Philippa Kaur

Hui, C. & Richardson, D. M. Invasion Dynamics (Oxford University Press, 2017).Book 
MATH 

Google Scholar 
Clobert, J., Baguette, M., Benton, T. G. & Bullock, J. M. Dispersal Ecology and Evolution (Oxford University Press, 2012).Book 

Google Scholar 
Shigesada, N., Kawasaki, K. & Takeda, Y. Modeling stratified diffusion in biological invasions. Am. Nat. 146, 229–251 (1995).Article 

Google Scholar 
Chuang, A. & Peterson, C. R. Expanding population edges: Theories, traits, and trade-offs. Glob. Change Biol. 22, 494–512 (2016).Article 
ADS 

Google Scholar 
Cayuela, H. et al. Determinants and consequences of dispersal in vertebrates with complex life cycles: A review of pond-breeding amphibians. Q. Rev. Biol. 95, 36 (2020).Article 

Google Scholar 
Measey, G. J. et al. A global assessment of alien amphibian impacts in a formal framework. Divers. Distrib. 22, 970–981 (2016).Article 

Google Scholar 
Antonelli, A., Smith, R. J., Perrigo, A. L. & Crottini, A. Madagascar’s extraordinary biodiversity: Evolution, distribution, and use. Science 378, eabf0869 (2022).
Article 
CAS 
PubMed 

Google Scholar 
Marshall, B. M. et al. Widespread vulnerability of Malagasy predators to the toxins of an introduced toad. Curr. Biol. 28, R654–R655 (2018).Article 
CAS 
PubMed 

Google Scholar 
Licata, F. et al. Toad invasion of Malagasy forests triggers severe mortality of a predatory snake. Biol. Inv. 24, 1189–1198 (2022).Article 

Google Scholar 
Licata, F. et al. Abundance, distribution and spread of the invasive Asian toad Duttaphrynus melanostictus in eastern Madagascar. Biol. Inv. 21, 1615–1626 (2019).Article 

Google Scholar 
McClelland, P., Reardon, J. T., Kraus, F., Raxworthy, C. J. & Randrianantoandro, C. Asian toad eradication feasibility report for Madagascar (Te Anau, 2015).Smith, M. A. & Green, D. M. Dispersal and the metapopulation paradigm in amphibian ecology and conservation: Are all amphibian populations metapopulations?. Ecography 28, 110–128 (2005).Article 

Google Scholar 
Shine, R. et al. Increased rates of dispersal of free-ranging cane toads (Rhinella marina) during their global invasion. Sci. Rep. 11, 23574 (2021).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Myles-Gonzalez, E., Burness, G., Yavno, S., Rooke, A. & Fox, M. G. To boldly go where no goby has gone before: Boldness, dispersal tendency, and metabolism at the invasion front. Behav. Ecol. 26, 1083–1090 (2015).Article 

Google Scholar 
Van Petegem, K. H. P. et al. Empirically simulated spatial sorting points at fast epigenetic changes in dispersal behaviour. Evol. Ecol. 29, 299–310 (2015).Article 

Google Scholar 
Stuart, Y. E. et al. Rapid evolution of a native species following invasion by a congener. Science 346, 463–466 (2014).Article 
ADS 
CAS 
PubMed 

Google Scholar 
Licata, F., Andreone, F., Crottini, A., Harison, R. F. & Ficetola, G. F. Does spatial sorting occur in the invasive Asian toad in Madagascar? Insights into the invasion unveiled by morphological analyses. JZSER 2021, 1–9 (2021).
Google Scholar 
Schwarzkopf, L. & Alford, R. A. Nomadic movement in tropical toads. Oikos 96, 492–506 (2002).Article 

Google Scholar 
Brown, G. P., Kelehear, C. & Shine, R. Effects of seasonal aridity on the ecology and behaviour of invasive cane toads in the Australian wet–dry tropics. Funct. Ecol. 25, 1339–1347 (2011).Article 

Google Scholar 
Duellman, W. E. & Trueb, L. Biology of Amphibians (JHU Press, 1994).Book 

Google Scholar 
Wells, K. D. The Ecology and Behavior of Amphibians (University of Chicago Press, 2010). https://doi.org/10.7208/9780226893334.Book 

Google Scholar 
Shaw, A. K., Kokko, H. & Neubert, M. G. Sex difference and Allee effects shape the dynamics of sex-structured invasions. J. Anim. Ecol. 87, 36–46 (2018).Article 
PubMed 

Google Scholar 
Schwarzkopf, L. & Alford, R. A. Desiccation and shelter-site use in a tropical amphibian: Comparing toads with physical models. Funct. Ecol. 10, 193–200 (1996).Article 

Google Scholar 
Wogan, G. O. U., Stuart, B. L., Iskandar, D. T. & McGuire, J. A. Deep genetic structure and ecological divergence in a widespread human commensal toad. Biol. Lett. 12, 20150807 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Licata, F. Exploring the invasion dynamics and impacts of the invasive Asian common toad in Madagascar (University of Porto, 2022).
Google Scholar 
Reilly, S. B. et al. Toxic toad invasion of Wallacea: A biodiversity hotspot characterized by extraordinary endemism. Glob. Change Biol. 23, 5029–5031 (2017).Article 
ADS 

Google Scholar 
Jørgensen, C. B., Shakuntala, K. & Vijayakumar, S. Body size, reproduction and growth in a tropical toad, Bufo melanostictus, with a comparison of ovarian cycles in tropical and temperate zone anurans. Oikos 46, 379 (1986).Article 

Google Scholar 
Vences, M. et al. Tracing a toad invasion: Lack of mitochondrial DNA variation, haplotype origins, and potential distribution of introduced Duttaphrynus melanostictus in Madagascar. Amphib. Reptilia 38, 197–207 (2017).Article 

Google Scholar 
Ngo, B. V. & Ngo, C. D. Reproductive activity and advertisement calls of the Asian common toad Duttaphrynus melanostictus (Amphibia, Anura, Bufonidae) from Bach Ma National Park, Vietnam. Zool. Stud. 52, 12 (2013).Article 

Google Scholar 
Licata, F. et al. The Asian toad (Duttaphrynus melanostictus) in Madagascar: A report of an ongoing invasion. In Problematic Wildlife II: New Conservation and Management Challenges in the Human-Wildlife Interactions (eds Angelici, F. M. & Rossi, L.) 617–638 (Springer, 2020). https://doi.org/10.1007/978-3-030-42335-3_21.Chapter 

Google Scholar 
Moore, M., Solofo Niaina Fidy, J. F. & Edmonds, D. The new toad in town: Distribution of the Asian toad, Duttaphrynus melanostictus, in the Toamasina area of eastern Madagascar. Trop. Conserv. Sci. 8, 440–455 (2015).Article 

Google Scholar 
Licata, F. et al. Using public surveys to rapidly profile biological invasions in hard-to-monitor areas. Anim. Conserv. https://doi.org/10.1111/acv.12835 (2023).Article 

Google Scholar 
Zhang, M. et al. Automatic high-resolution land cover production in madagascar using sentinel-2 time series, tile-based image classification and google earth engine. Remote Sensing 12, 3663 (2020).Article 
ADS 

Google Scholar 
Peel, M. C., Finlayson, B. L. & Mcmahon, T. A. Updated world map of the Köppen-Geiger climate classification. Hydrol. Earth Syst. Sci. 4, 439–473 (2007).
Google Scholar 
Merkel, A. Toamasina Climate (Madagascar). Accessed 20 July 2022. https://en.climate-data.org/africa/madagascar/toamasina/toamasina-4029/
(2021).Gordon, A. Secondary sexual characters of Bufo melanostictus schneider. Copeia 1933, 204–207 (1933).Article 

Google Scholar 
Alford, R. & Rowley, J. Techniques for tracking amphibians: The effects of tag attachment, and harmonic direction finding versus radio telemetry. Amphib. Reptilia 28, 367–376 (2007).Article 

Google Scholar 
Lassueur, T., Joost, S. & Randin, C. F. Very high resolution digital elevation models: Do they improve models of plant species distribution?. Ecol. Modell. 198, 139–153 (2006).Article 

Google Scholar 
Abrams, M., Crippen, R. & Fujisada, H. ASTER global digital elevation model (GDEM) and ASTER global water body dataset (ASTWBD). Remote Sensing 12, 1156 (2020).Article 
ADS 

Google Scholar 
Brown, G. P., Phillips, B. L., Webb, J. K. & Shine, R. Toad on the road: Use of roads as dispersal corridors by cane toads (Bufo marinus) at an invasion front in tropical Australia. Biol. Conserv. 133, 88–94 (2006).Article 

Google Scholar 
Breiman, L. Random forests. Mach. Learn. 45, 5–32 (2001).Article 
MATH 

Google Scholar 
Hijmans, R. J. et al. raster: Geographic data analysis and modeling. https://CRAN.R-project.org/package=raster (2021).Yagi, K. T. & Green, D. M. Performance and movement in relation to postmetamorphic body size in a pond-breeding amphibian. J. Herpetol. 51, 482–489 (2017).Article 

Google Scholar 
Labocha, M. K., Schutz, H. & Hayes, J. P. Which body condition index is best?. Oikos 123, 111–119 (2014).Article 

Google Scholar 
Tingley, R. & Shine, R. Desiccation risk drives the spatial ecology of an invasive anuran (Rhinella marina) in the australian semi-desert. PLoS ONE 6, e25979 (2011).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Richards, S. J., Sinsch, U. & Alford, R. A. Radio Tracking. In Measuring and Monitoring Biological Diversity: Standard Methods for Amphibians (eds Heyer, R. et al.) 155–158 (Smithsonian Institution, 1994).
Google Scholar 
Altobelli, J. T., Dickinson, K. J. M., Godfrey, S. S. & Bishop, P. J. Methods in amphibian biotelemetry: Two decades in review. Austral. Ecol. 47, 1382–1395 (2022).Article 

Google Scholar 
Dormann, C. F. et al. Collinearity: A review of methods to deal with it and a simulation study evaluating their performance. Ecography 36, 27–46 (2013).Article 

Google Scholar 
Burnham, K. P. & Anderson, D. R. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach (Springer, 2002). https://doi.org/10.1007/978-1-4757-2917-7_3.Book 
MATH 

Google Scholar 
Richards, S. A., Whittingham, M. J. & Stephens, P. A. Model selection and model averaging in behavioural ecology: The utility of the IT-AIC framework. Behav. Ecol. Sociobiol. 65, 77–89 (2011).Article 

Google Scholar 
R Core Team. R: A language and environment for statistical computing. (2021).Bates, D. et al. lme4: Linear Mixed-Effects Models using ‘Eigen’ and S4. (2020).Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. lmerTest package: Tests in linear mixed effects models. J. Stat. Softw. 82, 1–26 (2017).Article 

Google Scholar 
Barton, K. MuMIn: Multi-Model Inference. (2022).Hodges, C. W., Marshall, B. M., Hill, J. G. & Strine, C. T. Malayan kraits (Bungarus candidus) show affinity to anthropogenic structures in a human dominated landscape. bioRxiv https://doi.org/10.1101/2021.09.08.459477 (2021).Article 

Google Scholar 
Muller, B. J., Cade, B. S. & Schwarzkopf, L. Effects of environmental variables on invasive amphibian activity: Using model selection on quantiles for counts. Ecosphere 9, e02067 (2018).Article 

Google Scholar 
Linsenmair, K. E. & Spieler, M. Migration patterns and diurnal use of shelter in a ranid frog of a West African savannah: A telemetric study. Amphib. Reptilia 19, 43–64 (1998).Article 

Google Scholar 
Clobert, J., Le Galliard, J.-F., Cote, J., Meylan, S. & Massot, M. Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations. Ecol. Lett. 12, 197–209 (2009).Article 
PubMed 

Google Scholar 
Ward-Fear, G., Greenlees, M. J. & Shine, R. Toads on lava: spatial ecology and habitat use of invasive cane yoads (Rhinella marina) in Hawai’i. PLoS ONE 11, e0151700 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Huang, W.-S., Lin, J.-Y. & Yu, J.Y.-L. Male reproductive cycle of the toad Bufo melanostictus in Taiwan. Zool. Sci. 14, 497–503 (1997).Article 

Google Scholar 
Brown, G. P., Phillips, B. L. & Shine, R. The straight and narrow path: the evolution of straight-line dispersal at a cane toad invasion front. Proc. R. Soc. B 281, 20141385 (2014).Article 
PubMed 
PubMed Central 

Google Scholar 
Perkins, T. A., Phillips, B. L., Baskett, M. L. & Hastings, A. Evolution of dispersal and life history interact to drive accelerating spread of an invasive species. Ecol. Lett. 16, 1079–1087 (2013).Article 
PubMed 

Google Scholar 
Ochocki, B. M. & Miller, T. E. X. Rapid evolution of dispersal ability makes biological invasions faster and more variable. Nat. Commun. 8, 14315 (2017).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Phillips, B. L., Brown, G. P., Travis, J. M. J. & Shine, R. Reid’s paradox revisited: The evolution of dispersal kernels during range expansion. Am. Nat. 172, S34–S48 (2008).Article 
PubMed 

Google Scholar 
Kot, M., Lewis, M. A. & van den Driessche, P. Dispersal data and the spread of invading organisms. Ecology 77, 2027–2042 (1996).Article 

Google Scholar 
Deguise, I. & Richardson, J. S. Movement behaviour of adult western toads in a fragmented, forest landscape. Can. J. Zool. 87, 1184–1194 (2009).Article 

Google Scholar 
Mitrovich, M. J., Gallegos, E. A., Lyren, L. M., Lovich, R. E. & Fisher, R. N. Habitat use and movement of the endangered Arroyo toad (Anaxyrus californicus) in coastal southern California. J. Herpetol. 45, 319–328 (2011).Article 

Google Scholar 
Urban, M. C., Phillips, B. L., Skelly, D. K. & Shine, R. A toad more traveled: The heterogeneous invasion dynamics of cane toads in Australia. Am. Nat. 171, E134–E148 (2008).Article 
PubMed 

Google Scholar 
Enriquez-Urzelai, U., Montori, A., Llorente, G. A. & Kaliontzopoulou, A. Locomotor mode and the evolution of the hindlimb in western mediterranean anurans. Evol. Biol. 42, 199–209 (2015).Article 

Google Scholar 
Junior, B. T. & Gomes, F. R. Relation between water balance and climatic variables associated with the geographical distribution of anurans. PLoS ONE 10, e0140761 (2015).Article 

Google Scholar 
Klockmann, M., Günter, F. & Fischer, K. Heat resistance throughout ontogeny: Body size constrains thermal tolerance. Glob. Change Biol. 23, 686–696 (2017).Article 
ADS 

Google Scholar 
Petrovskii, S., Mashanova, A. & Jansen, V. A. A. Variation in individual walking behavior creates the impression of a Lévy flight. PNAS 108, 8704–8707 (2011).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Lindström, T., Brown, G. P., Sisson, S. A., Phillips, B. L. & Shine, R. Rapid shifts in dispersal behavior on an expanding range edge. PNAS 110, 13452–13456 (2013).Article 
ADS 
PubMed 
PubMed Central 

Google Scholar 
Tingley, R. et al. New weapons in the toad toolkit: A review of methods to control and mitigate the biodiversity impacts of invasive Cane toads (Rhinella marina). Q. Rev. Biol. 92, 123–149 (2017).Article 
PubMed 

Google Scholar 
Novoa, A. et al. Invasion syndromes: A systematic approach for predicting biological invasions and facilitating effective management. Biol. Invasions 22, 1801–1820 (2020).Article 

Google Scholar 
DeVore, J. L., Crossland, M. R., Shine, R. & Ducatez, S. The evolution of targeted cannibalism and cannibal-induced defenses in invasive populations of cane toads. Proc. Natl. Acad. Sci. 118, e2100765118 (2021).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Muller, B. J. & Schwarzkopf, L. Relative effectiveness of trapping and hand-capture for controlling invasive cane toads (Rhinella marina). Int. J. Pest Manag. 64, 185–192 (2018).Article 
CAS 

Google Scholar  More

Test organisms and exposuresIn this study, we used test organisms and reagents according to the Acute Toxicity Test Method of Daphnia magna Straus(Cladocera, Crustacea); ES 04704.1b29. Daphnia magna were fostered at the National Institute of Environmental Research and were adopted. During the test, adult female Daphnia magna over two weeks of age, cultured over several generations, were transferred to a freshly prepared container the day before the test. Daphnia magna are neonates for less than 24 h after birth29. To maintain the sensitivity of the organism, young individuals less than 24 h old that reproduced the following day were used. Individuals of a similar size were selected for the test. Daphnia magna was fed YCT, which is a mixture of green algae in Chlorella sp., yeast, Cerophy II(R), and trout chow. Sufficient amounts of prey were supplied 2 h before the test to minimize the effects of prey during the test. The test medium was prepared by dissolving KCl (8 mg/L), (text {MgSO}_4) (120 mg/L), (text {CaSO}_4 cdot 2 text {H}_2 text {O} ) (120 mg/L), and (text {NaHCO}_3) (192 mg/L) in deionized water.Automatic high-throughput Daphnia magna tracking systemTo build an automatic high-throughput Daphnia magna tracking system, we equipped the system with a video analysis algorithm as well as flow cells (Fig. 1). In the tracking system, six flow cells filled with culture medium were installed in the device. Each flow cell contained 10 Daphnia magna. Subsequently, to automatically measure the state of Daphnia magna, the six flow cells were photographed at 15 frames per second using a camera (Industrial Development Systems imaging) equipped with a CMOSIS sensor capable of infrared imaging. A red light close to the infrared spectrum was placed at the back of the flow cells for uniform illumination and to minimize stress on Daphnia magna. To capture the size and movement of the Daphnia magna as accurately as possible, the camera was set to a frame rate of 15 fps and a resolution of 2048 (times ) 1088 (2.23 MB), using a 12 mm lens. The distance between the flow cell and the camera was set to 16 cm. To measure the number of mobile Daphnia magna, their lethality, and swimming inhibition automatically and simultaneously, one camera for every two cells was used to collect the status data of Daphnia magna. For assessing ecotoxicity, the video analysis system used images obtained from the six flow cells to track each Daphnia magna and estimate key statistics such as the number of mobile individuals, average distance, and radius of activity.Figure 1New automatic high-throughput video tracking system for behavioral analysis using Daphnia magna as a model organismFull size imageThe automatic high-throughput video tracking system in the ecotoxicity measuring device was designed to continuously measure the ecotoxicity of Daphnia magna (Fig. 2). Daphnia magna moves faster at high temperatures and is less active at low temperatures. Thus, a constant temperature module that can be set to an appropriate Daphnia magna habitat temperature (20 ± 2 (^{circ })C) was added to create a suitable culture environment for Daphnia magna29. Natural pseudo-light ((lambda >590) nm, 3000 k) was installed on the upper part of the detector for proper habitat light intensity (500 Lux–1000 Lux). The size of the flow cell was set as small as possible while observing the movement of the Daphnia magna. An automatic feeding system was installed so that food could be injected during the replacement cycle. The six independent multi-flow cells were designed with an automatic dilution injection module; therefore, these flow cells were diluted to six different concentrations (100%, 50%, 25%, 12.5%, 6.25%, and 0%).Figure 2Schematic representation of the automatic high-throughput video tracking systemFull size imageAutomatic tracking algorithmThe CPU used for Daphnia magna tracking was Intel i5-9300H @ 2.40 GHz, with 8 GB of memory and Windows 10 Pro 64-bit operating system. In this experiment, the algorithms were trained using 12 Daphnia magna videos and tested using an additional four Daphnia magna videos. Subsequently, the detection and tracking methods were compared. The videos, each of which had a duration of 30 s, were captured at a rate of 15 frames per second. Generally, for long-time or real-time videos, the following factors must be considered in tracking Daphnia magna: automatic binarization between the object and background, effective classification of Daphnia magna or noise, and the speed of the algorithm. Therefore, to develop an efficient tracking algorithm, we propose the following tracking process (Fig. 3A). In this process, each frame is initially converted into an image and the background is identified from the obtained video (Fig. 3B). The background is the average of the frames over the previous 20 s, and the tracking system takes 20 s to capture the first background image. The background is subtracted from the image for object detection (Fig. 3C). The objects include Daphnia magna and noise such as droplets and sediment. The difference between the background and frame images is binarized, and each area of the binarized values is regarded as an object. Conventionally, the binarized values are manually generated using specific thresholds. In this study, the images are automatically binarized using k-means clustering to select the threshold value. After binarization, several machine learning methods are used to classify the objects as Daphnia magna or noise (Fig. 3D). For a faster tracking algorithm, we use simple machine learning methods such as random forest (RF) and support vector machine (SVM). The predicted Daphnia magna are tracked using SORT24, which is a fast and highly accurate tracking algorithm (Fig. 3E). Finally, based on the tracked results, statistics for assessing ecotoxicity, such as the number of mobile individuals, average distance, and radius of activity, are estimated to evaluate the toxicity of the aquatic environment.Figure 3Automatic Daphnia magna tracking algorithm process. (A) Overview of automatic tracking algorithm process. (B) Image extraction step. (C) Background subtraction step. (D) Daphnia magna detection step. (E) Daphnia magna tracking step.Full size imagek-means clustering for automatic background subtractionMany tracking algorithms assume that the background is fixed. With fixed backgrounds, the difference between the frame and background can be used to identify objects. However, automatically selecting the precise threshold value for image pixel binarization becomes one of the key problems in identifying objects. The proposed method applies k-means clustering to the pixel values of the subtracted image30, and the center value of each calculated cluster mean is selected as the threshold value (Fig. 4). In the k-means clustering method, grouping is repeatedly performed using the distance between data points31. For binarization, two groups are formed. Let (mu _1 (t)) be the mean of pixels less than the threshold and (mu _2(t)) be the mean of pixels greater than the threshold. At first, (mu _1(t), mu _2(t)) are randomly initialized. Subsequently, each pixel is grouped into a closer mean of each group. The above steps are repeated several times until the group experiences a few changes. Finally, the threshold is calculated as an average of the two means.Figure 4Example of automatic threshold value setting for binarization between objects and background using k-means clusteringFull size imageClassification methodsObject detection based solely on the subtraction between the background and frame images may have low accuracy. As the background in the proposed process is the average value of the frame images, noise may occur. Although this noise is removed by threshold selection in binarization, using only the threshold selection is not efficient for long or real-time videos. Therefore, additional noise must be classified and removed using machine learning models, requiring the construction of a database. In the database, the obtained objects are manually labeled as noise or Daphnia magna and are called ground truth. For classification, the resized 8 (times ) 8 image of each object is stored in the database. The resized image is transformed into a feature using the Sobel edge detection algorithm32 and entered as inputs to the classification models. In this study, classification models such as RF33 SVM34 were used.RF is a model that integrates several decision tree models35. All training data are sampled with a replacement for training each decision tree model. The decision tree model is trained to split intervals of each independent variable by minimizing the gini index (Eq. 1) or entropy index (Eq. 2). The gini index and entropy index denote the impurity within the intervals.$$begin{aligned} G= & {} 1- sum _{i=1}^{c} p_i ^2 end{aligned}$$
(1)
$$begin{aligned} E= & {} – sum _{i=1}^{c} p_i log_2 p_i end{aligned}$$
(2)
where (p_i) is a probability within i-th interval, and c is the number of intervals. For better performance, the RF selects independent variables of training data randomly. This step serves to reduce the correlation of each model. If predictions of each decision tree are uncorrelated, then the variance of an integrated prediction of models is smaller than the variance of each model. RF integrates several model predictions using the voting method. An advantage of the RF method is that it avoids overfitting because the model uses the average of many predictions.SVM is a model designed to search for a hyperplane to maximize the distance, or margin, between support vectors. The hyperplane refers to the plane that divides two different groups, and the support vector represents the closest vector to the hyperplane. Let (D=({textbf{x}}_i, y_i), i=1, ldots , n, {textbf{x}}_i in {mathbb {R}}^p, y_n in { -1,1 }) be training data. Suppose that the training data are completely separated linearly by a hyperplane; then, the hyperplane is expressed as Eq. 3.$$begin{aligned} {textbf{w}}^T {textbf{x}} + b = 0, end{aligned}$$
(3)
where ({textbf{w}}) is a weight vector of the hyperplane, and b is a bias. The weight vector is updated by minimizing Eq. 4.$$begin{aligned} L = {1 over 2} {textbf{w}}^T {textbf{w}} text { subject to } y_i ({textbf{w}}^T {textbf{x}} + b) ge 1 end{aligned}$$
(4)
We can transform Eqs. 4 to  5 by using the Lagrange multiplier method.$$begin{aligned} L^* = {1 over 2} {textbf{w}}^T {textbf{w}} – sum _{i=1}^n a_i { y_i ({textbf{w}}^T x_i + {-}) – 1 }, end{aligned}$$
(5)
where (a_i) is the Lagrange multiplier. We can efficiently solve Eq. 5 using a dual form. Furthermore, Eq. 5 can be solved in a case where it is not completely separated using a slack variable and a kernel trick can be used to estimate the nonlinear hyperplane.SORT trackerSORT, one of the frameworks for solving the multiple object tracking (MOT) problem, aims to achieve efficient real-time tracking24. The SORT method framework is created by combining the estimation step and the association step. The estimation step forecasts the next position of each predicted Daphnia magna. The association step matches the forecasting position and next true position of each predicted Daphnia magna. In the estimation step, the SORT framework uses the Kalman filter to forecast the position of the predicted Daphnia magna in the next frame. The position of each predicted Daphnia magna is expressed as Eq. 6.$$begin{aligned} {textbf{x}} = [u,v,s,r,{dot{u}}, {dot{v}}, {dot{s}}]^T end{aligned}$$
(6)
where u and v are the center positions of each predicted Daphnia magna, s is the scale size of the bounding box, and r is the aspect ratio of the bounding box. ({dot{u}}), ({dot{v}}), and ({dot{s}}) are the amounts of change in each variable. In the association step, to associate the forecasting position and true position, the framework adopts the intersection-over-union (IOU)36 as the association metric. The Hungarian algorithm is loaded into the SORT framework to perform fast and efficient Daphnia magna association prediction. In this study, a mixed metric of IOU36 and Euclidean distance37 was used instead of only the IOU that is used in SORT (Eq. 7) for more efficient association.$$begin{aligned} C_{ij} = (1-lambda ) {max_d – d_{ij} over max_d} + lambda cdot IOU_{ij} end{aligned}$$
(7)
where (d_{ij}) is the Euclidean distance between the i-th predicted Daphnia magna in the before frame and the j-th predicted Daphnia magna in the next frame, and (lambda ) is the weight of (IOU_{ij}). (IOU_{ij}) is the IOU between the i-th predicted Daphnia magna in the before-frame and the j-th predicted Daphnia magna in the next frame.MetricsThe binary confusion matrix consists of true positive (TP), true negative (TN), false positive (FP), and false negative (FN)38. TP is the number of cases where the predicted Daphnia magna matches the actual Daphnia magna, TN is the number of cases where the objects predicted as noise are actual noise, FP is the number of cases where the predicted Daphnia magna differs from the actual Daphnia magna, and FN is the number of cases where the objects predicted as noise are not actual noise. In this study, accuracy, recall, precision, and F1 scores (Eq. 8) were used as the metrics for comparing the machine learning methods.$$begin{aligned} begin{aligned} Accuracy&= {TP + FP over TP + TN + FP + FN} \ Recall&= {TP over TP + TN} \ Precision&= {TP over TP + FP} \ F1 score&= 2 times {Precision times Recall over Precision + Recall} end{aligned} end{aligned}$$
(8)
Standard MOT metrics to evaluate tracking performance include multi-object tracking accuracy (MOTA) and multi-object tracking precision (MOTP). An important task of MOT is to identify and track the same object across two frames. Identification (ID) precision (IDP), ID recall (IDR), ID F1 measure (IDF1), and ID switches (IDs) may be used as measures for evaluating the identification and tracking of the same objects39,40.Data analysisThe toxicity test using Daphnia magna was performed following the Korean official Acute Toxicity Test Method29. The test medium was prepared by dissolving KCl (8 mg/L), (text {MgSO}_4) (120 mg/L), (text {CaSO}_4 cdot 2 text {H}_2 text {O} ) (120 mg/L), and (text {NaHCO}_3) (192 mg/L) in deionized water. Considering that Daphnia magna are neonates for less than 24 h after birth29, five neonates were exposed to 50 mL of different concentrations of heavy metals such as Potassium dichromate, Copper(II) sulfate pentahydrate, and Lead(II) sulfate (6.25, 12.5, 25, 50, and 100%) and 50 mL of culture media. Potassium dichromate is a common inorganic reagent used as an oxidizing agent in chemical industries. Copper(II) sulfate pentahydrate is a trace material widely used in industrial processes and agriculture. A significant amount of copper is emitted in semiconductor manufacturing processes, which adversely impacts the aquatic ecosystem. When present as an ion in water, copper can be acutely toxic to aquatic organisms such as Daphnia magna. Lead(II) sulfate is another nonessential and nonbiodegradable heavy metal. It is highly toxic to numerous organisms even at low concentrations and can accumulate in aquatic ecosystems41. Twenty Daphnia magna (four replicates of five each) were exposed to each test solution for 24 h. The term “immobility” means that the Daphnia magna remains stationary after exposure to chemicals such as Potassium dichromate, Copper(II) sulfate pentahydrate, and Lead(II) sulfate. In this study, immobility was used as an endpoint identifier, and the number of mobile Daphnia magna were counted to evaluate the EC50 values for the samples using the ToxCalc 5.0 program (Tidepoll Software, USA).The locomotory responses of Daphnia magna were tested after 0, 12, 18, and 24 h of exposure at different concentrations. Potassium dichromate ((text {K}_2text {Cr}_2text {O}_7)) at 2 mg/L was connected to the Daphnia magna tracking system, and standard toxic substances were automatically diluted to 100%, 50%, 25%, 12.5%, and 6.25%. The automatic high-throughput Daphnia magna tracking system automatically measured the tracking results of a 1-minute-long video at hourly intervals. The average moving distance for 20 s of each Daphnia magna in each chamber was analyzed using a repeated measures ANOVA (RMANOVA). RMANOVA was used for the analysis of data obtained by repeatedly measuring the same Daphnia magna42. It analyzes the concentration effect excluding the time effect at each hour. The time effect means the change in average distance per 20 s. RMANOVA was implemented using the agricolae package of the R 4.0.4 program43. To remove the noise affecting RMANOVA, the Daphnia magna that remained stationary for 20 s or more were removed from the observations. In this study, we used the significance level at 5%. More

A hippo swims in Colombia’s Magdalena River, near where Pablo Escobar’s compound was located.Credit: Fernando Vergara/AP/Shutterstock

Colombian environment minister Susana Muhamad has triggered fear among researchers that she will protect, rather than reduce, a growing population of invasive hippos that threaten the country’s natural ecosystems and biodiversity. Although she did not directly mention the hippos — a contentious issue in Colombia — Muhamad said during a speech in late January that her ministry would create policies that prioritize animal well-being, including the creation of a new division of animal protection.
Landmark Colombian bird study repeated to right colonial-era wrongs
The hippos escaped from drug-cartel leader Pablo Escobar’s estate after he died in 1993. Left alone, the male and three females that Escobar had illegally imported from a US zoo established themselves in Colombia’s Magdalena River and some small lakes nearby — part of the country’s main watershed. After years of breeding, the ‘cocaine hippos’ have multiplied to about 150 individuals, scientists estimate.Given that the hippos (Hippopotamus amphibius) — considered the largest invasive animal in the world — have no natural predators in Colombia and have been mating at a steady rate, their population could reach 1,500 in 16 years, according to a modelling study published in 20211. “I do not understand what the government is waiting for to act,” says Nataly Castelblanco Martínez, a Colombian conservation biologist at the Autonomous University of Quintana Roo in Chetumal, Mexico, and co-author of the study. “If we don’t do anything, 20 years from now the problem will have no solution.”Researchers have called for a strict management plan that would eventually reduce the wild population to zero, through a combination of culling some animals and capturing others, then relocating them to facilities such as zoos. But the subject of what to do with the hippos has polarized the country, with some enamoured by the animals’ charisma and value as a tourist attraction and others concerned about the threat they pose to the environment and local fishing communities.‘A bit surreal’Several studies and observations suggest how destructive it could be to allow the Colombian hippo population to explode. A 2019 paper2, for example, showed that, compared with lakes without hippos, those where the animals have taken up residence contain more nutrients and organic matter that favour the growth of cyanobacteria — aquatic microbes associated with toxic algal blooms. These blooms can reduce water quality and cause mass fish deaths, affecting local fishing communities.

A sign near Doradal, Colombia, warns passersby of the danger of invasive hippos.Credit: Juancho Torres/Anadolu Agency via Getty

Other scientists have predicted that the hippos could displace endangered species that are native to the Magdalena River, such as the Antillean manatee (Trichechus manatus manatus), by outcompeting them for food and space. They caution that traffic accidents and attacks on people caused by the hippos will become more common. And they warn that wildlife traffickers are already taking advantage of the situation by illegally selling baby hippos — a trend that could intensify.“It’s a bit surreal,” says Jorge Moreno Bernal, a vertebrate palaeontologist at the University of the North in Barranquilla, Colombia. “This is just a taste of what may come.”When Colombian authorities first recognized the speed at which the hippo population was growing, during the 2000s, they acted to reduce their numbers. But in 2009, when photos appeared online after soldiers gunned down Pepe, Escobar’s fugitive male hippo, the outcry from animal-rights activists and others plunged the environment ministry into an “institutional paralysis”, says Sebastián Restrepo Calle, an ecologist at Javeriana University in Bogotá.Researchers say that the hippos don’t belong in Colombia — they are native to sub-Saharan Africa. Simulations run by Castelblanco Martínez and her colleagues suggest that to reduce the population to zero by 2033, about 30 hippos would need to be removed from the wild population per year1. No other course of action, including sterilization or castration, would eradicate them, according to the modelling of various management scenarios, says Castelblanco Martínez.The cost of inactionThe worry now is that, instead of basing decisions on evidence and expertise in conservation, the government is listening to popular opinion, says Restrepo Calle. Neither Muhamad nor representatives of the environment ministry replied to Nature’s requests for comment.
Ancient stone tools suggest early humans dined on hippo
“Why prioritize one species over our own ecosystems?” — especially a species that isn’t native, asks Alejandra Echeverri, a Colombian conservation scientist at Stanford University in California. Along with her colleagues, Echeverri published a study last month showing that Colombia has few policies governing invasive species compared with its overall number of biodiversity policies3.Animals-rights advocates, meanwhile, argue that they aren’t ignoring environmental concerns. Luis Domingo Gómez Maldonado, an animal-rights activist and specialist in animal law at Saint Thomas University in Bogotá, says “It’s not about saving the hippos on a whim,” but rather about solving the issue while also giving the hippos justice. “My indisputable position is: let’s save as many individuals as possible, let’s do it ethically.”Researchers, too, say they have the animals’ best interests at heart. “Even if [advocates] don’t see it, we care about the hippos,” Castelblanco Martínez says. “The more time that passes, the more hippos will either have to be culled, castrated or captured.”The question is whether environmental authorities will act swiftly to draft and enforce a management plan that is both ethical and effective. Should they sit on the issue for too long, Castelblanco Martínez warns, rural communities that are most affected by the hippos might take matters into their own hands.If the government doesn’t cull them, she says, people will use shotguns to do it. More

Knowlton, N. The future of coral reefs. Proc. Natl. Acad. Sci. 98, 5419–5425 (2001).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Hughes, T. P. et al. Climate change, human impacts, and the resilience of coral reefs. Science 1979(301), 929–933 (2003).Article 
ADS 

Google Scholar 
Hoegh-Guldberg, O. et al. Coral reefs under rapid climate change and ocean acidification. Science 1979(318), 1737–1742 (2007).Article 
ADS 

Google Scholar 
Eakin, C. M. et al. Monitoring coral reefs from space. Oceanography 23, 118–133 (2010).Article 

Google Scholar 
Hughes, T. P. et al. Global warming transforms coral reef assemblages. Nature 556, 492–496 (2018).Article 
ADS 
CAS 
PubMed 

Google Scholar 
Baker, D. M., Freeman, C. J., Wong, J. C. Y., Fogel, M. L. & Knowlton, N. Climate change promotes parasitism in a coral symbiosis. ISME J. 12, 921–930 (2018).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Hughes, T. P. et al. Global warming and recurrent mass bleaching of corals. Nature 543, 373–377 (2017).Article 
ADS 
CAS 
PubMed 

Google Scholar 
Berkelmans, R. & van Oppen, M. J. H. The role of zooxanthellae in the thermal tolerance of corals: a ‘nugget of hope’ for coral reefs in an era of climate change. Proc. R. Soc. B Biol. Sci. 273, 2305–2312 (2006).Article 

Google Scholar 
Byler, K. A., Carmi-Veal, M., Fine, M. & Goulet, T. L. Multiple symbiont acquisition strategies as an adaptive mechanism in the coral Stylophora pistillata. PLoS One 8, e59596 (2013).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Cumbo, V., van Oppen, M. & Baird, A. Temperature and Symbiodinium physiology affect the establishment and development of symbiosis in corals. Mar. Ecol. Prog. Ser. 587, 117–127 (2018).Article 
ADS 
CAS 

Google Scholar 
Mundy, C. N. & Babcock, R. C. Role of light intensity and spectral quality in coral settlement: Implications for depth-dependent settlement?. J. Exp. Mar Biol. Ecol. 223, 235–255 (1998).Article 

Google Scholar 
Gleason, D. F., Edmunds, P. J. & Gates, R. D. Ultraviolet radiation effects on the behavior and recruitment of larvae from the reef coral Porites astreoides. Mar. Biol. 148, 503–512 (2006).Article 

Google Scholar 
Yusuf, S., Zamani, N. P., Jompa, J. & Junior, M. Z. Larvae of the coral Acropora tenuis (Dana 1846) settle under controlled light intensity. IOP Conf. Ser. Earth Environ. Sci. 253, 012023 (2019).Article 

Google Scholar 
Vermeij, M. J. A., Marhaver, K. L., Huijbers, C. M., Nagelkerken, I. & Simpson, S. D. Coral larvae move toward reef sounds. PLoS One 5, e10660 (2010).Article 
ADS 
PubMed 
PubMed Central 

Google Scholar 
Doropoulos, C. et al. Characterizing the ecological trade-offs throughout the early ontogeny of coral recruitment. Ecol. Monogr. 86, 20–44 (2016).Article 

Google Scholar 
Morse, D. E., Hooker, N., Morse, A. N. C. & Jensen, R. A. Control of larval metamorphosis and recruitment in sympatric agariciid corals. J. Exp. Mar. Biol. Ecol. 116, 193–217 (1988).Article 

Google Scholar 
Price, N. Habitat selection, facilitation, and biotic settlement cues affect distribution and performance of coral recruits in French Polynesia. Oecologia 163, 747–758 (2010).Article 
ADS 
PubMed 
PubMed Central 

Google Scholar 
Ritson-Williams, R., Arnold, S. N., Paul, V. J. & Steneck, R. S. Larval settlement preferences of Acropora palmata and Montastraea faveolata in response to diverse red algae. Coral Reefs 33, 59–66 (2014).Article 
ADS 

Google Scholar 
Negri, A., Webster, N., Hill, R. & Heyward, A. Metamorphosis of broadcast spawning corals in response to bacteria isolated from crustose algae. Mar. Ecol. Prog. Ser. 223, 121–131 (2001).Article 
ADS 

Google Scholar 
Webster, N. S. et al. Metamorphosis of a scleractinian coral in response to microbial biofilms. Appl. Environ. Microbiol. 70, 1213–1221 (2004).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Erwin, P. M., Song, B. & Szmant, A. M. Settlement behavior of Acropora palmata planulae: effects of biofilm age and crustose coralline algal cover. In Proceedings of 11th International Coral Reef Symposium 24, (2008).Siboni, N. et al. Crustose coralline algae that promote coral larval settlement harbor distinct surface bacterial communities. Coral Reefs 39, 1703–1713 (2020).Article 

Google Scholar 
Petersen, L.-E. et al. Mono- and multispecies biofilms from a crustose coralline alga induce settlement in the scleractinian coral Leptastrea purpurea. Coral Reefs 40, 381–394 (2021).Article 

Google Scholar 
Jorissen, H. et al. Coral larval settlement preferences linked to crustose coralline algae with distinct chemical and microbial signatures. Sci. Rep. 11, 14610 (2021).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Tebben, J. et al. Induction of larval metamorphosis of the coral Acropora millepora by tetrabromopyrrole isolated from a pseudoalteromonas bacterium. PLoS One 6, e19082 (2011).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Tebben, J. et al. Chemical mediation of coral larval settlement by crustose coralline algae. Sci. Rep. 5, 10803 (2015).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Tran, C. & Hadfield, M. Larvae of Pocillopora damicornis (Anthozoa) settle and metamorphose in response to surface-biofilm bacteria. Mar. Ecol. Prog. Ser. 433, 85–96 (2011).Article 
ADS 

Google Scholar 
Sneed, J. M., Sharp, K. H., Ritchie, K. B. & Paul, V. J. The chemical cue tetrabromopyrrole from a biofilm bacterium induces settlement of multiple Caribbean corals. Proc. R. Soc. B Biol. Sci. 281, 20133086 (2014).Article 

Google Scholar 
Petersen, L.-E., Kellermann, M. Y., Nietzer, S. & Schupp, P. J. Photosensitivity of the bacterial pigment cycloprodigiosin enables settlement in coral larvae—light as an understudied environmental factor. Front. Mar. Sci. 8, 749070 (2021).Article 

Google Scholar 
Heyward, A. J. & Negri, A. P. Natural inducers for coral larval metamorphosis. Coral Reefs 18, 273–279 (1999).Article 

Google Scholar 
Harrington, L., Fabricius, K., Death, G. & Negri, A. Recognition and selection of settlement substrata determine post-settlement survival in corals. Ecology 85, 3428–3437 (2004).Article 

Google Scholar 
Da-Anoy, J. P., Villanueva, R. D., Cabaitan, P. C. & Conaco, C. Effects of coral extracts on survivorship, swimming behavior, and settlement of Pocillopora damicornis larvae. J. Exp. Mar. Biol. Ecol. 486, 93–97 (2017).Article 

Google Scholar 
Morse, D. E. & Morse, A. N. C. Enzymatic characterization of the morphogen recognized by Agaricia humilis (Scleractinian Coral) larvae. Biol. Bull. 181, 104–122 (1991).Article 
CAS 
PubMed 

Google Scholar 
Kitamura, M., Koyama, T., Nakano, Y. & Uemura, D. Characterization of a natural inducer of coral larval metamorphosis. J. Exp. Mar. Biol. Ecol. 340, 96–102 (2007).Article 

Google Scholar 
Kitamura, M., Schupp, P. J., Nakano, Y. & Uemura, D. Luminaolide, a novel metamorphosis-enhancing macrodiolide for scleractinian coral larvae from crustose coralline algae. Tetrahedron Lett. 50, 6606–6609 (2009).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Maru, N. et al. Relative configuration of luminaolide. Tetrahedron Lett. 54, 4385–4387 (2013).Article 
CAS 

Google Scholar 
Nietzer, S., Moeller, M., Kitamura, M. & Schupp, P. J. Coral larvae every day: Leptastrea purpurea, a brooding species that could accelerate coral research. Front. Mar. Sci. 5, 466 (2018).Article 

Google Scholar 
Moeller, M., Nietzer, S. & Schupp, P. J. Neuroactive compounds induce larval settlement in the scleractinian coral Leptastrea purpurea. Sci. Rep. 9, 2291 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Petersen, L.-E., Kellermann, M. Y. & Schupp, P. J. Secondary metabolites of marine microbes: from natural products chemistry to chemical ecology. In YOUMARES 9 – The Oceans: Our Research, Our Future: Proceedings of the 2018 Conference for Young Marine Researcher in Oldenburg, Germany (eds Jungblut, S. et al.) 159–180 (Springer International Publishing, 2020). https://doi.org/10.1007/978-3-030-20389-4_8.Chapter 

Google Scholar 
Fiegel, L. J. et al. A detailed visualization of the early development stages of Leptastrea purpurea reveals distinct bio-optical features. Front. Mar. Sci. 10, 1–10 (2023).
Google Scholar 
Strader, M. E., Aglyamova, G. V. & Matz, M. V. Molecular characterization of larval development from fertilization to metamorphosis in a reef-building coral. BMC Genom. 19, 17 (2018).Article 

Google Scholar 
Puisay, A. et al. Parental bleaching susceptibility leads to differences in larval fluorescence and dispersal potential in Pocillopora acuta corals. Mar. Environ. Res. 163, 105200 (2021).Article 
CAS 
PubMed 

Google Scholar 
Perez-Tomas, R. & Vinas, M. New insights on the antitumoral properties of prodiginines. Curr. Med. Chem. 17, 2222–2231 (2010).Article 
CAS 
PubMed 

Google Scholar 
You, Z. et al. Insights into the anti-infective properties of prodiginines. Appl. Microbiol. Biotechnol. 103, 2873–2887 (2019).Article 
CAS 
PubMed 

Google Scholar 
Kellermann, M. Y., Yoshinaga, M. Y., Valentine, R. C., Wörmer, L. & Valentine, D. L. Important roles for membrane lipids in haloarchaeal bioenergetics. Biochim. Biophys. Acta (BBA) Biomembr. 1858, 2940–2956 (2016).Article 
CAS 

Google Scholar 
Hirose, M., Yamamoto, H. & Nonaka, M. Metamorphosis and acquisition of symbiotic algae in planula larvae and primary polyps of Acropora spp.. Coral Reefs 27, 247–254 (2008).Article 
ADS 

Google Scholar 
Bollati, E. et al. Green fluorescent protein-like pigments optimize the internal light environment in symbiotic reef building corals. Elife 11, e73521 (2022).Article 
PubMed 
PubMed Central 

Google Scholar 
Palmer, C. V., Modi, C. K. & Mydlarz, L. D. Coral fluorescent proteins as antioxidants. PLoS One 4, e7298 (2009).Article 
ADS 
PubMed 
PubMed Central 

Google Scholar 
Alegado, R. A. et al. A bacterial sulfonolipid triggers multicellular development in the closest living relatives of animals. Elife 1, e00013 (2012).Article 
PubMed 
PubMed Central 

Google Scholar 
Woznica, A. et al. Bacterial lipids activate, synergize, and inhibit a developmental switch in choanoflagellates. Proc. Natl. Acad. Sci. 113, 7894–7899 (2016).Article 
ADS 
CAS 
PubMed 
PubMed Central 

Google Scholar 
He, J. et al. Bacterial nucleobases synergistically induce larval settlement and metamorphosis in the invasive mussel Mytilopsis sallei. Appl. Environ. Microbiol. 85, e01039 (2019).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Guo, H., Rischer, M., Westermann, M. & Beemelmanns, C. Two distinct bacterial biofilm components trigger metamorphosis in the colonial hydrozoan Hydractinia echinata. MBio 12, e00401 (2021).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Ross, C., Fogarty, N. D., Ritson-Williams, R. & Paul, V. J. Interspecific variation in coral settlement and fertilization success in response to hydrogen peroxide exposure. Biol. Bull. 233, 206–218 (2017).Article 
CAS 
PubMed 

Google Scholar 
Boettcher, A. A., Dyer, C., Casey, J. & Targett, N. M. Hydrogen peroxide induced metamorphosis of queen conch, Strombus gigas: tests at the commercial scale. Aquaculture 148, 247–258 (1997).Article 
CAS 

Google Scholar 
Covarrubias, L., Hernández-García, D., Schnabel, D., Salas-Vidal, E. & Castro-Obregón, S. Function of reactive oxygen species during animal development: Passive or active?. Dev. Biol. 320, 1–11 (2008).Article 
CAS 
PubMed 

Google Scholar 
Gauron, C. et al. Hydrogen peroxide (H2O2) controls axon pathfinding during zebrafish development. Dev. Biol. 414, 133–141 (2016).Article 
CAS 
PubMed 

Google Scholar  More

Study site, tree selections, and drought-simulation experimentThis research was performed in Guangzhou (22°26′-23°56′N, 112°57′-114°03′E), which is a core city located in subtropical zones. With an area of 7434.4 km2 and a population of 18.87 million, Guangzhou’s urbanization rate has reached 86.46%. To cope with multiple environmental challenges, several urban-forest nurseries were established to cultivate and introduce various tree species. Among them, we selected the one in Tianhe District as our study site. This nursery was not only abundant with native and exotic tree species but also equipped with similar edatope in cities, which was ideal for our research.Tilia cordata Mill. (Tc) and Tilia tomentosa Moench (Tt), originating from the west of Britain and southeast of Europe, were common urban tree species planted in European cities. Based on their performance in providing ecological and landscape functions, these two tree species were considered to be introduced for urban greening. Therefore, Tilia cordata Mill. (Tc) and Tilia tomentosa Moench (Tt) were selected as our objectives, which were investigated for their growth and ecosystem services to evaluate their adaption in Guangzhou. In addition, a native tree species Tilia miqueliana Maxim (Tm) was also implemented concurrent measurement as a comparison.For each of the three surveyed tree species, ten trees with a diameter at breast height (DBH) around 5.5 cm and tree height around 2.5 m were chosen for our experiment, which were thought to possess similar initial statuses. To investigate the impact of drought on the growth and ecosystem services of the three selected tree species, a controlled experiment was launched from January to December in 2020. For each tree species, five trees were planted in the common environment as the controlled group, while the other five trees were under the precipitation-exclusion installation (PEI) as the drought-simulation group. Consisting of several water-proof tents, PEI was adequately large and could completely prevent trees from obtaining rainfalls, which created a precipitation-exclusive environment to simulate an enduring drought event within the whole research period (Fig. 1).Figure 1Schematic diagram of the drought simulation experiment for the three surveyed tree species.Full size imageEnvironmental monitoring systemsClimatic data were sampled every 10 min with a weather station (WP3103 mesoscale automatic weather station, China) located at an unshaded site in the nursery. The data were stored in the logger and copied to our laboratory to produce daily or monthly data. All the climatic variables, including photosynthetically active radiation (PAR, µmol m-2 s-1), wind speed (m s-1), precipitation (mm), and air temperature (°C) were calculated from January to December in 2020.For volumetric soil water content (%; VWC), the HOBO MX2307 system (Onetemp, Adelaide, Australia), placed in a shaded box in the nursery, was applied for all the three tree species from both the controlled and drought-simulation groups. For each individual tree, the sensing probe was inserted horizontally at the depths of 30 cm and located 20 cm in the northern direction from the tree stems. Based on the daily readings, monthly means were calculated from January to December in 2020.Measurement of above-ground growthTo investigate the above-ground growth of the three tree species from both the controlled and drought-simulation groups, their DBH (diameter at breast height, cm), tree height (m), and LAI (leaf area index) were measured at the beginning of each month in 2020. DBH was measured with the help of a caliper (Altraco Inc., Sausalito, California, USA), and their tree heights were measured using a standard tape. The crown analytical instrument CI-110 (Camas, Washington State, USA) was used to capture an accurate image of tree crowns and calculate LAI. Sufficient numbers of points were measured and recorded to describe each tree’s average crown shape. The software FV2200 (LICOR Biosciences, Lincoln, NE) helped compute each tree’s crown width and crown area.Measurement of below-ground growthFine root coring campaigns were launched for all the trees of the three tree species from both the controlled and drought treatment groups every three to four months, i.e., in February, May, September, and December. Although the coring campaign might damage part of the roots, the fine roots obtained each time were a mere portion of the whole root system, not affecting the general development of trees’ underground processes. For every individual tree, two 30-cm soil cores were applied in each direction of north, south, east, and west, of which one was located at 20 cm to the trunk (paracentral roots) and the other one was located at 40 cm (outer roots). In addition, the soil samples were evenly divided into three horizons which were 0–10 cm (shallow layer), 10–20 cm (middle layer), and 20–30 cm (deep layer). Then a sieve with 2-mm mesh size was used to filter all the fine roots. The fine roots were washed carefully to remove the adherent soils and dried in an oven at 65 ℃ for 72 h. Finally, all the samples were weighed using a balance with an accuracy of four decimal places to obtain the dry weight. The fine root biomass at different depths was calculated using the dry weight divided by the cross-sectional area of the auger20.Model’s simulation of ecosystem servicesThe process-based model City-Tree was used to predict the ecosystem services of the three tree species from both the controlled and drought-simulation groups23. The model required the data of tree growth parameters including tree height, DBH, and crown area together with environmental conditions such as edaphic and climatic data24. In this research, cooling, evapotranspiration and CO2 fixation of the three surveyed tree species in the controlled and drought-treatment groups were simulated at the end of 2020.The actual evapotranspiration eta was calculated from the potential evapotranspiration using fetp[t], Tilia’s factors fetp[t], and the reduction factor fred:$${mathrm{et}}_{mathrm{a}}={mathrm{f}}_{mathrm{red}}*{mathrm{f}}_{mathrm{etp}}left[mathrm{t}right]*{mathrm{et}}_{mathrm{p}}$$The process of tree’s evapotranspiration (etp) was calculated on the basis of SVAT algorithm together with Penman formula in the module on water balance as below:$${mathrm{et}}_{mathrm{p}}=left[mathrm{s }/ left(mathrm{s}+upgamma right)right]*left({mathrm{r}}_{mathrm{s}}-{mathrm{r}}_{mathrm{L}}right) /mathrm{ L}+left[1-mathrm{s }/ left(mathrm{s}+upgamma right)right]*{mathrm{e}}_{mathrm{s}}*mathrm{f }left({mathrm{v}}_{mathrm{u}}right)$$with γ: psychrometric constant in hPa K−1; s: the slope of the saturation vapour pressure curve in hPa K−1; rs: short wave radiation balance in W m−2; rL: long-wave radiation balance in W m−2; L: specific evaporation heat in W m−2 mm−1 d; es: saturation deficit in hPa; f (vu): ventilation function with vu being the daily average wind speed in m s−1.Within the module cooling, the energy needed for the transition of water from liquid to gaseous phase was calculated based on the crown area (CA) and the transpiration eta sum:$${mathrm{E}}_{mathrm{A}}= {mathrm{et}}_{mathrm{a}}*mathrm{CA}-left({mathrm{L}}_{mathrm{O}}* -0.00242*mathrm{temp}right) / {mathrm{f}}_{mathrm{con}}$$with EA: energy released by a tree through transpiration (kWh tree-1), LO: energy needed for the transition of the 1 kg of water from the liquid to gaseous phase = 2.498 MJ (kgH2O)-1 and temp = temperature in ℃, fcon: 0.5.The calculation of new assimilation in the module of photosynthesis and respiration was on the basis of the approach of Haxeltine and Prenticem25. The model assumed that 50% of the incoming short-wave radiation is photosynthetic active radiation (PAR). Using the LAI and a light extinction factor of 0.5, the radiation amount of 1 m2 leaf area can be estimated based on an exponential function according to the Lambert–Beer law. This way, the gross assimilation per m2 leaf area as the daily mean of the month can be derived from:$${text{A}} = {text{d}}*{{left[ {left( {{text{J}}_{{text{p}}} + {text{J}}_{{text{r}}} – {text{sqrt}} left( {left( {{text{J}}_{{text{P}}} + {text{J}}_{{text{r}}} } right)^{2} – 4*uptheta *{text{J}}_{{text{p}}} *{text{J}}_{{text{r}}} } right)} right)} right]} mathord{left/ {vphantom {{left[ {left( {{text{J}}_{{text{p}}} + {text{J}}_{{text{r}}} – {text{sqrt}} left( {left( {{text{J}}_{{text{P}}} + {text{J}}_{{text{r}}} } right)^{2} – 4*uptheta *{text{J}}_{{text{p}}} *{text{J}}_{{text{r}}} } right)} right)} right]} {left( {2*uptheta } right)}}} right. kern-0pt} {left( {2*uptheta } right)}}$$with A: gross assimilation [g C m−2 d−1]; d: mean day length of the month [h]; Jp: reaction of photosynthesis on absorbed photosynthetic radiation [g C m−2 h−1]; Jr: rubisco limited rate of photosynthesis [g C m−2 h−1]; θ: form factor = 0.7.Jp was defined as a function of the photosynthetic active radiation PAR in mol m−2 h−1 and the efficiency of carbon fixation per absorbed PAR [g C mol−1].$${text{J}}_{{text{p}}} = {text{c}}_{{text{p}}} {text{*PAR}}$$$${text{c}}_{{text{p}}} = alpha *left( {{text{p}}_{{{text{ci}}}} – {text{r}}} right){ /}left( {{text{p}}_{{{text{ci}}}} – {text{r}}} right)*gamma *{text{m}}_{{{text{co}}_{2} }} *{text{i}}left[ {text{t}} right]$$with α: intrinsic quantum efficiency for CO2 uptake = 0.08; pci: partial pressure of the internal CO2 [Pa]; r: CO2 compensation point [Pa]; ϒ: species dependent adjustment function for tree age; m CO2: molecular mass of C = 12.0 g mol−1; i[t]: influence of temperature on efficiency.Net assimilation AN [g C m−2 d−1] was then derived from the gross assimilation A and the dark respiration Rd by:$${text{A}}_{{text{N}}} = {text{A}} – {text{R}}_{{text{d}}}$$$${text{R}}_{{text{d}}} =upbeta *{text{V}}_{{text{m}}}$$where Vm was calculated as:$${text{V}}_{{text{m}}} = {1 mathord{left/ {vphantom {1 upbeta }} right. kern-0pt} upbeta } * {{{text{c}}_{{text{p}}} } mathord{left/ {vphantom {{{text{c}}_{{text{p}}} } {{text{c}}_{{text{r}}} * {text{PAR}} * left[ {left( {2uptheta – 1} right) * beta * {{text{d}} mathord{left/ {vphantom {{text{d}} {{text{d}}_{max } }}} right. kern-0pt} {{text{d}}_{max } }} – left( {2uptheta *upbeta *{{text{d}} mathord{left/ {vphantom {{text{d}} {{text{d}}_{max } }}} right. kern-0pt} {{text{d}}_{max } }} – {text{c}}_{{text{r}}} } right)*varsigma } right]}}} right. kern-0pt} {{text{c}}_{{text{r}}} * {text{PAR}} * left[ {left( {2theta – 1} right) * beta * {{text{d}} mathord{left/ {vphantom {{text{d}} {{text{d}}_{max } }}} right. kern-0pt} {{text{d}}_{max } }} – left( {2theta *upbeta *{{text{d}} mathord{left/ {vphantom {{text{d}} {{text{d}}_{max } }}} right. kern-0pt} {{text{d}}_{max } }} – {text{c}}_{{text{r}}} } right)*varsigma } right]}}$$By multiplying AN, the number of days and the total leaf area, the entire monthly net assimilation of the tree can be obtained. In this study, we assumed a fixed share of 50% as respiration based on the gross primary production that the resulting net primary production NPP was transformed in the content of fixed carbon by multiplying the value with the carbon conversion factor 0.524.$${mathrm{Carbon}}_{mathrm{fix}}=0.5*mathrm{NPP}$$Statistical analysesThe software package R was used for statistical analysis. To investigate the differences between means, two-sampled t-test and analysis of variance (ANOVA) with Tukey’s HSD (honestly significant difference) test were used. All the cases, the means were reported as significant when P  More

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.This is a summary of: Zhu, L. et al. Comparable biophysical and biogeochemical feedbacks on warming from tropical moist forest degradation. Nat. Geosci. https://doi.org/10.1038/s41561-023-01137-y (2023). More

Terraces are a land type that is defined by its shape. They have a distinct morphological structure and edge features that distinguish them from other land types. In this study, we define terraces as agricultural land with strip or wavy sections built on slopes greater than 2° along the contour direction. Figure 1 depicts Google Maps satellite images of terraces in the Loess Plateau region. Terraces can be distinguished from other features in remote sensing images based on their colour, morphology, texture, and structure. Terraces can be distinguished from construction land, water, glaciers, and deserts by their colours. Figure 1b–d shows terraces that are primarily green and yellow. Furthermore, terraces are generally distributed along the contour direction, and can therefore be identified based on their morphology. Terraced field ridges curve downward and resemble strips in Fig. 1b,d or circles or ovals in Fig. 1c rather than a neat grid-like distribution. These features differ in morphology from the flat land shown in Fig. 1h. Based on texture and structure, the field area of terraces can be identified based on their strong edge features, as shown in Fig. 1b–d. The edges of terraces have dark stripes caused by oblique illumination received from the sun, and the field ridge of terraces often intercepts part of the sunlight due to their height. Sloping cultivated land, as shown in Fig. 1g, has no evident terraced wall. The outline of sloping cultivated land in the high-resolution image is curved, with no prominent edge features. These findings are critical differences distinguishing terraces and sloping land in high-resolution images.Fig. 1The spatial location of the Loess Plateau and images of various types of cultivated land. (a) The spatial location of the Loess Plateau and Spatial distribution of various cultivated land types images, (b) wide strip-mounted terraces in Longxi, (c) circular wide terraces in central Yulin, (d) high resolution image of Zhuanglang County in July 2019, (e) Zhuanglang County in February 2020, (f) narrow terraces in Shangbao, Chongyi, Jiangxi Province, (g) sloping cropland in Zhenjiang Town, Laibin, Guangxi, and (h) horizontal cropland in the North China Plain.Full size imageDeep learning-based terrace extraction modelThe DLTEM is a terrace extraction model that uses deep learning algorithms and other supplementary information. Initially, a preliminary terrace distribution map was obtained using a deep learning algorithm. It was then combined with the spectral and digital elevation model (DEM) elevation information to fine-tune the results. The final spatial distribution of the terraces was produced by manual correction (Fig. 2). Traditional land classification models or methods typically superimpose spectral, elevation, and morphological texture information from remote sensing images together for training, such as random forest, which is easily ignored in training since morphological texture information accounts for a relatively small amount of the total information. This leads to significant errors while identifying land classes with textural characteristics. In contrast, the DLTEM focuses on morphological texture information from remote sensing images and classifies it into land classes, followed by auxiliary correction through additional information. Thus, this method is more suitable to extract terraces enriched with texture structure information.Fig. 2Flow chart of the deep learning-based terrace extraction model.Full size imageThe UNet++ network is a classic deep learning algorithm that is uniquely unrivaled in extracting colour, morphology, texture, and structure features from images and applying them for classification. In comparison with other Convolutional Neural Network (CNN) classification models (e.g., Fully Convolutional Networks (FCN)), it has high classification accuracy, fast computation speed, strong robustness, and provides variable importance metrics. Therefore, in this study, the UNet++ network was adopted as the network framework for deep learning; the primary data source used was high-resolution satellite imagery from 2019. DEM (SRTM v4.1) data were used to obtain the elevation information and GlobeLand30 data were used to obtain the spectral information. The results were corrected to construct the final map of the distribution of terraces in the Loess Plateau.Study areaThe Loess Plateau, one of China’s four major plateaus, is located in northern central China (34°–40° N and 103°–114° E) (Fig. 1). It is covered by a thick loess layer that ranges in thickness from 50 to 80 m, and is the world’s largest loess deposition area, covering 648,700 km2. The altitude of the Loess Plateau ranges from 800 to 3,000 m, its average annual temperature is 6–14 °C, and its average annual precipitation is 200–700 mm. Since ancient times, the Loess Plateau has been used for agriculture because of its fine grains, fluffy soil texture, and rich soluble mineral nutrients, all of which are conducive to crop cultivation. However, long-term unsustainable land use caused the degradation of the vegetation cover in the Loess Plateau. Moreover, the land is degrading due to considerable nutrient loss caused by long-term water erosion in conjunction with natural conditions, such as arid climate, loose soil, concentrated and heavy rainfall. The fragmented ground in the region has made it susceptible to soil erosion. It has also become the primary source of Yellow River sediment as a result of the massive flow of eroded sediment into the Yellow River, posing a serious threat to the economic and social development of the lower Yellow River basin.Terracing is one of the main measures used to enhance crop yield and conserve soil and water in the region. Since the 1980s, the Chinese government has implemented many large-scale slope-to-terrace projects in the Loess Plateau. Especially in recent years, the outline of the comprehensive management plan for the Loess Plateau area (2010–2030) has been promulgated with a planned area of 2.608 million hectares for slope to terrace conversion, making it the core area of slope to terrace conversion projects in the country.Data preparationAlthough high-resolution satellite images can be an important data source for the spatial distribution of terraces on the Loess Plateau, they are not ideal for terraces classification. On the one hand, a higher resolution image requires more storage space. On the other hand, it reduces the efficiency, prolongs the interpretation time, and increases the noise in the image, affecting the interpretation accuracy. Most of the terraces on the Loess Plateau are wider than 7 m (Fig. 1b–d). These are wide terraces in comparison with the narrow terraces of southern China (Fig. 1f), which are less than 2 m wide. Furthermore, it is also easy to mistake the fish-scale pits constructed for soil and water conservation for terraces because of their similarity in form. However, as the width of their field surface is less than 1.5 m, remote sensing images with a 2 m resolution can effectively prevent the false extraction of such features. Based on the actual situation of this study area, we chose a high-resolution image with a spatial resolution of 1.89 m from Google Maps 16 level as the data source. The colour, texture, and morphological features of terraces in the images show seasonal variations. In autumn and winter, the weather is dry, and the vegetation is less shaded in the Loess Plateau. During this time, even the edge features become more visible and easier to identify. As a result, we selected images from October 2018 to February 2019 whenever possible (Fig. 1c,d).Deep learning network selectionLand classification is the extraction of land types from remote sensing images using image segmentation techniques. As the key technology of image segmentation, the Fully Convolutional Network (FCN) classifies images at the pixel level. FCN follows the network structure pattern of encoding and decoding, which adopts AlexNet as the encoder of the network and then employs transposed convolution to up-sample the feature map output from the final convolutional layer of the encoder to the resolution of the input image to achieve pixel-level image segmentation. However, due to the large error in image pixel boundary localization, Ronneberger et al.29 improved the FCN structure in 2015 by expanding the capacity of the network decoder by adding a contracting path to the encoding and decoding modules to achieve more accurate pixel boundary localisation29. The U-Net network is commonly used in medical image processing because it requires a small number of training samples and is effective in classifying objects with a fixed structure and limited semantic information. This network is comparable to natural image semantic segmentation such as Deeplab v3+, which has a smaller number of model parameters and the same effect.Since the texture and morphological features of terraces and human organs have certain similarities, they are primarily manifested by simple semantic information contained within the terrace images themselves. Thus, high-level semantic information and low-level features of such images become more important. However, high-resolution images are more complicated and variable than medical image patterns, and errors in terrace extraction edge identification using the U-Net network, such as boundary segmentation of terraces and flatlands, still occur. To fully utilize the semantic information of the network, we adopted a nested U-Net architecture, namely the UNet++ network proposed by Zhou et al.28. The network integrates long-connected and short-connected architectures to capture features at different levels by adding a shallower U-Net structure and integrates them via feature superposition to make the scale difference of feature maps smaller when fused to enhance the correct rate of image segmentation edges. However, because the U-Net++ network increases the number of model parameters, this study adopted the sparse matrix approach to accelerate model training and decrease the number of parameters.Data pre-processingData pre-processing is a prerequisite for UNet++ network training, that is, valid input according to the standard format annotation before training can be performed. Since the UNet++ network proposed by Zhou et al.28. is primarily used for medical images, which have characteristics such as fixed image structure, no spatial information, and less pattern variation, labelling medical images is comparatively easier using this method. In contrast, high-resolution remote sensing images have a large number of rasters, many pattern changes, irregular image structure, and spatial information. Therefore, determining how to better annotate high-resolution remote sensing images and reduce the annotation workload becomes critical. First, we vectorized the training sample area and generated the terrace vector dataset using ArcGIS with a high-resolution remote sensing image as the primitive map. Second, we converted the terrace vector dataset into raster data. The information of the raster had to be identical to that of the primitive map, including the size of the raster, its processing range, and its coordinate system. The output was converted to TIFF format to complete the image annotation. Since the raster size input to UNet++ network training is a fixed size, it is much smaller than the original image. To simplify the process of inputting the original image and its annotation information, we added an image import module to DLTEM, which was a sliding window of 400*400, and read the image automatically by setting the corresponding judgement conditions. Finally, the entire high-resolution image was processed automatically into the model in accordance with the established rules for training.The goal of the data enhancement was to improve the universality and robustness of the UNet++ network training results. As mentioned above, the high-resolution images taken simultaneously often included clouds or other anomalies in some areas, as the images were stitched together using multiple sources of data fusion. This can easily form evident stitching traces (Fig. 1c,d) due to the different shooting times and image quality of various data sources, i.e., brightness, saturation, and colour contrast of the images. Thus, the model trained on the original image data has strong limitations, and in many scenes, there are notable matrix-type misclassification regions due to image differences, making extraction work challenging. Therefore, in this study, we first adjusted the brightness, grayscale, and contrast of the training data after input to enhance its colour feature recognition ability. We then altered the scaling of the image, and rotated and transformed the training image from 0° to 360° to enhance morphological feature recognition and the accuracy of the training network in terrace extraction.Parameter settingThe network parameter setting is the most critical hyperparameter for UNet++ network training. They are mainly divided into input image size, batch size, learning rate, number of iterations, objective function, gradient descent strategy, momentum, decay rate, and activation function. Among them, we set the image size to 400*400 pixels based on the actual situation of the terraced area, where the UNet++ network has four scaling times, and the image size must be a multiple of 16. The batch size primarily affects the convergence of the model. If the batch limit is set to one, the model is easily affected by the random perturbation phenomenon and cannot converge to find the optimal solution. Since the batch size is determined by the size of the video memory, the value of the batch is limited by equipment constraints. The model in this study used a 2080Ti video card with 11 GB of video memory, and the batch was set to 8. The learning rate, gradient descent strategy, and objective function play a role in whether the network can find the best classification model better and faster. The learning rate was set to 0.001 for the first 500 generations, with the goal of achieving fast convergence to the target region. The learning rate was then set to 0.0001 for 500–1,000 generations, and the model was fine-tuned by choosing a smaller learning rate to find the model with the highest classification accuracy. Adam was chosen for the gradient descent strategy. The momentum and adaptive learning rate were used to increase the convergence rate. The cross-entropy classification loss function was chosen as the objective function to improve the differentiation between terraced and non-terraced areas. Momentum, decay rate, and activation function were all adopted from the previous default settings of the UNet++ network.Data correctionIn this study, we primarily used high-resolution images from Google Earth as the data source to extract the distribution of terraces on the Loess Plateau. Because this image source only contains a large amount of texture structure information and no vegetation information, it is easy to misjudge and misclassify features with the same morphological structure and edge features, such as permanent snow and ice, water bodies, bare land, and artificial surfaces. Vegetation information was generally processed based on waveband data from multispectral/hyperspectral images. It requires topographic correction, atmospheric correction, radiometric calibration, de-clouding, and other operational processes, which are extremely sophisticated30.GlobeLand30 is a 30 m spatial resolution global surface coverage dataset developed by the National Geomatics Center of China. The most recent GlobeLand30 dataset (v2020) has been updated with data sources from 2017 to the present. Its extensive data sources enable effective reduction of the impacts of cloud cover, with an overall accuracy of 85.72%. The classification accuracy of permanent snow and ice, water bodies, bare land, and artificial surfaces of this dataset is as high as 75.79%, 84.70%, 81.76%, and 86.70%, respectively. Since the update time of v2020 data is similar to that of high-resolution images, it can be used as correction data for vegetation information31.Since the training image data are two-dimensional planar data with no elevation or slope information (Fig. 1g), certain flat fields with visible field bumps are easily misclassified as terraces. The Space Shuttle Radar Topography Mission (SRTM v4.1) DEM has a spatial resolution of 30 m and ranges from 60° N to 56° S, completely covering the Loess Plateau32,33. In this study, these data were treated as terrain correction data. The amendment standard corrects the areas that have been extracted as terraces below 2° to non-terraced areas according to the requirements of the Ministry of Natural Resources of China.The spatial resolution of our extracted terraces is 1.89 m, whereas the spatial resolution of GlobeLand30 and DEM as correction data sources is 30 m, which is difficult to meet the requirements of data processing. Hence, we up-sample the two correction data sources, and then used multi-source data fusion. First, we extracted and up-sampled the terraced areas of glaciers, rivers, and deserts from GlobeLand30 to a spatial resolution of 1.89 m. Secondly, we up-sampled the DEM to 1.89 m using spatial interpolation for its raster centre as the true value of the region and performed a slope calculation for the up-sampled DEM. Further, the spatial distribution maps of glaciers, rivers, deserts, and slope maps of the Loess Plateau with the same resolution as the spatial distribution maps of terraces were available. Finally, we superimposed these images, used the terrace range in the TDMLP as a mask, and assessed the pixels in the mask area one by one. If a pixel belonged to permanent snow and ice, a water body, bare land, or an artificial surface, or had a slope less than 2°, it was modified to the background value. Otherwise, the original value was retained.We made artificial corrections to the data based on the extracted results for the arid areas of the Loess Plateau as well as for the flatter basins, given that these areas do not feature terraces.Training and validation dataFor supervised classification, the selection of sample areas and sample features is crucial. The focus and core of any land classification work is representative and effective training sample selection. To obtain a better sample area selection, we considered the selection of sample areas from three perspectives, i.e., colour texture features, topographic features, and spatial distance of the training samples. First, the terraces in this study are in agricultural land, including cultivated land, woodland, grassland, and other types of land; thus, different types of land will present different texture details. At the same time, high-resolution images from Google Earth are mosaicked. Because of the different acquisition times, the same region and land type will have visible colour differences and stitching traces, which is more common in the Loess Plateau region. Therefore, these factors should be considered in the selection of training samples as much as possible to improve the generability of the model and the correct rate of its extraction. Second, the state of the terraces varies according to topographic features. Among them, gradient, direction, altitude, and climate are the most significant factors. Terraces can be categorised as shallow-slope or steep-slope terraces. Based on slope aspect, altitude, and climate characteristics, they can also be categorised as either easy to identify or hard to identify. Thus, the sample should be inclusive of these types of terraces. According to the first law of geography, terraces in different spatial locations have different morphologies. Therefore, the spatial location of the samples should also be at a certain distance.In summary, we selected one county in each region based on the geomorphic zoning characteristics of the Loess Plateau. In addition, we added one more in the area where the density of terraces may be higher. Finally, we selected the whole area of seven counties (Fig. 3) as the training sample area distribution, covering 2.18% of the overall Loess Plateau area. The colour morphological features, topographic features, spatial location, and imaging quality of terrace images in these regions are highly representative. This method was unique from other classification methods. Most of the traditional methods are based on the single-pixel information of feature layers such as random forests, which tend to ignore the neighbouring information around the point, and thus are subject to misclassification and under classification for land types with outstanding texture information. In our study, we adopted the visual interpretation of the whole domain, which can cover the neighbourhood information of each pixel point more comprehensively. To ensure the uniformity and correctness of visual interpretation, the terraces in the training area were visually interpreted by seven interpreters after uniform professional training. For the disputed and uncertain areas, the seven interpreters carried out interactive interpretation and scoring according to the interpretation results. Finally, two other interpretation experts made the final review and corrections. The interpretation results of the training area were re-examined and revised based on the results of the later interpretations.Fig. 3Distribution of training sample areas and validation sites in terraces on the Loess Plateau.Full size imageTo better assess and compare the validity and correctness of the terraced agricultural area datasets on the Loess Plateau in quantitatively, the validation dataset was divided into two parts: a per-pixel point-based validation set and a field validation dataset of terraces with location information. The extracted datasets were comprehensively evaluated in terms of both pixel scale and field validation.We constructed a single-pixel validation point that evaluates the TDMLP. We applied the Icosahedral Snyder Equal Area Discrete Global Grid created by ArcGIS. Based on this strategy, the study area was partitioned into 972 regions (Fig. 3). To better validate the terrace classification results (excluding non-terrace classes), we placed more validation points within the grid where the terrace distribution is more concentrated. First, we calculated the proportion of terraces in each hexagonal grid to the total area of the hexagonal grid. Second, we separated the terraces into four levels according to the proportion of terraces to the whole grid area as 0–20%, 20–50%, 50–80%, and 80–100% and the number of validation points was 10, 20, 40, and 50, respectively.Since the proportion of the extracted terraced area to the total area was only 14%, direct random point deployment would have led to fewer terraced validation sets and thus would have affected the final data evaluation. Therefore, in the deployment strategy, we ensured that the validation points distributed in the extracted terraces in each grid account for at least one-fifth of the total number of validation points, but for the grid with a smaller proportion of terraces or even 0, this practice was meaningless. Hence, we stipulated that in the grid with a proportion of terraces ≤1%, direct random scattering was to be performed. The final scattered verification points in the terraced and non-terraced areas were 5,194 and 6,226, respectively, with a ratio close to 1:1 for easy verification. The spatial distribution is shown in Fig. 3.We validated the spatial distribution map of terraces on the Loess Plateau from 14 April 2021 to 1 May 2021 and constructed a field validation dataset of terraces with location information. Considering the longitudinal, latitudinal, and vertical heterogeneities of the Loess Plateau, the verification route was divided into two sections, north to south and east to west, to more comprehensively cover all regions of the Loess Plateau. The verification route started at Hohhot in the northeast of the Loess Plateau. It passed through the Datong Basin, followed the Yellow River to the south and the Weihe Plain, and then travelled westward through Mount Liupan to the westernmost part of the Loess Plateau. The route was through 54 counties/districts in 16 cities and six provinces on the Loess Plateau, with a total distance of 3,680 km, covering 15.8% of the counties on the Loess Plateau (total of 341 counties). We also surveyed and sampled the verification points approximately every 5 km along the route and collected data from a total of 815 sample points, covering various types of terraces on the Loess Plateau. The results are shown in Fig. 3. More

All procedures involving animals were conducted in accordance with the guidelines and regulations from Institutional Animal Care and Use Committee (IACUC) of the University of Florida (protocol #201509019). Tis manuscript is reported in accordance with ARRIVE guidelines.Site descriptionThis study was carried out at the North Florida Research and Education Center, in Marianna, FL (30°46′35″N 85°14′17″W, 51 m.a.s.l). The trial was performed in two experimental years (2019 and 2020) in a greenhouse.The soil used was collected from a pasture of rhizoma peanut (Arachis glabrata Benth.) and Argentine bahiagrass (Paspalum notatum Flügge) as the main forages. Without plant and root material, only soil was placed into buckets, as described below in the bucket assemblage section. Soil was classified as Orangeburg loamy sand (fine-loamy-kaolinitic, thermic Typic Kandiudults), with a pHwater of 6.7, Mehlich-1-extratable P, K, Mg and Ca concentrations of 41, 59, 63, 368 mg kg−1, respectively. Average of minimum and maximum daily temperature and relative humidity in the greenhouse for September and November (September for beetle trial due seasonal appearance of beetles, and October and November to the Pear Millet trial) in 2019 and 2020 were 11 and 33 °C, 81%; 10 and 35 °C, 77%, respectively.Biological material determinationTo select the species of beetles, a previous dung beetle sampling was performed in the grazing experiment in the same area (grass and legume forage mixture) to determine the number of dung beetle species according to the functional groups as described by Conover et al.44. Beetles were pre-sampled from March 2017 to June 2018, where Tunnelers group were dominant and represented by Onthophagus taurus (Schreber), Digitonthophagus gazella (Fabricius), Phanaeus vindex (MacLeay), Onthophagus oklahomensis (Brown), and Euniticellus intermedius (Reiche). Other species were present but not abundant, including Aphodius psudolividus (Linnaeus), Aphodius carolinus (Linnaeus), and Canthon pilularius (Linnaeus) identified as Dweller and Roller groups, respectively. The pre-sampling indicated three species from the Tunneler group were more abundant, and thereby, were chosen to compose the experimental treatments (Fig. 4).Figure 4Most abundant dung beetle species in Marianna, FL used in the current study. Credits: Carlos C.V. García.Full size imageBeetles collection and experimental treatmentsThree species of common communal dung beetles were used: O. taurus (1), D. gazella (2), and P. vindex (3). Treatments included two treatments containing only soil and soil + dung without beetles were considered as Control 1 (T1) and Control 2 (T2), respectively. Isolated species T3 = 1, T4 = 2, T5 = 3 and their combinations T6 = 1 + 2 and T7 = 1 + 2 + 3. Dung beetles were trapped in the pasture with grazing animals using the standard cattle-dung-baited pitfall traps, as described by Bertone et al.41. To avoid losing samples due to cattle trampling, 18 traps were randomized in nine paddocks (two traps per paddock) and installed protected by metal cages, and after a 24-h period, beetles were collected, and the traps removed. Table 1 shows the number of dung beetles, their total mass (used to standardize treatments) per treatment, and the average mass per species. To keep uniformity across treatments we kept beetle biomass constant across species at roughly 1.7 to 1.8 g per assemblage (Table 1). Twenty-four hours after retrieving the beetles from the field traps, they were separated using an insect rearing cage, classified, and thereafter stored in small glass bottles provided with a stopper and linked to a mesh to keep the ventilation and maintaining the beetles alive.Table 1 Total number and biomass of dung beetles per treatment.Full size tableBuckets assemblageThe soil used in the buckets was collected from the grazing trial in two experimental years (August 2019 and August 2020) across nine paddocks (0.9 ha each). The 21 plastic buckets had a 23-cm diameter and 30-cm (0.034 m2) and each received 10 kg of soil (Fig. 5). At the bottom of the recipient, seven holes were made for water drainage using a metallic mesh with 1-mm diameter above the surface of the holes to prevent dung beetles from escaping. Water was added every four days to maintain the natural soil conditions at 60% of the soil (i.e., bucket) field capacity (measured with the soil weight and water holding capacity of the soil). Because soil from the three paddocks had a slightly different texture (sandy clay and sandy clay loam), we used them as the blocking factor.Figure 5Bucket plastic bucket details for dung beetle trial.Full size imageThe fresh dung amount used in the trial was determined based on the average area covered by dung and dung weight (0.05 to 0.09 m2 and 1.5 to 2.7 kg) from cattle in grazing systems, as suggested by Carpinelli et al.45. Fresh dung was collected from Angus steers grazing warm-season grass (bahiagrass) pastures and stored in fridge for 24 h, prior to start the experiment. A total of 16.2 kg of fresh dung was collected, in which 0.9 kg were used in each bucket. After the dung application, dung beetles were added to the bucket. To prevent dung beetles from escaping, a mobile plastic mesh with 0.5 mm diameter was placed covering the buckets before and after each evaluation. The experiment lasted for 24 days in each experimental year (2019 and 2020), with average temperature 28 °C and relative humidity of 79%, acquired information from the Florida Automated Weather Network (FAWN).Chamber measurementsThe gas fluxes from treatments were evaluated using the static chamber technique46. The chambers were circular, with a radius of 10.5 cm (0.034 m2). Chamber bases and lids were made of polyvinyl chloride (PVC), and the lid were lined with an acrylic sheet to avoid any reactions of gases of interest with chamber material (Fig. 6). The chamber lids were covered with reflective tape to provide insulation, and equipped with a rubber septum for sampling47. The lid was fitted with a 6-mm diameter, 10-cm length copper venting tube to ensure adequate air pressure inside the chamber during measurements, considering an average wind speed of 1.7 m s−148,49. During measurements, chamber lids and bases were kept sealed by fitting bicycle tire inner tubes tightly over the area separating the lid and the base. Bases of chambers were installed on top of the buckets to an 8-cm depth, with 5 cm extending above ground level. Bases were removed in the last evaluation day (24th) of each experimental year.Figure 6Static chamber details and instruments for GHG collection in the dung beetle trial.Full size imageGas fluxes measurementsThe gas fluxes were measured at 1000 h following sampling recommendations by Parkin & Venterea50, on seven occasions from August 28th to September 22nd in both years (2019 and 2020), being days 0, 1, 2, 3, 6, 12, and 24 after dung application. For each chamber, gas samples were taken using a 60-mL syringe at 15-min intervals (t0, t15, and t30). The gas was immediately flushed into pre-evacuated 30-mL glass vials equipped with a butyl rubber stopper sealed with an aluminium septum (this procedure was made twice per vial and per collection time). Time zero (t0) represented the gas collected out of the buckets (before closing the chamber). Immediately thereafter, the bucket lid was tightly closed by fitting the lid to the base with the bicycle inner tube, followed by the next sample deployment times.Gas sample analyses were conducted using a gas chromatograph (Trace 1310 Gas Chromatograph, Thermo Scientific, Waltham, MA). For N2O, an electron capture detector (350 °C) and a capillary column (J&W GC packed column in stainless steel tubing, length 6.56 ft (2 M), 1/8 in. OD, 2 mm ID, Hayesep D packing, mesh size 80/100, pre-conditioned, Agilent Technologies) were used. Temperature of the injector and columns were 80 and 200 °C, respectively. Daily flux of N2O-N (g ha−1 day−1) was calculated as described in Eq. (1):$${text{F}}, = ,{text{A}}*{text{dC}}/{text{dt}}$$
(1)
where F is flux of N2O (g ha−1 day−1), A is the area of the chamber, and dC/dt is the change of concentration in time calculated using a linear method of integration by Venterea et al.49.Ammonia volatilization measurementAmmonia volatilization was measured using the open chamber technique, as described by Araújo et al.51. The ammonia chamber was made of a 2-L volume polyethylene terephthalate (PET) bottle. The bottom of the bottle was removed and used as a cap above the top opening to keep the environment controlled, free of insects and other sources of contamination. An iron wire was used to support the plastic jar. A strip of polyfoam (250 mm in length, 25 mm wide, and 3 mm thick) was soaked in 20 ml of acid solution (H2SO4 1 mol dm−3 + glycerine 2% v/v) and fastened to the top, with the bottom end of the foam remaining inside the plastic jar. Inside each chamber there was a 250-mm long wire designed with a hook to support it from the top of the bottle, and wire basket at the bottom end to support a plastic jar (25 mL) that contained the acid solution to keep the foam strip moist during sampling periods (Fig. 7). The ammonia chambers were placed installed in the bucket located in the middle of each experimental block after the last gas sampling of the day and removed before the start of the next gas sampling.Figure 7Mobile ammonia chamber details for ammonia measurement in dung beetle trial. Adapted from Araújo et al.51.Full size imageNutrient cyclingPhotographs of the soil and dung portion of each bucket were taken twenty-four hours after the last day of gas flux measurement sampling to determine the dung removal from single beetle species and their combination. In the section on statistical analysis, the programming and statistical procedures are described. After this procedure, seeds of pearl millet were planted in each bucket. After 5 days of seed germination plants were thinned, maintaining four plants per bucket. Additionally, plants were clipped twice in a five-week interval, with the first cut occurring on October 23rd and the second cut occurring on November 24th, in both experimental years. Before each harvest, plant height was measured twice in the last week. In the harvest day all plants were clipped 10 cm above the ground level. Samples were dried at 55 °C in a forced-air oven until constant weight and ball-milled using a Mixer Mill MM 400 (Retsch, Newton, PA, USA) for 9 min at 25 Hz, and analyzed for total N concentration using a C, H, N, and S analyzer by the Dumas dry combustion method (Vario Micro Cube; Elementar, Hanau, Germany).Statistical analysisTreatments were distributed in a randomized complete block design (RCBD), with three replications. Data were analyzed using the Mixed Procedure from SAS (ver. 9.4., SAS Inst., Cary, NC) and LSMEANS compared using PDIFF adjusted by the t-test (P  More