination of prey hair in faeces of free-ranging Namibian cheetahs with a simple method. Cat News 44, 8–9 (2006).
Google Scholar 
Marker, L. L., Muntifering, J. R., Dickman, A. J., Mills, M. G. L. & Macdonald, D. W. Quantifying prey preferences of free-ranging Namibian cheetahs. South Afr. J. Wildl. Res. 33, 43–53 (2003).
Google Scholar 
Wacher, T. et al. Sahelo-Saharan Interest Group Wildlife Surveys, Part 4: Ahaggar Mountains, Algeria (March 2005). (2005).Thuo, D. et al. An insight into the prey spectra and livestock predation by cheetahs in Kenya using faecal DNA metabarcoding. Zoology 143, 125853 (2020).PubMed 
Article 

Google Scholar 
Broekhuis, F., Thuo, D. & Hayward, M. W. Feeding ecology of cheetahs in the Maasai Mara, Kenya and the potential for intra- and interspecific competition. J. Zool. 304, 65–72 (2018).Article 

Google Scholar 
Cooper, A. B., Pettorelli, N. & Durant, S. M. Large carnivore menus: factors affecting hunting decisions by cheetahs in the Serengeti. Anim. Behav. 73, 651–659 (2007).Article 

Google Scholar 
Mills, M. G. L. Living near the edge: A review of the ecological relationships between large carnivores in the arid Kalahari. African J. Wildl. Res. 45, 127–137 (2015).Article 

Google Scholar 
Rostro-García, S., Kamler, J. F. & Hunter, L. T. B. To kill, stay or flee: The effects of lions and landscape factors on habitat and kill site selection of cheetahs in South Africa. PLoS ONE 10, e0117743 (2015).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Laurenson, M. K. Behavioural costs and constraints of lactation in free-living cheetahs. Anim. Behav. 50, 815–826 (1995).Article 

Google Scholar 
Farhadinia, M. S. & Hemami, M.-R. Prey selection by the critically endangered Asiatic cheetah in central Iran. J. Nat. Hist. 44, 1239–1249 (2010).Article 

Google Scholar 
Farhadinia, M. S. et al. Feeding ecology of the Asiatic cheetah Acinonyx jubatus venaticus in low prey habitats in northeastern Iran: Implications for effective conservation. J. Arid Environ. 87, 206–211 (2012).ADS 
Article 

Google Scholar 
Zahedian, B. & Nezami, B. Cheetah (Acinonyx jubatus venaticus) (Felidae: Carnivora) feeding ecology in Central Plateau of Iran and effects of prey poor management. J. Wildl. Biodivers. 3, 22–30 (2019).
Google Scholar 
Zamani, N. et al. Predation of montane deserts ungulates by Asiatic cheetah Acinonyx jubatus venaticus in Central Iran. Folia Zool. 66, 50–57 (2017).Article 

Google Scholar 
Monterroso, P. et al. Factors affecting the (in)accuracy of mammalian mesocarnivore scat identification in South-western Europe. J. Zool. 289, 243–250 (2013).Article 

Google Scholar 
Morin, D. J. et al. Bias in carnivore diet analysis resulting from misclassification of predator scats based on field identification. Wildl. Soc. Bull. 40, 669–677 (2016).Article 

Google Scholar 
Caro, T. M. Cheetahs of the Serengeti Plains: Group Living in an Asocial Species (University of Chicago Press, 1994).
Google Scholar 
Floyd, T. J., Mech, L. D. & Jordan, P. A. Relating wolf scat content to prey consumed. J. Wildl. Manage. 42, 528–532 (1978).Article 

Google Scholar 
Jethva, B. D. & Jhala, Y. V. Computing biomass consumption from prey occurrences in Indian wolf scats. Zoo Biol. 23, 513–520 (2004).Article 

Google Scholar 
Pompanon, F. et al. Who is eating what: diet assessment using next generation sequencing. Mol. Ecol. 21, 1931–1950 (2012).CAS 
PubMed 
Article 

Google Scholar 
Taberlet, P., Coissac, E., Pompanon, F., Brochmann, C. & Willerslev, E. Towards next-generation biodiversity assessment using DNA metabarcoding. Mol. Ecol. 21, 2045–2050 (2012).CAS 
PubMed 
Article 

Google Scholar 
Mata, V. A. et al. How much is enough? Effects of technical and biological replication on metabarcoding dietary analysis. Mol. Ecol. 28, 165–175 (2019).CAS 
PubMed 
Article 

Google Scholar 
Shehzad, W. et al. Prey preference of Snow Leopard (Panthera uncia) in South Gobi Mongolia. PLoS ONE 7, e32104 (2012).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Monterroso, P. et al. Feeding ecological knowledge: the underutilised power of faecal DNA approaches for carnivore diet analysis. Mamm. Rev. 49, 97–112 (2019).Article 

Google Scholar 
Shehzad, W. et al. Carnivore diet analysis based on next-generation sequencing: Application to the leopard cat (Prionailurus bengalensis) in Pakistan. Mol. Ecol. 21, 1951–1965 (2012).CAS 
PubMed 
Article 

Google Scholar 
Thuo, D. et al. Food from faeces: Evaluating the efficacy of scat DNA metabarcoding in dietary analyses. PLoS ONE 14, e0225805 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Araujo, M. S., Bolnick, D. I. & Layman, C. A. The ecological causes of individual specialisation. Ecol. Lett. 14, 948–958 (2011).PubMed 
Article 

Google Scholar 
Balme, G. A., Roex, N., Rogan, M. S. & Hunter, L. T. B. Ecological opportunity drives individual dietary specialization in leopards. J. Anim. Ecol. 89, 589–600 (2020).PubMed 
Article 

Google Scholar 
Bolnick, D. I. et al. The ecology of individuals: incidence and implications of individual specialization. Am. Nat. 161, 1–28 (2003).MathSciNet 
PubMed 
Article 

Google Scholar 
Harrington, L. A., Harrington, A. L., Hughes, J., Stirling, D. & Macdonald, D. W. The accuracy of scat identification in distribution surveys: American mink, Neovison vison, in the northern highlands of Scotland. Eur. J. Wildl. Res. 56, 377–384 (2010).Article 

Google Scholar 
Weiskopf, S. R., Kachel, S. M. & McCarthy, K. P. What are snow leopards really eating? Identifying bias in food-habit studies. Wildl. Soc. Bull. 40, 233–240 (2016).Article 

Google Scholar 
Durant, S. M., Caro, T. M., Collins, D. A., Alawi, R. M. & Fitzgibbon, C. D. Migration patterns of Thomson’s gazelles and cheetahs on the Serengeti Plains. Afr. J. Ecol. 26, 257–268 (1988).Article 

Google Scholar 
Lindsey, P. A. et al. Minimum prey and area requirements of the vulnerable cheetah Acinonyx jubatus: implications for reintroduction and management of the species in South Africa. Oryx 45, 587–599 (2011).Article 

Google Scholar 
Farhadinia, M. S., Akbari, H., Eslami, M. & Adibi, M. A. A review of ecology and conservation status of Asiatic cheetah in Iran. Cat News Spec. Issue 18–26 (2016).Asadi, H. Some Observation on Hunting Behaviours of the Iranian Cheetah in Captivity. (1997).Heptner, V. G. & Sludskii, A. A. Mammals ofthe Soviet Union volume II part 2 Carnivora (hyaenas and cats). (Vysshaya Shkola Publishers, 1974).Ziaie, H. A Field Guide to the Mammals of Iran. (Iran Wildlife Center, 2008).Wilson, J. W. et al. Cheetahs, Acinonyx jubatus, balance turn capacity with pace when chasing prey. Biol. Lett. 9, 20130620 (2013).PubMed 
PubMed Central 
Article 

Google Scholar 
Grohé, C., Lee, B. & Flynn, J. J. Recent inner ear specialization for high-speed hunting in cheetahs. Sci. Rep. 8, 2301 (2018).ADS 
PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Cheraghi, F. et al. Inter-dependent movements of Asiatic Cheetahs Acinonyx jubatus venaticus and a Persian Leopard Panthera pardus saxicolor in a desert environment in Iran (Mammalia: Felidae). Zool. Middle East 65, 283–292 (2019).Article 

Google Scholar 
Ghoddousi, A., Soofi, M., Hamidi, A. K. & Lumetsberger, T. Assessing the role of livestock in big cat prey choice using spatiotemporal availability patterns. PLoS ONE 11, e0153439 (2016).PubMed 
PubMed Central 
Article 
CAS 

Google Scholar 
Khorozyan, I., Ghoddousi, A., Soofi, M. & Waltert, M. Big cats kill more livestock when wild prey reaches a minimum threshold. Biol. Conserv. 192, 268–275 (2015).Article 

Google Scholar 
Zeder, M. A. Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion, and impact. Proc. Natl. Acad. Sci. 105, 11597–11604 (2008).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Daberger, M. Systematic prioritization of livestock grazing rights buyout in the last viable population of Asiatic cheetah (Acinonyx jubatus venaticus) in Iran. (Humboldt University Berlin, 2021).Wolf, C. & Ripple, W. J. Prey depletion as a threat to the world’s large carnivores. R. Soc. Open Sci. 3, 160252 (2016).ADS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Melzheimer, J. et al. Communication hubs of an asocial cat are the source of a human—carnivore conflict and key to its solution. Proc. Natl. Acad. Sci. 117, 33325–33333 (2020).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Malakoutikhah, S., Fakheran, S., Tarkesh, M. & Senn, J. Assessing future distribution, suitability of corridors and efficiency of protected areas to conserve vulnerable ungulates under climate change. Divers. Distrib. 26, 1383–1396 (2020).Article 

Google Scholar 
Long, R. A., Donovan, T. M., Mackay, P., Zielinski, W. J. & Buzas, J. S. Comparing scat detection dogs, cameras, and hair snares for surveying carnivores. J. Wildl. Manage. 71, 2018–2025 (2007).Article 

Google Scholar 
Becker, M. S. et al. Using dogs to find cats: Detection dogs as a survey method for wide-ranging cheetah. J. Zool. 302, 184–192 (2017).Article 

Google Scholar 
Johnson, W. E. & O’Brien, S. J. Phylogenetic reconstruction of the Felidae using 16S rRNA and NADH-5 mitochondrial genes. J. Mol. Evol. 44, S98–S116 (1997).ADS 
CAS 
PubMed 
Article 

Google Scholar 
Reese, E. M., Winters, M., Booth, R. K. & Wasser, S. K. Development of a mitochondrial DNA marker that distinguishes domestic dogs from Washington state gray wolves. Conserv. Genet. Resour. 12, 497–501 (2020).Article 

Google Scholar 
Ormerod, S. J. Applied issues with predators and predation: Editor’s introduction. J. Appl. Ecol. 39, 181–188 (2002).Article 

Google Scholar 
Boast, L. K., Good, K. & Klein, R. Translocation of problem predators: Is it an effective way to mitigate conflict between farmers and cheetahs Acinonyx jubatus in Botswana?. Oryx 50, 537–544 (2016).Article 

Google Scholar 
Darvish Sefat, A. A. Atlas of Protected Areas of Iran (University of Tehran, 2006).
Google Scholar 
Yusefi, G. H., Faizolahi, K., Darvish, J., Safi, K. & Brito, J. C. The species diversity, distribution, and conservation status of the terrestrial mammals of Iran. J. Mammal. 100, 55–71 (2019).Article 

Google Scholar 
Karami, M., Ghadirian, T. & Faizolahi, K. The Atlas of the Mammals of Iran. (Iran Department of the Environment, 2016).Abangah Consulting Engineer Company. Reconvene expanded Livestock Control Committee (LCC) in Touran and establish the LCC for Miandasht with participation of all stakeholders. (2017).Mills, M. G. L. & Hofer, H. Hyaenas. Status Survey and Conservation Action Plan. (IUCN/SSC Hyaena Specualist Group, 1998).Maudet, C., Luikart, G., Dubray, D., Von Hardenberg, A. & Taberlet, P. Low genotyping error rates in wild ungulate faeces sampled in winter. Mol. Ecol. Notes 4, 772–775 (2004).CAS 
Article 

Google Scholar 
Deagle, B. E., Kirkwood, R. & Jarman, S. N. Analysis of Australian fur seal diet by pyrosequencing prey DNA in faeces. Mol. Ecol. 18, 2022–2038 (2009).CAS 
PubMed 
Article 

Google Scholar 
Frantz, A. C. et al. Reliable microsatellite genotyping of the Eurasian badger (Meles meles) using faecal DNA. Mol. Ecol. 12, 1649–1661 (2003).CAS 
PubMed 
Article 

Google Scholar 
Boom, R. et al. Rapid and simple method for purification of nucleic acids. J. Clin. Microbiol. 28, 495–503 (1990).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Rosel, P. E. & Kocher, T. D. DNA-based identification of larval cod in stomach contents of predatory fishes. J. Exp. Mar. Bio. Ecol. 267, 75–88 (2002).Article 

Google Scholar 
Deagle, B. E. et al. Molecular scatology as a tool to study diet: analysis of prey DNA in scats from captive Steller sea lions. Mol. Ecol. 14, 1831–1842 (2005).CAS 
PubMed 
Article 

Google Scholar 
Riaz, T. et al. ecoPrimers: inference of new DNA barcode markers from whole genome sequence analysis. Nucleic Acids Res. 39, e145 (2011).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Luikart, G. et al. Multiple maternal origins and weak phylogeographic structure in domestic goats. Proc. Natl. Acad. Sci. 98, 5927–5932 (2001).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Menotti-Raymond, M. et al. A genetic linkage map of microsatellites in the domestic cat (Felis catus). Genomics 57, 9–23 (1999).CAS 
PubMed 
Article 

Google Scholar 
Charruau, P. et al. Phylogeography, genetic structure and population divergence time of cheetahs in Africa and Asia: Evidence for long-term geographic isolates. Mol. Ecol. 20, 706–724 (2011).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Driscoll, C. A., Menotti-Raymond, M., Nelson, G., Goldstein, D. & O’Brien, S. J. Genomic microsatellites as evolutionary chronometers: A test in wild cats. Genome Res. 12, 414–423 (2002).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Kotze, A., Ehlers, K., Cilliers, D. C. & Grobler, J. P. The power of resolution of microsatellite markers and assignment tests to determine the geographic origin of cheetah (Acinonyx jubatus) in Southern Africa. Mamm. Biol. 73, 457–462 (2008).Article 

Google Scholar 
Marker, L. L. et al. Molecular genetic insights on cheetah (Acinonyx jubatus) ecology and conservation in Namibia. J. Hered. 99, 2–13 (2008).CAS 
PubMed 
Article 

Google Scholar 
Taberlet, P. et al. Reliable genotyping of samples with very low DNA quantities using PCR. Nucleic Acids Res. 24, 3189–3194 (1996).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Egeter, B. et al. Challenges for assessing vertebrate diversity in turbid Saharan water-bodies using environmental DNA. Genome 61, 807–814 (2018).CAS 
PubMed 
Article 

Google Scholar 
Magoc, T. & Salzberg, S. L. FLASH: Fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957–2963 (2011).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Rognes, T., Flouri, T., Nichols, B., Quince, C. & Mahé, F. VSEARCH: A versatile open source tool for metagenomics. PeerJ 4, e2584 (2016).PubMed 
PubMed Central 
Article 

Google Scholar 
Godinho, R. et al. Real-time assessment of hybridization between wolves and dogs: Combining noninvasive samples with ancestry informative markers. Mol. Ecol. Resour. 15, 317–328 (2015).CAS 
PubMed 
Article 

Google Scholar 
Valière, N. GIMLET: A computer program for analysing individual identification data. Mol. Ecol. 2, 377–379 (2002).
Google Scholar 
Wachter, B. et al. An advanced method to assess the diet of free-ranging large carnivores based on scats. PLoS ONE 7, e38066 (2012).ADS 
CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Breuer, T. Diet choice of large carnivores in northern Cameroon. Afr. J. Ecol. 43, 181–190 (2005).Article 

Google Scholar 
Wilson, M. F. J., O’Connell, B., Brown, C., Guinan, J. C. & Grehan, A. J. Multiscale terrain analysis of multibeam bathymetry data for habitat mapping on the continental slope. Mar. Geodesy 30, 2 (2007).Article 

Google Scholar 
Fick, S. E. & Hijmans, R. J. WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).Article 

Google Scholar  More

Overview of the study areaBased on the actual cultivation of peanuts, the Huang-Huai-Hai region is selected as the study area (Fig. 1). The main body of the study area is the Huang-Huai-Hai Plain (North China Plain), which is a typical alluvial plain resulting from extensive sediment deposition carried by the Yellow River, the Huaihe River and the Haihe River and their tributaries, and the hills in central and southern Shandong Peninsula adjacent to it. Administrative zones include 5 provinces, 2 cities, 53 cities and 376 counties (districts). In China, The Huang-Huai-Hai region is an important production and processing centre for agricultural products, with a total land area of 4.10 × 105 square kilometers and cultivated fields of 2.15 × 107 hm2, accounting for 4.3% and 16.3% of the total amount of the country, respectively. It belongs to temperate continental monsoon climate with distinct seasons, accumulated temperature of 3600–4800 degrees above 10 °C, frost-free period of 170–200 days and annual precipitation of 500–950 mm27. The Huang-huai-hai region is the largest peanut growing area, accounting for more than 50% of the country’s peanut production and area28.Figure 1Location of the study areas. The figure was made in the ArcGIS 10.2 platform (https://www.esri.com/en-us/home).Full size imageData sourcesThe data used in the study mainly include meteorological data, geographic information data and crop data. The meteorological data comes from China Meteorological Information Center (http://data.cma.cn), including the daily maximum temperature (℃), daily minimum temperature (℃), daily average temperature (℃), daily precipitation (mm) and daily average wind speed (M/s) observed by 186 ground observation meteorological stations in the Huang-Huai-Hai region from 1960 to 2019 (Fig. 1). Geographic information data include elevation DEM data (resolution of 1 km × 1 km) and land use data in the study area, which are from the resource and environmental science and data center of Chinese Academy of Sciences (http://www.resdc.cn). Crop data, including peanut sowing area and yield data, are derived from the statistical yearbooks of provinces and cities in the study area and China Agricultural Technology Network (http://www.cast.net.cn).Data processingMeteorological data processingAnusplin software is a tool to interpolate multivariate data based on ordinary thin disks and local thin disk spline functions, enabling the introduction of covariates for simultaneous spatial interpolation of multiple surfaces, suitable for meteorological data time series29. First, the Anusplin software is used to spatially interpolate the meteorological data and suitability data of the peanut growing season (April to September) from 1960 to 2019 based on the elevation data with a resolution of 1 km × 1 km. The Inverse Distance Weight (IDW) interpolation can make the meteorological data after Anusplin interpolation maintain consistency with the original data, and is able to improve the interpolation accuracy. Finally, the meteorological and suitability data set with a resolution of 1 km × 1 km is obtained. ArcGIS and MATLAB software were used to count the median of regional meteorological factors in agricultural fields of different cities (counties), and the meteorological factors and suitability of different periods of peanut growth season in each city (county) were obtained.Yield data processingMany factors affect crop yield formation, which can be generally divided into three main categories: meteorological conditions, agronomic and technological measures, and stochastic factors. Agricultural technical measures reflect the development level of social production in a certain historical period and become time technology trend output, which is referred to as trend output for short, and meteorological production reflects short period yield components that are affected by meteorological elements. Stochastic factors account for a small proportion and are often ignored in actual calculations30. The specific calculation is as follows:$$Y={Y}_{t}+{Y}_{w}$$
(1)

where Y is the actual yield (single production) of the crop, Yt is the trend yield, and Yw is the meteorological yield.In this paper, a straight-line sliding average method is used to simulate the trend yield. The straight-line sliding average method is a very commonly used method to model yield, and it considers the change in the time series of yield within a certain stage as a linear function, showing a straight line, as the stage continuously slides, the straight line continuously changes the position, and the backward slip reflects the continuous change in the evolution trend of the yield history31. The regression models in each stage are obtained in turn, and the mean value of each linear sliding regression simulation value at each time point is taken as its trend yield value. The linear trend equation at some stage is:$${Y}_{i}left(tright)={a}_{i}+{b}_{i}t$$
(2)
where i = n-k + 1, is the number of equations; k is the sliding step; n is the number of sample sequences; t is the time serial number. Yi(t) is the function value of each equation at point t. there are q function values at point t. the number of q is related to n and k. Calculate the average value of each function value at each point:$$overline{{Y }_{i}(t)}=frac{1}{q}sum_{j=1}^{q}{Y}_{i}left(tright)$$
(3)
Connecting the (overline{{Y }_{i}(t)}) value of each point can represent the historical evolution trend of production. Its characteristics depend on the value of k. Only when k is large enough, the trend yield can eliminate the influence of short cycle fluctuation. After comparison and considering the length of yield series, k is taken as 5 in this paper.After the trend yield is obtained, the meteorological yield is calculated using Eq. (1), then the relative meteorological production is$${Y}_{r}=frac{{Y}_{w}}{{Y}_{t}}$$
(4)
The relative meteorological yield shows that the relative variability of yield fluctuation deviating from the trend, that is, the amplitude of yield fluctuation, is not affected by time and space, and is comparable. However, when the value is negative, it indicates that the meteorological conditions are unfavorable to the overall crop production, and the crop yield reduction, that is, the yield reduction rate32.Characteristics of spatial and temporal distribution of climatic resources in the Huang-Huai-Hai regionCollect meteorological resource data from 1960 to 2019. Taking 1960–1989 as the first three decades of the study and 1990–2019 as the last three decades, the climatic resource changes of peanut growth in the Huang-Huai-Hai region are analyzed by interpolation of heat resources (average temperature), water resources (precipitation) and light resources (sunshine hours) in the study area in two periods combined with topographic factors.Establishment of suitability modelAccording to the definition of phenological time and growth period of peanut planting practice in the Huang-Huai-Hai region, the growth season of peanut is divided into three growth periods and five growth stages (Table 1). Temperature, precipitation and sunshine hours are the necessary meteorological factors to determine the normal development of peanut. Therefore, combined with climatic resources in the study area, temperature, precipitation and sunshine suitability model was introduced to quantitatively analyze the suitability of peanut planting.Table 1 Division of peanut growth periods.Full size tableTemperature suitability modelTemperature is a very important factor in the growth period of peanut, and the change of temperature in different growth periods will have a great influence on the yield and quality of peanut. As a warm-loving crop, accumulated temperature plays a decisive role in the budding condition and nutrient growth stage of peanut. Temperature determines the quality of fruit and the final yield of peanut. Beta function33 is used to calculate temperature suitability, which is universal for crop-temperature relationship. The specific calculation is as follows:$${F}_{i}left(tright)=frac{(t-{t}_{1}){({t}_{h}-t)}^{B}}{({t}_{0}-{t}_{1}){({t}_{h}-{t}_{0})}^{B}}$$
(5)
where the value of B is shown in$$B=frac{{t}_{h}-{t}_{0}}{{t}_{0}-{t}_{1}}$$
(6)
where Fi(t) is the temperature suitability of a certain growth period; t is the daily average temperature of peanut at a certain development stage; t1, th and t0 are the lower limit temperature, upper limit temperature and appropriate temperature required for each growth period of peanut. Refer to the corresponding index system and combined with the peanut production practice in Huang-Huai-Hai region34,35,36, determine the three base point temperature of peanut in each growth period, as shown in the Table 2.Table 2 Three fundamental points temperature and crop coefficient of peanut at each growth stage in the study area.Full size tablePrecipitation suitability modelPeanut has a long growth period, which is nearly half a year. Insufficient or excessive water during the growth period has a great impact on the growth and development, pod yield and quality of peanut. Combined with the actual situation of Huang-Huai-Hai region and peanut precipitation / water demand index, the water suitability function is determined and calculated as follows:$${text{F}}_{{text{i}}} left( {text{r}} right) = left{ {begin{array}{*{20}l} {frac{{text{r}}}{{0.9{text{ET}}_{{text{c}}} }}} hfill & {r < 0.9E{text{T}}_{{text{c}}} } hfill \ 1 hfill & {0.9E{text{T}}_{{text{c}}} le r le 1.2E{text{T}}_{{text{c}}} } hfill \ {frac{{1.2{text{ET}}_{{text{c}}} }}{{text{r}}}} hfill & {r > 1.2E{text{T}}_{{text{c}}} } hfill \ end{array} } right.$$
(7)
where Fi(r) is the water suitability of a certain growth period; r is the accumulated precipitation of peanut in a certain development period; ETc is the water demand of peanut in each growth period.$${mathrm{ET}}_{mathrm{c}}={mathrm{K}}_{mathrm{c}}cdot {mathrm{ET}}_{0}$$
(8)
where Kc is the peanut crop coefficient (Table 2) and ET0 is the crop reference evapotranspiration, which is calculated by the Penman Monteith method recommended by the international food and Agriculture Organization (FAO).Sunshine suitability modelSunshine hours are an important condition for photosynthesis. The “light compensation point” and “light saturation point” of peanut are relatively high, and more sunshine hours are required for photosynthesis. Under certain conditions of water, temperature and carbon dioxide, photosynthesis increases or decreases with the increase or decrease of light. Relevant studies show that when the sunshine hours reach more than 55% of the available sunshine hours, the crops reach the appropriate state to reflect the light37. The following formula is used to calculate the sunshine suitability of peanut in each growth period.$${mathrm{F}}_{mathrm{i}}left(mathrm{s}right)=left{begin{array}{l}frac{mathrm{S}}{{mathrm{S}}_{0}} quad S{mathrm{S}}_{0}end{array}right.$$
(9)
where Fi(s) is the sunshine suitability of peanut in a certain development period, S is the actual sunshine hours in a certain growth period, S0 is 55% of the sunshine hours (L0), and the calculation method of L0 refers to the following formula.$${mathrm{L}}_{0}=frac{2mathrm{t}}{15}$$
(10)
$$mathrm{sin}frac{mathrm{t}}{2}=sqrt{frac{mathrm{sin}(45^circ -frac{mathrm{varnothing }-updelta -upgamma }{2})times mathrm{sin}(45^circ +frac{mathrm{varnothing }-updelta -upgamma }{2})}{mathrm{cosvarnothing }times mathrm{cosdelta }}}$$
(11)
where Φ is the geographic latitude, δ is the declination, γ is the astronomical refraction, t is the angle.Comprehensive suitability modelPeanut has different needs for meteorological elements such as temperature, sunshine and precipitation in different growth periods. In order to analyze the impact of meteorological factors in different growth periods on yield, correlation analysis was conducted between the suitability of temperature, precipitation and sunshine in each growth period and the relative meteorological yield of peanut, and the correlation coefficient of each growth period divided by the sum of the correlation coefficients of the whole growth period was used as the weight coefficient of the suitability of temperature, precipitation and sunshine in each growth period (Table 3). The climatic suitability of each single element in peanut growing season is calculated by using formulas (12) and (13):Table 3 The weight coefficients of climatic suitability at each growth stage.Full size table$$left{begin{array}{c}{mathrm{b}}_{mathrm{ti}}=frac{{mathrm{a}}_{mathrm{ti}}}{sum_{mathrm{i}=1}^{mathrm{n}}{mathrm{a}}_{mathrm{ti}}}\ {mathrm{b}}_{mathrm{ri}}=frac{{mathrm{a}}_{mathrm{ri}}}{{sum }_{mathrm{i}=1}^{mathrm{n}}{mathrm{a}}_{mathrm{ri}}}\ {mathrm{b}}_{mathrm{si}}=frac{{mathrm{a}}_{mathrm{si}}}{{sum }_{mathrm{i}=1}^{mathrm{n}}{mathrm{a}}_{mathrm{si}}}end{array}right.$$
(12)
$$left{begin{array}{c}F(t)={sum }_{mathrm{i}=1}^{mathrm{n}}left[{mathrm{b}}_{mathrm{ti}}{mathrm{F}}_{mathrm{i}}(mathrm{t})right]\ F(r)={sum }_{mathrm{i}=1}^{mathrm{n}}left[{mathrm{b}}_{mathrm{ri}}{mathrm{F}}_{mathrm{i}}(mathrm{r})right]\ F(s)={sum }_{mathrm{i}=1}^{mathrm{n}}left[{mathrm{b}}_{mathrm{si}}{mathrm{F}}_{mathrm{i}}(mathrm{s})right]end{array}right.$$
(13)
where bti, bri and bsi are the weight coefficients of temperature, precipitation and sunshine suitability in the i growth period respectively, ati, ari and asi are the correlation coefficients between temperature, precipitation and sunshine suitability and meteorological impact index of peanut yield in the i growth period respectively, and F(t), F(r) and F(s) are the temperature, precipitation and sunshine suitability in peanut growth season respectively.Then, the geometric average method is used to obtain the comprehensive suitability of peanut growth season, as shown in formula (14).$$F(S)=sqrt[3]{F(t)times F(r)times F(s)}$$
(14)
Verification of climatic zoning resultsDrought and flood disaster indexOn the basis of previous studies, in view of the different water demand of peanut in different development stages, this paper adds the water demand of peanut in different development stages as an important index to calculate, and constructs a standardized precipitation crop water demand index (SPRI) that can comprehensively characterize the drought and flood situation of peanut, so as to judge and analyze the occurrence of drought and flood disasters of peanut.Step 1: calculate the difference D between precipitation and crop water demand at each development stage$${D}_{i}={P}_{i}-{ET}_{ci}$$
(15)
where Pi is the precipitation in the i development period (mm), and ETci is the crop water demand in the i development period (mm).Step 2: normalize the data sequence.Since there are negative values in the original sequence, it is necessary to normalize the data when calculating the standardized precipitation crop water demand index. The normalized value is the SPRI value. The normalization method and drought and flood classification are consistent with SPEI index38,39,40.Chilling injury indexBased on the results of previous studies41, the abnormal percentage of caloric index was selected as the index of low-temperature chilling injury of peanut to judge and analyze the occurrence of chilling injury in different growth stages. The specific calculation process and formula are as follows:Step 1: calculate the caloric index of different development stages.Combined with the growth and development characteristics of peanut and considering the appropriate temperature, lower limit temperature and upper limit temperature at different growth stages of peanut, the caloric index can reflect the response of crops to environmental heat conditions. The average value of daily heat index is taken as the heat index of growth stage to reflect the influence of heat conditions in different growth stages on crop growth and development. Refer to formulas (5) and (6) to calculate the heat index Fi(t) at different development stages.Step 2: calculate the percentage of heat index anomaly$${I}_{ci}=frac{{F}_{i}(t)-overline{{F }_{i}(t)}}{overline{{F }_{i}(t)}}times 100%$$
(16)
where Ici is the Chilling injury index of stage i, Fi(t) is the heat index of stage i, and (overline{{F }_{i}(t)}) is the average value of the heat index of stage i over the years.Heat injury indexBased on the results of previous studies42, taking the average temperature of 26 °C, 30 °C and 28 °C and the daily maximum temperature of 35 °C, 35 °C and 37 °C as the critical temperature index to identify the heat damage of peanut in three growth stages, if this condition is met and lasts for more than 3 days, it will be recorded as a high temperature event.Disaster frequencyDisaster frequency (Pi) is defined as the ratio of the number of years of disaster at a certain station to the total number of years in the study period43, which is calculated by formula (17).$${P}_{i}=frac{n}{N}times 100%$$
(17)
where n is the number of years of disaster events to some extent at a certain growth period at a certain station, and N is the total number of years. More

Testing H1 and H2 at community composition levelAs noted above, the simple fact that fungi grow more slowly than bacteria is the basis of the hypotheses that (H1) fungal communities should be more resistant than bacterial communities to drought stress, and (H2) that fungal communities should be less resilient than bacterial communities when the stress is relieved by rewetting18. In addition to growth rate, these two hypotheses may be related to differences in the form of growth between fungi and bacteria. For example, multicellular hyphal growth versus unicellular division or the greater thickness of fungal cell walls as compared to those of bacteria47,48. We tested H1 and H2 at the community composition level by blending the fungal and bacterial datasets generated from the same leaf, root, rhizosphere and soil samples collected from field-grown sorghum that had been either irrigated as a control, or subjected to preflowering drought followed by regular wetting beginning at flowering10,11.We followed the approach of Shade et al.17 to detect resistance and resilience, which had been developed for univariate variables, e.g., richness. For multivariate data, e.g., community composition, we modified it by calculating pairwise community dissimilarity for two groups: within-group (control-control pairs, drought-drought pairs, or rewetting-rewetting pairs), and between-group (control-drought pairs, or control-rewetting pairs). Ecological resistance to drought stress is detected by comparing compositional dissimilarity of between-group pairs (control-drought pairs) against within-group pairs (control-control pairs and drought-drought pairs) for each of the droughted weeks (weeks 3–8). Ecological resilience to rewetting is detected by assessing, from before to after rewetting, the change in the difference of compositional dissimilarity between within-group pairs and between-group pairs. Here, the point just before rewetting was week 8 and the points after rewetting were weeks 9–17. A t-test was used to assess the statistical significance of the differences in resistance or resilience between bacterial and fungal communities at each time point for each compartment.To account for the different resolutions of ITS and 16 S, we compared bacterial 16 S OTUs against both fungal ITS, species-level OTUs as well the fungal family level (Supplementary Fig. 1). The results of analyses using either fungal families or OTUs are consistent. Out of 36 comparisons (15 roots, 15 rhizospheres and 6 soils), different family and OTUs results were detected in four instances. In two of these, significances detected by OTUs were not detected by family (root, weeks 4 and 17) and, in the other two cases, significances detected by family were not detected by OTUs (rhizosphere, weeks 7 and 8). (Fig. 1). We report only results that are consistent at both the species and family levels (Fig. 1).In line with our first hypothesis, H1, we found that the resistance to drought stress for fungal mycobiomes was consistently stronger than that for bacterial microbiomes for weeks 5 in root, weeks 4–6 in rhizosphere, and weeks 4 and 6–8 in soil (Fig. 1, Supplementary Fig. 1 and Supplementary Table 2). In support of our second hypothesis, H2, when the stress of pre-flowering drought was relieved by rewetting, we found that the resilience of the bacterial communities was consistently higher than that for the fungi in weeks 9–16 in root, and weeks 11–17 in rhizosphere (Fig. 1, Supplementary Fig. 1 and Supplementary Table 2).Surprisingly, we found that resilience was stronger for fungal than bacterial communities in the first week (week 9) of rewetting in the rhizosphere (Fig. 1, Supplementary Fig. 1 and Supplementary Table 2). This high resilience of fungi may be associated with the quick growth of sorghum roots when rewetted. The rhizosphere zone around these newly formed roots may be quickly colonized by soil fungi, a community that was weakly affected by drought. This result suggests that re-assembly of the rhizosphere microbial community is more complex than previously expected.The finding that fungal community composition in the soil is not shaped by drought prevented us from further detecting resilience (Fig. 1). Note fungal community in early leaves was excluded from analysis due to the high proportion of non-fungal reads in sequencing11.Testing H1 and H2 at all-correlation levelNext, we moved from the comparison of whole communities to correlation among individual bacterial and fungal taxa to test the hypotheses about resistance, H1, and resilience, H2. As noted above, previous research provided the foundation for the stress gradient hypothesis, which predicts an increase in positive associations in stress32,33,34,35,36,37. Further, ecological modeling predicts that negative associations promote stability40. Concerning specific associations, studies of Arabidopsis and associated microbes reported that positive associations are favored within kingdoms, i.e., within bacteria or within fungi, while negative associations predominate between kingdoms38,39. Given these foundations, concerning H1, we expected an increase in the proportion of positive correlation by drought stress that would be strongest for B-B, followed by F-F, and lastly by B-F; for H2 we expected rewetting to cause a decrease in the proportion of positive correlation, again most strongly for B-B, followed by F-F, and lastly by B-F.Overall, at the all-correlation level, we found no consistent support for the differences postulated for bacterial and fungal responses in H1. For example, strong increases in the proportion of positive correlations under drought could be found in all microbial pairings for some compartments (B-B in leaf and root, F-F in rhizosphere and soil, and B-F in root and rhizosphere) (Fig. 2a, Supplementary Figs. 2, 3). Neither did we find consistent support for the differences ascribed to bacteria and fungi in H2 as the strongest decreases in the proportion of positive correlations during rewetting occurred at F-F in rhizosphere and soil, and B-B in leaf and root (Fig. 2b, Supplementary Figs. 2, 3).Fig. 2: Correlations of microbes in drought stress and drought relief.Estimates of combined correlations (row a) show an increase in positive correlations under drought stress across the four compartments (root, black; rhizosphere, blue; soil, red; leaf, green). Data points underlying the lines in the figure are provided in the alternative version in Supplementary Fig. 2. This result is in line with the stress gradient hypothesis which posits that stressful environments favor positive associations because competition will be less intense than in benign environments32,33,36,37. Note that positive trends in combined correlations can arise in two ways. First, from an increase of positive correlations (row b) that exceeds the rise in negative correlations (row c), e.g., Leaf bacterial-bacterial (Bac-Bac) correlations or rhizosphere fungal-fungal (Fun-Fun) correlations in the drought period (Negative correlations in row C values are multiplied by −1 to facilitate comparison). Second, from a decrease in negative correlations that exceeds a decrease in positive correlations, e.g., root bacterial-bacterial correlations or root bacterial-fungal (Bac-Fun) correlations in drought. Combined (a), positive (b) and negative (c) estimates of correlation (Spearman’s rho, ρ) are given for four compartments (root, rhizosphere, soil and leaf), and three types of correlations (Bacterium-Bacterium, Fungus-Fungus, Bacterium-Fungus). T-tests (two sided) were carried out for linear mixed effect modelling that incorporates link type and compartments as random factors. Detailed distribution densities of correlations are presented in Supplementary Fig. 3. Source data are provided as a Source Data file.Full size imageWe found support for the stress gradient hypothesis because drought increased the relative frequency of positive correlations among microbial taxa (Fig. 2a, Supplementary Figs. 2, 3). The increases were due, largely, to B-B correlations in leaf and F-F correlations in the rhizosphere during drought, when the relative frequency of positive correlations was increased (Fig. 2b, Supplementary Figs. 2, 3) and the frequencies of negative correlations were decreased or weakly increased (Fig. 2c, Supplementary Figs. 2, 3). Less obvious increases in the relative frequency of positive correlations (such as B-B in root, F-F in soil, and B-F in root and rhizosphere) occurred where drought reduced both positive and negative correlations, but the losses of negative correlations exceeded those of positive correlations (Fig. 2, Supplementary Figs. 2, 3).In support of the expectation that correlations would be more negative between taxonomic groups than within taxonomic groups, we found that the relative frequency of positive correlations was generally lower for B-F than B-B and F-F correlations (Fig. 2, Supplementary Figs. 2, 3). Moreover, as ecological modeling has indicated that negative associations should promote stability of communities40, we hypothesize that B-F correlations would be more stable than B-B and F-F networks in response to drought stress. However, we found no support for this hypothesis, as B-F correlations (for example in root) did not always show the least response to drought stress (Fig. 2, Supplementary Figs. 2, 3).Testing H1 and H2 at species co-occurrence levelFor our final test of H1 (resistance) and H2 (resilience) we focused on co-occurrence networks based on significant, positive correlations. These networks have been reported to be destabilized for bacteria but not for fungi in mesocosms subject to drought stress19, and shown to be disrupted for bacteria in natural vegetation studied over gradients of increasing aridity41,42. Using these results as guides, for H1 we expected that drought stress should disrupt co-occurrence networks most strongly for B-B, followed by F-F, and lastly by B-F. For H2 we expected that relief of stress by rewetting should strengthen microbial co-occurrence networks most strongly for B-B, followed by F-F, and lastly by B-F.For this test we constructed microbial co-occurrence networks using significant positive pairwise correlations between microbial taxa, B-B, F-F and B-F, and compared the network complexity between fully irrigated control and drought, and between control and rewetting following drought. In general, we found no consistent support for the difference between bacteria and fungi inherent in H1. Rhizosphere was the one compartment where B-B vertices dropped and F-F vertices rose in response to drought, as expected, but this result was offset in root and soil, where vertices dropped in all networks, B-B, F-F and B-F (Figs. 3, 4; Supplementary Figs. 4, 5). Analysis by co-occurrence networks highlighted the differences between plant compartments. In root drought strongly disrupted networks of B-B, B-F and F-F, but in the other three compartments, network disruption was weaker, and networks were even enhanced by drought for F-F in rhizosphere and B-B in leaf (Figs. 3, 4).Fig. 3: Networks of significant positive cross-taxonomic group correlations (bacteria and fungi).a Fungal operational taxonomic units (OTUs) (blue) and bacterial OTUs (black) are graphed as nodes. Significant positive Spearman correlations are graphed as edges (ρ  > 0.6, false discovery rate adjusted P  More

Population genetic data collection from primary data sourcesFigure 4 describes the overall data collection workflow for the four datasets that comprise CaliPopGen. We first identified literature potentially containing population genetic data for California by querying the Web of Science Core Collection (https://webofknowledge.com/) for relevant literature from 1900 to 2020 with the terms: topic = (California*) AND topic = (genetic* OR genomic*) AND topic = (species OR taxa* OR population*). We included only empirical peer-reviewed literature and excluded unreviewed preprints. In using these search terms, our goal was to broadly identify genetic papers focused on California with population or species-level analyses, while avoiding purely phylogenetic studies or those focused on agricultural or model species. This resulted in 4,942 unique records.Fig. 4Flow chart of the data collection process that generated the CaliPopGen databases.Full size imageWe next screened titles and abstracts to retain articles that: (1) provided data on populations of species which are self-sustaining without anthropogenic involvement; (2) included at least some eukaryote species; (3) included population(s) sampled within California; (4) mentioned measures of genetic diversity or differentiation; and (5) were not reviews (thus restricting our search to only primary literature). We retained 1869 studies after this first pass of literature screening (see Technical Validation for estimate of inter- and intra-screener bias).Our second, more in-depth screening pass involved reading the full text of these 1869 studies. We had two goals. First, we confirmed that retained papers fully met all five of our inclusion criteria (the first screen was very liberal with respect to these criteria, and many papers failed to meet at least one criterion after close reading). Second, we eliminated papers where the data were not presented in a way that allowed us to extract population-level information. For example, many of the more systematics-focused studies pooled samples from large, somewhat ill-defined regions (“Sierra Nevada” or “Southern California”); if such regions were larger than 50 km in a linear dimension, we deemed them unusable for making geographically-informative inferences. Other studies presented summaries of population data, often in the form of phylogenetic networks or trees, but did not include information on actual population genetic parameters and therefore were not relevant to our database. We retained 528 publications after this second pass.From this set of papers, we extracted species, locality, and genetic data for each California population or sampling locality described in each study (Fig. 3A). This included Latin binomial/trinomial, English common name, population identifiers, and geographic coordinates of sampling sites. We also noted population/sampling localities that were interpreted as comprised of interspecific hybrids, and listed both parental species. We collected population genetic diversity and differentiation statistics for each unique genetic marker for each population/sampling locality; as a result, a sampling locality may have multiple entry rows, one for each locus or marker type. Parameters extracted for each population/marker combination include sample size, genetic marker type, gene targets, number of loci, years of sampling, and reported values for effective population size (Ne), expected (HE) and observed (HO,) heterozygosity, nucleotide diversity (π, pi), alleles-per-locus (APL), allelic richness (AR), percent polymorphic loci (PPL), haplotype diversity (HDIV), inbreeding coefficient (e.g. FIS, FIT, GIS), and pairwise population genetic comparison parameters (FST, GST, DST, Nei’s D, Jost’s D, or phi). We note that while there are technical differences between allelic richness and alleles-per-locus, source literature often used the terms interchangeably, and we include the parameters and their values as named in the source. We define marker type as the general category of genetic marker used (e.g., “microsatellite” or “nuclear”), while gene targets are the specific locus/loci (e.g., “COI”). We present these data in two separate datasets, one containing all population-level genetic summary statistics (Dataset 121, see Fig. 3C and detailed description in Table 1) and a second for estimates of pairwise genetic differentiation (Dataset 221, see Fig. 3D and detailed description in Table 2).Table 1 Description of the population genetic data in Dataset 121.Full size tableTable 2 Description of the pairwise genetic distance data in Dataset 221.Full size tableAll genetic data were extracted directly from the source literature. However, we also updated or added to the metadata for these population genetic values in several ways. We included kingdom, phylum, and a lower-level taxonomic grouping for each species (usually class), and updated scientific and common names based on the currently accepted taxonomy of the Global Biodiversity Information Facility22. When geographic coordinates were not provided for a sampling locality, as was frequently the case in the older literature, we used Google Maps (https://www.google.com/maps) to georeference localities based on either in-text descriptions or embedded figure maps guided by permanent landmarks like a bend in a river or administrative boundaries. Because this can only yield approximate coordinates, we recorded estimated accuracy as the radius of our best estimate of possible error in kilometers. If coordinates were provided in degree/minute/seconds, we used Google Maps to translate them to decimal degrees. In cases where coordinates were not provided and locality descriptions were too vague to determine coordinates with less than 50 km estimated coordinate error, we did not attempt to extract coordinates but still provide the genetic data. All coordinates are provided in the web Mercator projection (EPSG:3857). We excluded studies that reported genetic parameter values only for samples aggregated regionally (“Southern California” or “Sierra Nevada”). If marker type was not explicitly included, we classified marker type based on the gene targets reported, if provided.Life history trait data collectionTo increase the utility of CaliPopGen, we also assembled data on life history traits for all animal (Dataset 321) and plant (Dataset 421) species contained in Datasets 121 and 221. We assembled trait data that have previously been shown to correlate with genetic diversity, including those related to reproduction, life cycle, and body size, as well as conservation status (e.g.23,24,25,26,). Life history data were compiled by first referencing large online repositories, often specific to taxonomic groups, like the TRY plant trait database27, and the Royal Botanic Gardens Kew Seed Information Database28. If trait data for species of interest were unavailable from these compilations, we conducted keyword literature searches for each combination of species and life history trait, and extracted data from the primary literature. When data were not available for the subspecies or species for which we had genetic data, we report values for the next closest taxonomic level, up to and including family, as available in the literature.For both animals and plants, we defined habitat types as marine, freshwater, diadromous, amphibious, or terrestrial. Marine species include those that are found in brackish or wetland-marine habitats, as well as bird species that primarily reside in marine habitats. Freshwater species include those that are found in wetland-freshwater habitats, as well as species that primarily reside in freshwater. The diadromous category includes fish species that are catadromous or anadromous. We considered species to be amphibious if they have an obligatory aquatic stage in their life cycle, but also spend a significant portion of their life cycle on land. Terrestrial species were defined as those that spend most of their life cycle on land and are not aquatic for any portion of their life cycle. In a few cases (e.g., waterbirds that are both freshwater and marine, semi-aquatic reptiles), a species could reasonably be placed in more than one category, and we did our best to identify the primary life history category for such taxa. If the taxonomic identity of an entry was hybrid between species or subspecies, this was noted in the speciesID column and no life history data were reported.The CaliPopGen Animal Life History Traits Dataset 321 (description of dataset in Table 3) includes habitat type, lifespan, fecundity, lifetime reproductive success, age at sexual maturity, number of breeding events per year, mode of reproduction, adult length and mass, California native status, listing status under the US Endangered Species Act (ESA), listing status under the California Endangered Species Act (CESA), and status as a California Species of Special Concern (SSC). For some traits, value ranges were recorded–for example, minimum to maximum lifespan. In other cases, we recorded single values and, when available, a definition of this single value, (for example, minimum, average, or maximum lifespan). We report either the range of the age of sexual maturity (minimum to maximum), or a single value, depending on the available literature. For sexually dimorphic species, we report female adult length and weight when available, because female body size often correlates with fecundity. Across animal taxonomic groups, different measures of body size and length measurements are often used, reflecting community consensus on how to measure size. Given this variation, we report the type of length measurement, if available, as Standard Length (SL), Fork Length (FL), Total Length (TL), Snout-to-Vent Length (SVL), Straight-Line Carapace (SLC), or Wingspan (WS).Table 3 Description of the animal life-history data in Dataset 321.Full size tableThe CaliPopGen Plant Life History Traits Dataset 421 (description of dataset in Table 4) includes habitat type, lifespan, life cycle, adult height, self-compatibility, monoecious or dioecious, mode of reproduction, pollination and seed dispersal modes, mass per seed, California native status, NatureServe29 element ranks (global and state ranks, see Table 5 for definitions), listing status under the Federal Endangered Species Act (ESA), and listing status under the California Endangered Species Act (CESA). In contrast to most animal species, plant lifespan was typically reported as a single value. We define life cycles as the following: Annual: completes full life cycle in one year; Biennial: completes full life cycle in two years; Perennial: completes full life cycle in more than two years; Perennial-Evergreen: perennial and retains functional leaves throughout the year; Perennial-Deciduous: perennial and loses all leaves synchronously for part of the year. Some species are variable (for example, have annual and biennial individuals), and in those cases we attempted to characterize the most common modality.Table 4 Description of the plant life-history data in Dataset 421.Full size tableTable 5 Description of the Conservation status (Heritage Rank) from California Natural Diversity Database29.Full size tableBecause of the paucity of data available for chromists and fungi, we did not extract life history trait data for the relatively few species in these taxonomic groups.Data visualization and summaryWe used the R-package raster (v3.1–5) to visualize the spatial extent of the data in CaliPopGen in Fig. 3. Panel (A) shows a summary plot of all unique populations of both the Population Genetic Diversity in Dataset 121 and the Pairwise Population Differentiation in Dataset 221. Panel (B) shows the total number of unique populations in each California terrestrial ecoregion. Panel (C) depicts all data entries of Population Genetic Diversity Dataset 121, summed for each 20×20 km grid cell. Panel (D) shows the density of pairwise straight lines drawn between pairs of localities in the Pairwise Population Differentiation Dataset 221, depicted as the total number of lines per 20×20 km grid cell. The number of populations and species of both Datasets 121 & 221 are summarized for each marine and terrestrial ecoregion in Table 6.Table 6 Summary of total numbers of populations and species per California ecoregion, separately for population genetic and pairwise datasets.Full size table More